communications medicine Article

https://doi.org/10.1038/s43856-025-00738-5

Vital signs-based healthcare kiosks for

screening chronic and infectious
diseases: a systematic review
Check for updates
1,2,3 4 5 6 6
Saksham Bhutani , Aymen Alian , Richard Ribon Fletcher , Hagen Bomberg , Urs Eichenberger ,
Carlo Menon 1,7 & Mohamed Elgendi 1,2,3,7

Abstract Plain language summary

1234567890():,;
1234567890():,;

Background Increasing demands, such as the COVID-19 pandemic, have presented Our healthcare systems face increasing
substantial challenges to global healthcare systems, resulting in staff shortages and demands due to staff shortages,
overcrowded emergency rooms. Health kiosks have emerged as a promising solution to overcrowded emergency rooms, and new
improve overall efficiency and healthcare accessibility. However, although kiosks are challenges from infectious diseases such as
commonly used worldwide for access to information and financial services, the health kiosk COVID-19. This study reviews research on
industry, valued at $800 million, accounts for just 1.9% of the $42 billion global kiosk market. health kiosks, a type of self-service technol-
This review aims to bridge the research-to-practice gap by examining the development of ogy designed to help address these chal-
health kiosk technology from 2013 to 2023. lenges. Health kiosks can measure vital signs,
Methods We conducted a systematic search across PubMed, IEEE Xplore, and Google assist in diagnosing conditions such as high
Scholar databases, identifying 5,537 articles, with 36 studies meeting inclusion criteria for blood pressure and diabetes, and offer
detailed analysis. We evaluated each study based on kiosk purpose, targeted diseases, remote care to ease the strain on healthcare
measured vital signs, and user demographics, along with an assessment of limitations in workers. We reviewed studies on kiosks
participant selection and data reporting. developed between 2013 and 2023, finding
Results The findings reveal that blood pressure is the most frequently measured vital sign, utilized that they show potential for improving access
in 34% of the studies. Furthermore, cardiovascular disease detection emerges as the primary and efficiency in healthcare. However, the
motivation in 56% of the included studies. The United States, India, and the United Kingdom are widespread use of health kiosks is limited by
notable contributors, accounting for 43% of the reviewed articles. Our assessment reveals technical, financial, and regulatory chal-
considerable limitations in participant selection and data reporting in many studies. Additionally, lenges. Addressing these issues could enable
several research gaps remain, including a lack of performance testing, user experience evaluation, health kiosks to become a more common tool
clinical intervention, development standardization, and inadequate sanitization protocols. in healthcare, helping people access care
Conclusions This review highlights health kiosks’ potential to ease the burden on healthcare more easily and improving health outcomes.
system and expand accessibility. However, widespread adoption is hindered by technical,
regulatory, and financial challenges. Addressing these barriers could enable health kiosks to
play a greater role in early disease detection and healthcare delivery.

A global shortage of 18 million health workers is expected by 20301. Cur- assessing or treating patients already in the ED represent the most critical
rently, there are 11.89 nurses per 1000 people in the United States2. The ratio reason for ED overcrowding5. In its current form, the healthcare sector is
is even worse in China, with only 3.1 nurses per 1000 people3. Emergency slow, inefficient, understaffed, and ill-equipped. The situation is further
department (ED) crowding is a major healthcare issue. Due to ED over- compounded by some of the recent pandemics, such as Ebola and COVID-
crowding, 10-day mortality rates were seen to increase by 50%4. Delays in 19, when lapses in the healthcare industry became apparent. A total of 21%

1
Biomedical and Mobile Health Technology Research Lab, ETH Zürich, Zürich, Switzerland. 2Department of Biomedical Engineering and Biotechnology, Khalifa
University of Science and Technology, Abu Dhabi, UAE. 3Healthcare Engineering Innovation Group (HEIG), Khalifa University of Science and Technology, Abu
Dhabi, UAE. 4Yale School of Medicine, Yale University, New Haven, CT, USA. 5Massachusetts Institute of Technology, Cambridge, MA, USA. 6Department for
Anesthesiology, Intensive Care and Pain Medicine, Balgrist University Hospital, Zürich, Switzerland. 7These authors contributed equally: Carlo Menon, Mohamed
Elgendi. e-mail: [email protected]; [email protected]

Communications Medicine | (2025)5:28 1

Content courtesy of Springer Nature, terms of use apply. Rights reserved
https://doi.org/10.1038/s43856-025-00738-5 Article

of Sierra Leone’s health workforce died in the 2014-2015 Ebola outbreak6, Search strategy and study eligibility
while 20% of all healthcare workers left their jobs since the start of the The PubMed, Institute of Electrical and Electronics Engineers (IEEE
COVID-19 pandemic7. These facts reveal that our health force is over- Xplore), and Google Scholar were searched for articles published between
burdened and exposed. Jan 1, 2013, and June 1, 2023. The detailed strategy on IEEE Xplore was the
Kiosks can be an indispensable tool for addressing the growing burden following query: (("Full Text & Metadata”:kiosk OR “Full Text & Metada-
on our healthcare system. Kiosks are most commonly used to provide access ta”:terminal OR “Full Text & Metadata”:booth OR “Full Text & Metada-
to information or financial services; but kiosks employed in healthcare ta”:platform) AND ("All Metadata”:healthcare OR “All Metadata”:hospital
settings are commonly referred to as “health kiosks,” which generally OR “All Metadata”:clinic OR “All Metadata”:nursing home OR “All Meta-
measure vital signs, such as heart rate, blood pressure, oxygen saturation, data”:primary care) AND ("All Metadata”:vital sign OR “All Metadata”:-
respiratory rate, body temperature, as well as certain laboratory parameters physiological measurement OR “All Metadata”:biometric OR “All
such as hemoglobin (HbA1c) and low-density lipoprotein cholesterol Metadata”:health parameter)). On Pubmed, the following search terms were
(LDL). Such systems are advancing medical technology, making healthcare used: (kiosk OR terminal OR booth OR platform) AND (healthcare OR
efficient and accessible. Health kiosks are currently employed for a variety hospital OR clinic OR “nursing home” OR “primary care”) AND (vital sign
of functions including self-check-in, telemedicine, diagnostics, and OR “physiological measurement” OR biometric OR “health parameter”). We
reducing ED crowding. Modern kiosks are able to help diagnose patients included Google Scholar into our study due to its reputation as a compre-
or provide early detection of problems ranging from hypertension and hensive database, particularly for grey literature sources16. However, Google
diabetes to eye and teeth diseases. The use of kiosks streamlines patient Scholar lacks “reliable and scalable methods to extract data”17. On Google
processing while reducing labor requirements and minimizes the exposure Scholar the following terms were used: (kiosk) AND (healthcare OR hospital
of staff to diseases while adding convenience. The ultimate goal is that OR clinic OR “nursing home” OR “primary care”) AND ("vital sign” OR
utilization of health kiosks can lead to improved efficiency, accuracy, “physiological measurement” OR biometric OR “health parameter”). These
and effectiveness in early disease detection and early intervention, search terms brought 2,160 search outcomes, but only 1000 records were
ultimately contributing to enhanced healthcare outcomes and patient displayed. Two reviews11,12 have already addressed literature predating 2013.
well-being. Moreover, this timeframe was chosen to reflect advances in smart sensors,
Global interactive kiosk market size is valued at USD 42 billion at a artificial intelligence technologies, and their kiosk applications in medicine.
compound annual growth rate (CAGR) of 6% between 2022 to 20308. The The search for this review was completed in June 2023.
healthcare kiosk market is calculated at USD 0.8 billion and is expected to
reach USD 1.81 billion by 2028 at a CAGR of 12.49%9. In 2018, sales of 432 Inclusion and exclusion criteria
thousand units were recorded, with the Americas as the biggest adopter, In this study, we have used the following definition for healthcare kiosk: “any
followed by Europe, Asia-Pacific, and the Middle East10. freestanding units containing computer programs capable of measuring
Two reviews, one published in 200911 and another in 201412, addressed vitals for the purpose of conducting disease screening.” Health application
literature predating 2013. This review focuses on literature published in the software (apps) were deemed eligible for inclusion only if they were speci-
last decade. During this period, only one other systematic review on the fically designed and accessible on publicly accessible devices. Conversely,
topic has been published, conducted by Letafat-nejad et al.13, which exam- apps installed exclusively on personally owned devices, including smart-
ines articles up until 2018. However, this review adopts a broad definition of phones, tablets, laptops, and desktop computers, were excluded from the
health kiosks, encompassing various functionalities such as patient regis- study. Studies without participants were also included. Articles were exclu-
tration, feedback, information dissemination, and education. In contrast, ded (a) if the focus was not on the kiosk for medical applications, (b) if it was
our review is distinctly focused on a comprehensive evaluation of health not a health kiosk, (c) if the kiosk was not directly measuring vitals (asking
kiosks designed specifically for the screening of chronic and infectious users to self-report symptoms), and (d) if the article was a review article. Two
diseases through the measurement of vital signs. By narrowing our scope to reviewers (SB and ME) independently conducted the literature search,
this specific area of study, we aim to provide a detailed and comprehensive screened the titles, abstracts, and full texts for potentially eligible studies.
analysis of the effectiveness and utility of these specialized health kiosks in
disease screening. Study selection and data extraction
The main objective of this review is to analyze scientific publications The literature search was carried out using the Rayyan software18. Two
from January 2013 to June 2023 and identify critical factors for future authors (SB and ME) independently screened the titles and abstracts of
research. These factors include global trends, the diseases or syndromes potential studies without employing any automation tools. Any disagree-
screened by health kiosks, the vital signs monitored, the sensors used ments regarding the eligibility of an article were resolved through discus-
for these measurements, and the clinical outcomes of the kiosks. sion. Each study deemed potentially eligible underwent full-text screening,
Additionally, the review compares the techniques employed in kiosks to during which study-specific information and data were extracted. Various
validated gold standards, examines the settings in which kiosks are perspectives were considered for the included articles, including the pub-
deployed, and evaluates methods for measuring user experience. The lication year, author(s), author(s)’s country, study purpose, sensor type
review acknowledges and addresses the challenges pertaining to the acces- (contact or contactless), gold standard clinical measurement, evaluation
sibility of health kiosks in terms of their hardware and software design, the metrics, healthcare setting and adherence to design standards. Additionally,
sanitization of kiosks between users, and the regulatory considerations a sub-group analysis based on the purpose of the kiosks was conducted.
associated with privacy and data collection. This review also comments on
the limitations of various papers and provides recommendations for future Limitations assessment of individual studies
investigations. A protocol was established to assess limitations in individual studies. We
selected five types of limitations to address the research questions of this
Methods review. These included: limitations in the selection of participants that could
Study Guidelines arise due to under or over-representation of participants of a particular
This review was conducted according to the Preferred Reporting Items for gender, age, ethnicity, or other factors, limitations due to a small sample size
Systematic Reviews and Meta-Analyses statement (PRISMA)14. A prior of participants that could lead to chance findings, limitations due to non-
review protocol was drafted using the Preferred Reporting Items for Sys- response because of unwillingness or inability of participants to complete
tematic Reviews and Meta-Analyses Protocols15. This review has been the study, limitation due to selective result reporting, and limitations in
registered on PROSPERO [CRD42022351687]. The review protocol has measurement of outcome due to poor performance of the system in diag-
been provided in Supplementary Note 1. nosing over dataset or compared to a clinical reference measurement. Two

Communications Medicine | (2025)5:28 2

Content courtesy of Springer Nature, terms of use apply. Rights reserved
https://doi.org/10.1038/s43856-025-00738-5 Article

Fig. 1 | Flow diagram of the exclusion criteria used in this study. Out of the initial ineligible. Each of the 338 potentially eligible studies underwent full-text screening,
total of 5537 articles, 157 duplicates were removed. The remaining 5380 articles were leading to the exclusion of 302 studies. Ultimately, 36 studies were identified as
screened based on their titles and abstracts, resulting in 5042 articles being deemed eligible.

Fig. 2 | Overall trends of publications using kiosks

in a healthcare setting from Jan 2013 to June 2023.
The graph indicates a rise in interest in the topic,
especially in 2020, which is suspected to be due to
COVID-19. Data for this graph can be found in
Supplementary Data 3.

authors (SB and ME) independently assessed each study and marked it as excluded if they were not using health kiosks (n = 241), if they were review
having a high, moderate, or low level of limitation or could not be deter- articles (n = 19), if they were not directly measuring vitals (n = 42). All the
mined (due to insufficient information) for all elements. For limitations due results (eligible and ineligible) are provided in Supplementary Data 2.
to the small sample size, studies with less than 25 participants were deemed
high limitation level, between 25 and 120 to be moderate limitation level, Study Characteristics
and more than 120 to be low limitation level. The limitation in the outcome All papers included in this analysis are summarized in Supplementary
measurement is considered a high limitation level if the error is greater than Data 1. Figure 2 shows the cumulative sum of publications and the total
10%, moderate limitation level if the error is between 10% and 5%, and a low number of studies published per year that used kiosks in a healthcare setting.
limitation level if the error is less than 5%. For limitation due to non- The findings indicate that the total number of publications on healthcare
response, if the non-response is more than 30%, then it is considered high kiosks is increasing yearly. Figure 3 displays the worldwide distribution of
limitation level, between 30% and 10%, then it is considered a moderate literature pertaining to the topic under review. Notably, the United States
limitation level, and for less than 10%, it is considered low limitation level. emerged as the leading contributor with 19% of the studies, followed by
India at 14%, and the United Kingdom at 10%. Most articles did not report
Results the location of the study, in which case the locations of the institute authors
Study selection are affiliated with were considered. In the case of different locations for
As shown in Fig. 1, 5537 publications were found using PubMed (n = 2961), different authors, we considered all the locations.
IEEE (n = 1576), and Google Scholar (n = 1000) databases. We used the While health kiosks can be organized by function (e.g. triage-support,
aforementioned search queries and applied filters to identify publications disease screening, chronic disease monitoring, etc.), it was found that the
from the last ten years (January 2013 to June 2023). Studies were also primary function of all papers reviewed was for the purpose of disease

Communications Medicine | (2025)5:28 3

Content courtesy of Springer Nature, terms of use apply. Rights reserved
https://doi.org/10.1038/s43856-025-00738-5 Article

Fig. 3 | The global distribution of scientific articles on kiosks in healthcare is based on the 36 included studies. Detailed information about this analysis can be found in
Supplementary Data 3.

screening. Therefore, we organized the papers studied in this review into six optimized a healthcare kiosk to measure the patients’ blood pressure,
distinct disease categories, which are listed below. Some published studies oxygen saturation, pulse rate, HbA1c, ECG, height, and weight. The
failed to mention a specific purpose for the health kiosk, in which case we authors designed an ergonomic kiosk while trying to minimize mea-
inferred the purpose based on the vital signs and sensors used in the study. surement errors.
Figure 4 shows the number of studies detecting each disease.
Respiratory disease. Four articles20,29,30,37 screened for common pul-
19–45
Cardiovascular disease (CVD). In total, we identified 27 studies monary or respiratory diseases. Four studies19,20,28,29 proposed remote
potentially screening for cardiovascular diseases. Gómez et al.23 presented photoplethysmography (rPPG) kiosks that can use an RGB camera to
a kiosk for the detection and prevention of CVDs, including atrial determine the heart rate, respiratory rate, oxygen saturation, and blood
fibrillation (AF). The kiosk recorded the blood pressure, an electro- pressure. These studies claimed that rPPG could be used to non-
cardiogram (ECG), oxygen saturation, respiration rate, and body tem- invasively measure the tissue blood volume pulses in the microvascular
perature and administered a questionnaire about the patient’s family and tissue bed underneath the skin. The kiosk by Rizal et al.29 in particular
personal history. All these parameters were fed into two machine learning displayed the calculated value of all the vitals and some graphs. The study
models (logistic regression and random forest) to identify whether the validated the measurements on 11 subjects and found a mean absolute
patient had cardiovascular disease or not. After testing the kiosk on 54 error of 1.7 Beats per minute, 0.41 Breaths per minute, and 8.15 mmHg
participants, the authors reported the measurement error and root mean for the pulse rate, respiratory rate, and systolic blood pressure, respec-
square error for all vital signs and the accuracy, F1 score, recall, precision, tively. Pap et al.30 presented an “eHealth Data Acquisition Kiosk” that
and specificity of the classifiers. recorded blood pressure, oxygen saturation, air flow, and galvanic skin
response. The platform allowed remote data rendering for remote con-
Metabolic syndrome. Seven studies31,34,35,37,38,41,43 also monitored vitals sultation with a physician.
that could be potentially used to screen for metabolic syndrome. All of
these studies measured HbA1c and other hematological parameters. The Infectious disease. Seven of the articles25,46–51 used the proposed kiosks
HbA1c test measures the amount of blood sugar (glucose) attached to to screen for infectious diseases transmitted horizontally from human to
hemoglobin, and it is a good indicator of diabetes. Ng et al.31,35 showcased human. Notably, five of these articles25,46–48,50 specifically emphasized the
a kiosk for screening chronic assessment that measured a patient’s blood utilization of the kiosks for COVID-19 screening purposes. Khetan et al.50
pressuxre, LDL-C, HbA1c, height, and weight. Next, the kiosk stratified showcased the “NeelKavach” kiosk for efficiently screening people
the patients into high-, medium-, or low-limitation level categories using entering any premises. The kiosk used a neural network to identify
a simple conditional algorithm based on clinical practice guidelines. whether people entering the premises were wearing masks. Each person’s
Based on the four categories, the kiosk advised the patient to refill their body temperature was then checked using a thermal camera. The data
medicine, see a nurse, or see a doctor. Similarly, Bahadin et al.34 proposed were sent to the concerned authorities if the body temperature was above
a kiosk for follow-up consultations for stable chronic diseases. This kiosk a preset threshold. The kiosk also offered optional hand and luggage
used blood pressure, LDL-C, and HbA1c in addition to the pulse rate. disinfection using an alcohol-based sanitizer and UV-C light, respec-
Their algorithm combined patients’ physiological parameters and recent tively. The authors claimed that the entire process was contactless. The
laboratory results to classify them into good, suboptimal, or poor-control kiosk was tested on 261 subjects, and an overall accuracy of 99% for the
groups using pre-defined rules. Based on these classifications, the kiosk mask detection neural network was found. In another example, Ganesh
produced a result slip for the patient with instructions to continue their et al.25,48 introduced “AutoImpilo,” an automated health machine for
current medications for those with good disease control or to see a nurse virtual health checkups and self-screening in rural areas. The kiosk was
or doctor for further management. Finally, Liu et al.35 designed and equipped with sensors for measuring body temperature, heart rate,

Communications Medicine | (2025)5:28 4

Content courtesy of Springer Nature, terms of use apply. Rights reserved
https://doi.org/10.1038/s43856-025-00738-5 Article

Fig. 4 | Visualization depicting the relationship between the disease categories sign (n = 29), followed by heart rate (n = 19), body temperature (n = 16), oxygen
being-supported, the vital signs being used, and the sensors employed in the saturation (n = 15) and respiration rate (n = 7). Cuff or sphygmomanometer was the
studies analyzed in this review. Most studies (n = 27) detected cardiovascular most used sensor (n = 24) followed by pulse oximeter (n = 22), infrared (IR) tem-
diseases, followed by the detection of metabolic syndrome (n = 7), infectious perature (n = 13), RGB camera (n = 11), glucometer (n = 5), thermal camera (n = 3),
diseases-horizontal transmission (Infectious-H) (n = 7), respiratory (n = 4), infec- electrocardiogram (ECG) (n = 2), airflow sensor (n = 2), digital stethoscope (n = 1),
tious diseases-vector borne transmission (Infectious-VB) (n = 1), dental diseases respiration radar (n = 1), and doppler blood flow sensor (n = 1).
(n = 1), and eye diseases (n = 1). Blood Pressure is the most dominantly used vital

oxygen saturation, and blood pressure. This information was then stored chikungunya, the system exhibited a commendable ability to correctly
on a smart card that users could provide to their physician. The kiosk identify the unlikelihood of the presence of these mosquito-borne diseases,
could also offer patients a range of common medications based on preset achieving an accuracy rate of 91.67%.
thresholds for the measured vitals. The authors further extended the use
case of this kiosk to detect COVID-19. To this end, an infrared tem- Dental disease. Pentti et al.53 developed and tested a kiosk named the
perature sensor was used to determine the user’s body temperature. The “DentoCam Extraoral Imaging System” to identify oral health care needs.
user’s hands would then be sanitized using UV-C light if the body tem- The DentoCam comprises five machine vision cameras to produce five
perature was below the threshold. Otherwise, the kiosk would ask the user images: intraoral clinical daylight and fluorescence images of the upper
a few questions about their symptoms if the temperature was above the and lower dental arches and daylight photographs of the front view with
threshold. Based on the severity of the symptoms, the kiosk would initiate the posterior teeth biting together. DentoCam is accompanied by
an online video call with a physician or book a COVID-19 test at a nearby mechanical guides to help the patients assume the correct position for
hospital and an ambulance to take the patient there. photographing and a live video to show the patients how visible their
Health kiosks for infectious disease screening also included vector- teeth are. The system was tested on 21 participants. A trained dentist
borne diseases, such as malaria, which afflicts many tropical regions of the analyzed all the images taken by the kiosk. The dentist marked all tooth
world. Magwili et al.52 developed a kiosk specifically designed for the fillings, missing teeth, and caries lesions. To analyze the reliability of the
screening of mosquito-borne diseases in the Philippines. The kiosk effec- annotator, 13 of the images were re-annotated by a trained dentist.
tively measures various physiological parameters such as body temperature, Cohen’s kappa between the DentoCam images and the clinical evaluation
blood pressure, and heart rate. In addition, the system incorporates a was calculated to be 0.32.
comprehensive questionnaire that captures essential symptoms including
fever, skin manifestations, vomiting, loss of appetite, and nausea, among Eye disease. One of the studies that we analyzed was that by Kapoor
others. To enhance the diagnostic accuracy, an expert system employing et al.54, who introduced a tele-ophthalmology kiosk named “GlobeChek”
fuzzy logic and weighted rules was integrated into the kiosk, enabling the for detecting common eye diseases. GlobeChek is a globe-shaped kiosk
prediction of the likelihood of chikungunya, dengue, or malaria infection. designed to perform the following screenings: intraocular pressure,
Through a series of evaluations involving a total of 80 tests (20 tests per pachymetry, anterior segment optical coherence tomography, posterior
disease and 20 without), the system’s preliminary diagnoses were compared segment optical coherence tomography, and non-mydriatic fundus
against those determined by medical experts. The results revealed a photography. The patients were categorized into one or more of the
remarkable level of accuracy, with the system correctly pre-diagnosing following categories based on the screenings: healthy ocular, abnormal
dengue, chikungunya, and malaria in 71.67%, 83.33%, and 91.67% of cases, ocular, primary glaucoma suspects, narrow-angle glaucoma suspects,
respectively. Moreover, for diseases not classified as dengue, malaria, or diabetic retinopathy (DR), macular degeneration suspects, and other eye

Communications Medicine | (2025)5:28 5

Content courtesy of Springer Nature, terms of use apply. Rights reserved
https://doi.org/10.1038/s43856-025-00738-5 Article

conditions. Over 4 months, 326 participants were screened, of which 133 department crowding, finding that 80% of participants found the kiosk easy
(40.79%) were detected to have a sight-threatening eye disease or con- to use, with younger and more educated users completing tasks faster. Brizio
dition, 47 (14.41%) had more than one disease, while the results of 192 et al.47 used a kiosk for automatic vitals recording during triage procedures,
(58.89%) came out normal. Of the 133 (40.79%) participants identified saving an average of nine minutes compared to conventional methods.
with at least one eye disease, 70 (21.47%) had primary glaucoma, 37 Tompson et al.27 evaluated a kiosk for blood pressure measurement and data
(11.34%) had narrow-angle glaucoma, 6 (1.84%) had DR, 4 (1.22%) had upload to electronic medical records, concluding that while the system could
macular degeneration, and 43 (13.10%) had other eye diseases. These 133 reduce healthcare provider workload, it required higher patient utilization to
participants were recommended to see an ophthalmologist; however, be cost-effective.
only 71 patients followed up, and the physician confirmed 33 of the 71
diseases (46.47%). Standards of accessibility
Another critical factor that many articles overlooked is following the stan-
Clinical outcomes dards of accessibility. Making kiosks accessible requires accounting for
Ten studies22,23,25,28,30,32,38,44,46,48 employed kiosks for telemedicine or tele- visual, auditory, physical, speech, cognitive, language, learning, and neu-
consultation, primarily in community settings with limited access to rological disabilities. According to the WHO55, over 1 billion people (15% of
healthcare, such as rural areas. For instance, Vaidya et al.44 tested their kiosk the world’s population) are expected to experience disabilities, and disabled
in rural India and found that it substantially supported the rural healthcare people are far more likely to require healthcare services than people without
system through remote diagnosis. The kiosk’s audio/video call feature disabilities. Thus, it is crucial to ensure that all potential kiosk users,
enabled doctors to provide enhanced healthcare services to rural popula- including people with disabilities, have a decent user experience. For kiosks,
tions in a cost-effective manner, effectively improving healthcare accessi- accessibility standards must be followed for software and hardware design.
bility and quality in these underserved regions. Some of the oldest yet relevant standards are the Americans with
Thirteen studies20,23,25,31,33–35,37,50–54 introduced kiosks capable of asses- Disabilities Act56 (ADA) standards for accessible design, released in 2010.
sing patient conditions or making diagnoses. These kiosks typically mea- The ADA, a federal civil rights law in the USA, recommends a minimum
sured vital signs and, in many cases, also incorporated historical electronic and maximum height for interactive touchpoints based on the depth of
health records or solicited additional information about the patient’s current obstructions in front of the kiosk and the forward and side reach of people
or past conditions. on wheelchairs. The ADA also requires the use of braille and the option of an
Ten studies20,23,25,33,37,50–54 developed kiosks for autonomous diagnosis. audio jack with graphic applications for people with visual disabilities.
For instance, Chong et al.37 developed a kiosk for automated triage aimed at Furthermore, the ADA has certain recommendations for keypads and
reducing emergency department overcrowding. Their system utilized vital keyboards, such as having keys with different colors and shapes to make
signs measured by sensors, along with syndrome information and the chief them readily identifiable and appropriate spacings between keys. The
complaint collected through a patient questionnaire. The kiosk, employing a European Union also has a law, the Accessibility Requirements for ICT
random forest algorithm, predicted the triage level in approximately six Products and Services57 (EN301-549). This law is similar to the ADA and
minutes on average, a duration comparable to manual triage times. Yao uses the same minimum and maximum heights for interactive touchpoints.
et al.51 conducted a comparison of classification algorithms for a multimodal The most recent version of EN301-549, released in 2021, includes additional
infection screening device based solely on vital signs. They identified support stipulations for font enlargement, real-time text (RTT), and visual indicators
vector machines and quadratic discriminant analysis as the most effective with audio, to name a few. Another comprehensive set of guidelines is the
methods, both achieving an error rate of 9.8%. Khetan et al.50 reported faster Implementation Guide Regarding Automated Self-Service Kiosks58 by the
and safer early COVID-19 detection with their kiosk, tested on over 1000 Canadian Transport Agency. These guidelines recommend a set of mini-
users. Gómez et al.23 achieved F1 scores of 0.81 and 0.83 in screening car- mum and maximum heights for interactive touchpoints similar to those of
diovascular disease risk using random forest and logistic regression classi- the ADA. Still, they also include a series of specific recommendations, such
fiers, and 0.83 for arrhythmia detection using a deep neural network. as the use of light text on the dark background for the graphical user
Three studies31,34,35 focused on chronic condition assessment in interface (GUI), visual and audible cues for successful and unsuccessful
healthcare settings. Bahadin et al.34 deployed algorithms on a commercially events, and color contrast for insertion slots, among others.
available kiosk for chronic disease management and found the algorithms The Web Content Accessibility Guidelines59 (WCAG) focused on web
decisions in agreement with nurse clinicians (Cohen’s κ = 0.575), with over content accessibility and was developed in cooperation with individuals and
96% of patients preferring the kiosk over nurse visits. Ng et al.31,35 reported organizations worldwide. These guidelines apply only to software for kiosks
that their kiosk’s blood pressure measurements were equivalent to those by using a web agent but are recommended to be followed by all kinds of GUIs.
nurses, and that both physicians and patients were satisfied with the auto- The WCAG has three levels for conformance: A (lowest), AA (mid-range),
mated chronic disease care system. and AAA (highest). Level A sets a minimum level of accessibility and does
Fifteen studies19,21,24,26,27,29,36,39–43,45,47,49 utilized kiosks primarily to only not achieve broad accessibility for many situations. For this reason, it is
measure vital signs, often integrating this data with electronic health records. advised to achieve the AA level. It is not recommended that Level AAA
Out of these five studies21,26,29,40,41 focused on self-checkup capabilities in conformance be required as a general policy because it is not possible to
community setting. Rizal et al.29 developed a kiosk that displayed calculated satisfy all Level AAA success criteria for some content. AA level requires
vital signs and graphs, validating measurements on 11 subjects with a mean using text contrast, animations for interactions, use of simple language,
absolute error of 1.7 BPM for pulse rate, 0.41 breaths per minute for minimum timeout, etc.
respiratory rate, and 8.15 mmHg for systolic blood pressure. Three None of the reviewed articles mentioned whether accessibility stan-
papers21,26,40 tested commercially available kiosks in pharmacy settings to dards were followed in their kiosks. Only three articles31,35,38 provided the
allow patients to self-check their blood pressure. These studies found the dimensions of their kiosks, two of which31,35 were not in accordance with any
kiosks’ measurements to be close to the clinical gold standard measurements. accessibility standards. Thus, only one study38 was found to follow all the
Seven studies19,24,27,36,39,43,47 focused on diagnostic support in a health- accessibility standards56–58 in terms of the hardware design. However,
care setting. Bagula et al.36 presented a triage prioritization system that enough information was not provided to determine if any of the studies were
measured vital signs and employed a multivariate linear regression model to following the software standards for accessibility.
assign scores based on these signs, effectively quantifying their severity
levels. This system provided a quantitative measure of their medical con- Standards for sanitization
ditions, ensuring that the most urgent cases received timely attention. A crucial aspect often overlooked in most studies is the sanitization of
Pacheco et al.24 developed a self-service kiosk to reduce emergency healthcare kiosks, particularly the components that users interact with

Communications Medicine | (2025)5:28 6

Content courtesy of Springer Nature, terms of use apply. Rights reserved
https://doi.org/10.1038/s43856-025-00738-5 Article

Table 1 | Practical metrics to quantify the user experience of kiosks

Metric Calculation Purpose

Pn
Average Time on Task i¼0
Time on Task by the ith user It shows how quickly a user can complete a task.
Total number of users (n) Generally, the faster users can complete a task, the better
the user interface.
 
Task Completion Rate Number of completed task It reflects how effectively users can complete different
× 100
Total number of tasks tasks. It can be a crucial indicator to show if tasks are
clearly conveyed to the user via the interface.
 
Error Occurrence Rate Number of errors It indicates how user-friendly the kiosk is. The higher the
× 100
Total number of attempts error occurrence rate, the more usability problems in the
kiosk interface.
System Usability Determined using a ten-item questionnaire where the contribution of each question SUS, developed by John Brooke87, is a subjective
Scale (SUS) is between 1 and 5. For odd questions, the contribution is scale position minus one, assessment of the usability of the kiosk and sub-scales of
and for even questions, the contribution is five minus the scale position. Multiply the satisfaction and learnability.
sum of a scorer with 2.5 to get SUS.
Customer Satisfaction Users are asked to rate their satisfaction level from 1 ’very unsatisfied’ to 5 ’very CSAT is a customer experience score that indicates the
(CSAT) Score satisfied’,
 and then CSAT is calculated using  the formula: satisfaction level of the user. It can also indicate the value
Number of satisfied customers (4 and 5) addition due to the kiosk for the user.
× 100
Total survey responses

directly. The surfaces of kiosks that patients interact with can harbor minimum, weekly, prior to use on another patient68. They recommend
pathogens, facilitating the transfer of these pathogens between consecutive wiping the device with detergent and clean water, removing any remaining
users. The accumulation of dirt, oils, and biological materials on sensors can detergent residue with a dry lint-free cloth, followed by cleaning with a
further compromise the accuracy and functionality of the sensing tech- disinfectant (as specified by the manufacturer) using a fresh cloth or
nologies. Regulatory bodies mandate stringent sanitization protocols for disposable wipe.
medical devices to ensure patient safety and device efficacy. Consequently, it Ex vivo technologies such as glucometers, which involve direct blood
is essential for healthcare kiosks to comply with these various standards and contact, attract more stringent safety requirements. The CDC advises
regulations to maintain hygiene and operational integrity. against sharing glucometers between patients unless the device is designed
WHO’s “Decontamination and Reprocessing of Medical Devices for for multi-patient use and can be properly cleaned and disinfected69.
Health-care Facilities”60 provides comprehensive guidelines for the cleaning, Notably, only one study addressed the sanitization of healthcare kiosks.
disinfection, and sterilization of medical devices. The Centers for Disease Brizio et al.47 implemented manual cleaning, wherein a staff member sani-
Control and Prevention (CDC) in its “Guideline for Disinfection and tized the kiosk between users. Additionally, six studies19,20,28,29,49–51 utilized
Sterilization in Healthcare Facilities”61 employs the spaulding classification contactless technologies to measure vital signs. While these technologies
to categorize items as critical, semicritical, or noncritical based on their risk might not necessitate frequent sanitization, maintaining cleanliness remains
of infection. This classification dictates the required level of disinfection or essential.
sterilization. For each of these categories the guidelines suggest disinfection
or sterilization methods. The Food and Drug Administration’s (FDA) Ease of use of the kiosks
“Reprocessing Medical Devices in Health Care Settings: Validation Methods Identifying objective parameters to analyze the ease of use of these kiosks is
and Labeling”62, the European Union Medical Device Regulation (MDR) critical. Ease of use can depend on many factors, such as the placement of
2017/74563, and the International Electrotechnical Commission (IEC) different interactive components of the kiosk or the clarity of instructions
60601-1:202464 mandate that manufacturers of medical devices provide and language used. It becomes even more critical to quantify the user
comprehensive, clear, and validated instructions for the cleaning, disinfec- experience of the kiosks in healthcare scenarios because many require users
tion, and/or sterilization of their devices. Specifically, the EU MDR 2017/ to attach medical equipment, such as blood pressure cuffs on arms, SpO2
74563 Annex I stipulates that devices must be designed to facilitate safe monitors on fingertips, etc. In the book Measuring the User Experience:
cleaning, disinfection, and/or re-sterilization to prevent healthcare- Collecting, Analyzing, and Presenting, Albert et al.70 suggested different
associated infections. techniques for measuring user experience performance. Some techniques
For non-contact sensing technologies such as infrared (IR) thermo- relevant to kiosks are time on task, the task completion rate, the error
meters, RGB cameras, and thermal cameras, the risk of pathogen transfer is occurrence rate, the system usability scale (SUS), and customer satisfaction
lower, and thus no particular sanitization may be required. However, (CSAT). We used time on task—the time from the start to the end for using a
maintaining cleanliness remains crucial for ensuring accuracy. Conversely, kiosk—to compare the usability of the kiosks in the reviewed articles simply
sensors that come into direct contact with the skin necessitate slightly because it was the most reported value. Thirteen articles reported the total
stricter protocols. The Association for the Advancement of Medical time required to use the kiosk in some form. By making multiple subjects use
Instrumentation (AAMI) publishes specific standards for the cleaning and their kiosks, ten of the articles20,24,28,33,34,37,47,50,51,53 provided the average time
disinfection of these devices. For instance, ANSI/AAMI ST58:201865 pro- for the task. One of article54 reported only the maximum time for the task,
vides guidelines for the selection and use of chemical disinfectants. For while two32,35 mentioned both the maximum and minimum time for the
sphygmomanometers, the American Heart Association (AHA) recom- task. In most studies32,34,35,37,47,47,50,53,54, the time on task was seen to increase
mends using disposable cuffs whenever possible66. The CDC advises that with an increase in the number of sensors being used. Table 1 introduces
reusable cuffs should be disinfected with an EPA-registered disinfectant some practical metrics, methods to calculate them, and their purpose in
between patients to prevent cross-contamination67. quantifying the user experience of kiosks.
The CDC categorizes the stethoscope as a noncritical surface and Pacheco et al.24 developed a self-service kiosk to reduce ED crowding
suggests that frequent disinfection with alcohol is acceptable unless the and thoroughly analyzed the user interface experience. The authors col-
device is visibly soiled61. ECG leads and cables with soap and water or a lected the time on task along with a questionnaire with general and system
disinfectant wipe between each use61. According to WHO, pulse oximeters usability questions. A proportion of 80% of the participants agreed that the
must be cleaned and disinfected after each individual use and, at a kiosk was easy to use. The authors also linked the time-on-task data to the

Communications Medicine | (2025)5:28 7

Content courtesy of Springer Nature, terms of use apply. Rights reserved
https://doi.org/10.1038/s43856-025-00738-5 Article

age and education level of the participants. They found that participants pulse rate, respiratory rate, and blood pressure, whereas Pap et al.30 chose a
between 18 and 29 years of age with a university-level education took less radically different approach: using commercially available sensors to mea-
time than the average time on task. Silva et al.33 introduced a kiosk for patient sure the same vital signs. Yao et al.51 used a 10 GHz respiration radar to
screening and continuous monitoring. In addition to time on task and the measure respiration rate, a laser Doppler blood flow meter to measure heart
responses to a system usability questionnaire, the authors recorded the rate, and a thermal camera to measure the body temperature.
number of interface clicks. By overlaying the location of clicks on the We selected the gold standards for measuring each vital sign based on
interface, the authors could identify whether the users were using the kiosk the relevant literature and compared the performance of the techniques used
as per the initial design expectations. in the kiosks with these values. Ogedegbe et al.75 reported ambulatory blood
pressure monitoring (ABPM) as the non-invasive gold standard for blood
Discussion pressure measurements. ABPM requires the measurement of an individual’s
The purpose of this review was to assess the current state of the literature and blood pressure at regular intervals for up to 24 hours. A blood pressure
to guide future research dedicated to kiosks in healthcare. We sought to monitor is attached to the waist and connected to a cuff placed around the
describe the purpose for which these kiosks are used, the setting in which person’s upper arm to be used for measurements as the individual leads their
they are used, the technology they use, the working principles behind them, everyday life. ABPM can give a clear idea of the change in blood pressure
and the future scope of these kiosks, as reported in the literature. Through throughout the day and also avoid problems of “white coat” syndrome, in
our analysis, we discovered several noteworthy points related to the acces- which patients’ blood pressure rises due to anxiousness about being tested.
sibility, user experience quantification, and testing of these kiosks, as well as None of the 18 papers using blood pressure as a vital sign used ABPM for
future implementation challenges described below. measurements due to the difficulty of setting up the ABPM device and the
A common evaluation metric, Accuracy (ACC), was selected to analyze long measurement times.
the selected studies. But, Accuracy has also been used in a different sense in For heart rate measurements, Nelson et al.76 considered the ECG the
different studies. Some studies report the accuracy of the prediction of a gold standard. ECG requires attaching multiple electrodes to the patient’s
model or algorithm (ACCM) which is the ratio of the number of correct chest, arms, and legs to measure the electrical activity due to heart muscle
predictions to the total number of input samples. Many studies report the depolarizations. Eighteen articles mentioned the use of heart rate as a vital
accuracy of a sensor (ACCS) which is the degree of closeness of measure- sign in their kiosk. Of these eighteen articles, two articles38,44 reported the use
ments of a quantity to that quantity’s true value. of ECGs for heart rate measurements, four articles22,25,31,34 did not report the
We selected a definition of vital signs from each identified study to method employed, one article29 reported the use of imaging PPG, and others
determine the key vital signs. Elliot et al.71 claimed that acute changes in a used a pulse oximeter to measure this vital sign. Because ECGs are slow and
patient’s physiology can be recognized by accurately assessing their vital require a trained medical professional, they are not ideal for kiosks.
signs. The authors believe that the five traditional vital signs may not be Plüddemann et al.77 considered invasive arterial blood gas (ABG)
adequate for detecting clinical changes in patients who have care needs that analysis as the gold standard for measuring blood oxygen saturation. ABG
are more complex than what nurses have encountered in the past. For the analysis requires the collection of a blood sample from an artery of the
aforementioned reason, the following vital signs have been considered in patient, followed by a measurement of oxygenated hemoglobin. Due to the
this paper: body temperature, heart rate, blood pressure, respiratory rate, invasive nature of ABG analysis, none of the kiosks studied in this review
oxygen saturation, pain, level of consciousness, and urine output. No other employed it. Sixteen articles22–25,30,33,36–39,41–44,46,47,52 used pulse oximeters
parameter measured in the articles was considered a vital sign. Figure 4 attached to the fingertip to measure oxygen saturation or pulse rate. Pulse
shows the percentage of studies using each vital sign. None of the studies oximeters use the concept of spectrophotometry, according to which oxy-
used pain, level of consciousness, or urine output as vital signs. The studies genated hemoglobin (HbO2) absorbs more infrared light and less red light
evaluated in this review employed kiosks for various use cases. Multiple than deoxygenated hemoglobin.
reviews72–74 have explored the potential of the Internet of Things enabled Liu et al.78 considered capnometry the gold standard for measuring the
platforms in enabling video consultations and tele-monitoring, thereby respiration rate. Capnometry requires the use of a mass spectrometer or an
enabling healthcare providers to deliver essential services, during challen- infrared analyzer to measure the CO2 concentration in a patient’s breath
ging pandemics such as COVID-19. This technological approach holds using an endotracheal tube (ET). Though not invasive, capnometry is a very
particular promise for aiding patients with chronic diseases in effectively cumbersome process that requires inserting the ET tube into the patient’s
managing their conditions. Vengadeshwaran et al.22, developed a kiosk for trachea by a medically trained professional. For this reason, none of the
telemedicine. In this case, the kiosk was used to facilitate a video call between seven articles using respiration rate as a vital sign used capnometry. Gómez
the patient and the doctor, but it also measured the patient’s vital signs and et al.23 utilized ECG signals to estimate the respiration rate, whereas Rizal
reported them in real-time to the doctor. This type of technology can be et al.29 used rPPG signals from the face and palm. Three studies30,46,52
incremental, especially for people living in remote areas or during a crisis, mentioned the use of an airflow sensor but did not mention the make,
such as COVID-19, when directly meeting the patient can expose the doctor model, or working of this sensor. Yao et al.51 used a Radar to estimate the
to the virus. Other studies, such as that by Ganesh et al.25, used a kiosk for respiratory rate.
end-to-end clinical visits, during which the kiosk screened the patient’s Sermet-Gaudelus et al.79 reported taking the rectal temperature using
vitals, diagnosed them based on these vitals, and provided them with a an electronic thermometer as the gold standard for measuring body tem-
consultation and medicine. In yet another example, Pacheco et al.24 used perature. However, due to sanitization issues and lubrication requirements,
kiosks to reduce emergency room crowding. The idea was that patients rectal temperatures are not used in kiosks. All the articles employed IR
could screen their vitals using kiosks and that the emergency care staff could temperature sensors22–25,30,36,37,43,46–48,52 or a thermal camera49–51 for measuring
focus on treating the patients. Kiosks can also provide a diagnostic, as shown body temperature due to the non-contact nature of this measurement.
by Rizal et al.29 instead of going to a diagnostic center, patients can visit this An important parameter often overlooked in many studies was cali-
kiosk and obtain a comprehensive digital report on their health. bration. Calibration is crucial for ensuring accurate measurements. It can
The purposes of kiosks were also quite varied, from diagnosing patients directly impact patient safety by minimizing the risks associated with
with hypertension and diabetes to recognizing dental and eye diseases. The incorrect readings. It may also aid in enhancing the longevity of the kiosk by
techniques used were also very diverse. Many of the studies relied on new- maintaining long-term performance and reducing the need for repairs or
generation sensors, such as RGB cameras, and novel algorithms for mea- replacements. Only four studies20,28,43,52 addressed the calibration of sensors
surement or diagnostics. In contrast, other studies used standardized, used in their kiosks. Of these, three studies20,28,43 conducted calibration in
medically approved sensors to perform the same tasks. For example, Rizal controlled environments different from the actual deployment settings.
et al.29 used an RGB camera and developed an algorithm to measure the Calibrating in the actual environment where the kiosk is intended to be

Communications Medicine | (2025)5:28 8

Content courtesy of Springer Nature, terms of use apply. Rights reserved
https://doi.org/10.1038/s43856-025-00738-5 Article

deployed, is essential for accounting for real-world conditions, including amounted to approximately nine minutes compared to the conventional
environmental factors such as temperature, humidity, and electromagnetic practice of nurses measuring vital signs at normal times, where the time
interference, as well as user influence. Only one study52 mentioned cali- taken was 22 minutes. This efficient and expedited process was not only
bration in the deployment setting, where they calibrated temperature, blood found to alleviate the burden on the hospital’s emergency department but
pressure, and pulse rate sensing technologies with five participants. also fosters a secure and trusted environment for the hospital staff and
In terms of validation, seven studies23,28,29,36,37,46,49 validated their kiosks healthcare workers.
in controlled conditions. Eleven studies21,26,27,31,35,40,44,50,51,53,54 validated their Within the Community setting, twelve studies22,23,25,28,29,33,41,44,46,48,50,54
kiosks in the actual settings where they were intended to be deployed. investigated the implementation of kiosks. Among these studies, Ahn et al.41
Validation is important as it ensures precision and reproducibility of the conducted a test of kiosks specifically targeting the older population residing
results and establishes performance benchmarks that can be used to com- in private apartments and rest homes, providing them with the means to
pare different kiosks. monitor their health conditions. Additionally, Vaidya et al.44 developed and
While conducting the literature survey, it was very difficult to perform evaluated a mobile healthcare system designed for remote consultation with
concrete comparisons between the selected studies due to the diversity in the doctors, which was tested in rural areas of India. These studies contribute to
nature of the problems that different studies were trying to solve and the the growing body of research focused on leveraging kiosk technology to
dissimilar evaluation metrics reported. It was also challenging to find a enhance healthcare accessibility and provision within community settings.
consistent definition of vital signs in order to consider all the types of Finally, only three studies25,32,40 specifically mentioned the utilization of
measurements taken in the studies. kiosks in Pharmacy settings for telemedicine and e-diagnostics.
The studies were also categorized based on the methods used to The use of health kiosks also raises new regulatory concerns regarding
transmit data to healthcare providers, a patient’s medical home, or a pre- the collection and processing of personal health information, which must
scribing provider. Sixteen studies21–23,25,26,30,32,36,39,41–44,46,48,54 utilized the adhere to relevant privacy laws and regulations. For instance, in Europe, the
internet as the primary mode of communication, sending data from the General Data Protection Regulation81, and in the United States, the Health
kiosk to the cloud, where it could then be accessed by relevant parties. Insurance Portability and Accountability Act82, dictate the safeguarding of
Telemedicine was a dominant application among these studies, with kiosks personal health data. Compliance with these regulations necessitates
enabling patients to measure their vital signs and consult with healthcare implementing measures such as secure data storage, encryption, access
providers remotely. controls, and obtaining clear patient consent. It also requires practices like
Bagule et al.36 tested a mesh network for rural areas with unreliable the use of pseudonyms and de-identification of personal data, ensuring
internet access, where each kiosk acts as a node transmitting data via radio to individuals’ rights to access their data and request its erasure, and promptly
other kiosks. Special relay nodes then aggregate this data and send it to the reporting any data breaches. In addition to privacy regulations, the
cloud using the internet. Khetan et al.50 employed SMS messages to inform deployment of healthcare kiosks may require compliance with specific
authorities about patients with a high likelihood of COVID-19 infection. regional or national healthcare standards. These standards ensure that the
Many studies19,20,24,27,28,31,33–35,37,38,40,45,47,49,51–53 did not mention a method of kiosks meet the necessary quality, safety, and performance criteria set forth
communication, or their kiosks did not require communication capabilities. by regulatory bodies. Moreover, the potential for incorrect or misleading
The studies included in the analysis were systematically categorized information from a kiosk leading to patient harm may introduce liability
based on the specific healthcare settings in which the kiosks were utilized or concerns. Various liability laws and legal principles come into play. Medical
proposed to be used. These settings encompassed Primary Care, Secondary malpractice laws may also apply if it can be demonstrated that the kiosk
Care, Community, and Pharmacy. Within the Secondary Care category, operator or manufacturer failed to exercise reasonable care in ensuring the
further subcategories were established, namely Speciality Clinic and accuracy and reliability of the device, resulting in harm to a patient.
Emergency Department. Some studies were categorized under multiple The financial business model and insurance reimbursement model for
healthcare settings, indicating their applicability in different contexts. the use of health kiosks remain an important challenge. This topic remains
Among the identified studies, six studies20,30,38,42,49,51 did not explicitly specify an active area of discussion for health care policy and funding, as healthcare
the intended healthcare setting for their respective kiosks. Conversely, six- infrastructure shifts from focusing on disease treatment to a new model that
teen studies19,21,25–27,31,33–36,39,41,43,45,48,52 explicitly mentioned the utilization of also includes disease prevention and wellness. Notably, only three
kiosks in a Primary Care setting. Considering the evidence presented in studies25,32,43 have incorporated some form of payment interface within the
included studies, focused on kiosks in a Primary Care setting, it suggests a kiosk. Sarkar et al.32 proposed a novel freemium model, wherein users access
growing level of readiness for the primetime use of healthcare vital sign basic services at no cost or a minimal fee, while advanced features require
kiosks in clinical care. The comparative accuracy of blood pressure mea- payment. Additionally, they introduced an innovative payment mechanism
surements obtained through kiosks, as assessed in multiple trials21,35, whereby users do not pay directly at the kiosk; instead, they incur the costs
demonstrated similar levels of error compared to clinical measurements when collecting their medications from the pharmacy.
when evaluated against the gold standard. This finding supports the relia- The papers included in this review had certain limitations pertaining to
bility and feasibility of kiosk-based measurements for vital signs. Moreover, the recruitment of the participants or data reporting. An analysis of these
the concordance between kiosk-based diagnoses and those made by clin- limitations was performed and the results are summarized in Fig. 5. An
icians for managing patients with stable chronic diseases, as reported in assessment of the limitations of the included studies was also performed.
other studies34, indicates the potential of these kiosks as a valuable tool in Figure 5a shows the limitations for the five selected elements across all the
clinical care. Furthermore, the validation of commercially available kiosk studies. Figure 5b shows the percentage of studies with low, moderate, and
models such as PharmaSmart PS2000 and H4D ConsultStation through high limitation levels for the five selected questions. The interrater reliability
clinical studies80 adds to their credibility, highlighting their readiness for use suggested excellent agreement between the raters, κ = 0.8 (See Supple-
in real-world healthcare settings. mentary Data 4 for overall assessment while Supplementary Data 5 for
Among the studies that highlighted the use of kiosks in Secondary detailed assessment along with the adjudication process). Limitation eva-
Care, one study specifically referred to a Speciality Clinic53 while the luation revealed that most studies had one or more limitations with medium
remaining six studies23,24,36,37,47,48 focused on Emergency Departments. In a or high limitation levels. Thirteen studies19,23,25,27,29,33,35,36,38,41,45,52,53 were ana-
thorough investigation conducted by Brizio et al.47, involving 1844 patients lyzed to have high or moderate limitation level due to the selection of
in the emergency department of a private hospital in France during the participants. Twenty-seven studies19,22–25,28–33,35–43,45,46,48,49,51–53 had high
COVID-19 pandemic, it was observed that the utilization of a telemedicine or moderate limitation levels due to the small sample size out of
kiosk markedly reduced the time interval between patient registration and which eight studies22,30–32,39,42,43,48 did not perform any testing and were
vital signs assessment. The study revealed that the average time saved assessed to have a high limitation level. Seven studies24,27,34,45,47,53,54 had

Communications Medicine | (2025)5:28 9

Content courtesy of Springer Nature, terms of use apply. Rights reserved
https://doi.org/10.1038/s43856-025-00738-5 Article

Fig. 5 | Subjective limitation assessment of included studies. a Reviewers' judg- included studies. Detailed information about these analyses can be found in Sup-
ments about the five elements of limitations in all included studies. b Reviewers' plementary Data 4 and 5.
judgments about the five elements of limitations presented as percentages across all

high or moderate limitation levels due to non-response. Fifteen In our opinion, although the proposed kiosks show promising appli-
studies22,23,25,29,30,32,36,39,42–44,46,48–50 did not pre-specify the analysis they are cations, they do not appear to be ready to become an integral part of the
going to perform; hence, limitations in the study due to selective result healthcare industry. Given the increasing interest in this topic (as shown in
reporting for these could not be determined. Eighteen Fig. 2), we strongly propose the following recommendations for researchers
studies21,23,24,26–28,33,34,37,38,40,41,45,47,51–54 had low limitation levels, while four in this field. First, all studies developing kiosks should strictly abide by
studies19,20,31,35 had high or moderate limitation levels due to selective result accessibility standards, which need to be incorporated in the early stages of
reporting. Seven studies23,29,35,36,51,52,54 were estimated to have high or mod- the kiosk design. These standards dictate the dimensions of the kiosk,
erate limitation levels in the measurement of outcome. the design and position of the interactive components, and the design of

Communications Medicine | (2025)5:28 10

Content courtesy of Springer Nature, terms of use apply. Rights reserved
https://doi.org/10.1038/s43856-025-00738-5 Article

the interface for the kiosk. We have discussed four standards from various over 24h during daily normal life is far more pragmatic and still accurate. In
countries across the globe in this section. Of these, we recommend following a randomized controlled trial21 comprising 510 participants, a comparison
the Canadian Standards58 for the hardware simply because these are very was made between blood pressure readings acquired from the commercially
specific and in accordance with all other standards. For software, we available PharmaSmart BP kiosk and those obtained through ABPM. The
recommend following the WCAG 2.159 AA conformance. Following study’s results indicated that the one-time readings from the kiosk exhibited
accessibility standards will allow people with different disabilities, who a difference of 5 mmHg and 9.5 mmHg higher for systolic and diastolic
constitute a considerable proportion of all users in healthcare settings, to use blood pressure, respectively, when compared to the daytime ABPM read-
kiosks. ings. Importantly, these discrepancies fell within the acceptable error limits
Next, establishing and validating comprehensive sanitization protocols established by the American Heart Association83. Moreover, it should be
is crucial for the safe operation of healthcare kiosks. Studies should develop ensured that the patient sits comfortably with their arm supported and that
and validate specific sanitization procedures, detailing both the methodol- the cuff is snug to obtain an accurate reading. For the measurement of heart
ogy and frequency of cleaning. These protocols should adhere to the sani- rate, the ECG is considered the gold standard due to its accuracy. However,
tization guidelines provided by the device manufacturers for each sensor the main challenge it presents is attaching the electrodes. It is difficult for
and component within the kiosk. We have elaborated on general sanitiza- untrained patients to attach these properly to the correct measured anato-
tion guidelines from various international organizations in section, which mical sites. One simple alternative is using fingertip-based PPG sensors,
should be considered during protocol development. The kiosk’s physical which are simpler and more practical to use and provide sufficiently similar
design should prioritize ease of sanitization, minimizing areas prone to results84. We recommend using pulse oximeters as practical gold standards
contamination and simplifying cleaning processes. Furthermore, we for blood oxygen saturation measurement. This technique is not invasive
encourage studies to explore automation solutions for sanitization to and does not require medical expertise, as most other methods do. To
enhance the practicality and efficiency of real-world deployment of kiosks. measure respiration rate, the ECG-derived respiration (EDR) is a practical
Calibration and validation of the kiosk should be performed in the option for respiratory rate measurements. According to AL-Khalidi et al.85,
deployment setting. All sensing technologies used to measure vital signs in the EDR-based single-lead respiration rate is a robust method for calculating
the kiosk should be calibrated to maximize accuracy. Routine calibration is the respiratory rate. Unfortunately, due to hygiene and sanitary reasons,
recommended to ensure consistent performance and increase the opera- measuring the rectal temperature (the gold standard) is not ideal in the
tional life of the kiosk. Additionally, the sensing technologies and algorithms context of kiosks to measure body temperature. For this reason, we
used in the kiosk should be validated under real-world conditions, and the recommend the use of IR ear thermometers due to their markedly higher
results should be published publicly. This practice will enhance reliability accuracy compared to other non-contact measurement techniques.
but also contribute to transparency and trust in the performance of the Lastly, we recommend that all studies testing kiosks on human subjects
kiosks. follow all ethical, legal, and regulatory norms that apply. Studies should
We recommend that the testing and reporting of the performance of adhere to the tenets of the Declaration of Helsinki: Ethical Principles for
kiosks should be done using standard evaluation metrics. For machine Medical Research Involving Human Subjects86 and should obtain appro-
learning models used in kiosks, we recommend splitting the data between priate permission from the relevant bodies. We recommend that studies do
training, validation, and testing sets before training. Furthermore, the data not record participant data that reveal their identities. If unavoidable, studies
should be fed into the model by random stratified sampling to avoid bias. should undertake proper measures to anonymize and dispose of the data
We also advise that the results of the testing stage be reported as accuracy, F1 after use. It is crucial for studies to comply with local data collection and
score, precision, and recall, along with the receiver operating characteristic privacy laws to ensure the lawful and responsible handling of participant
curve for a thorough analysis of the model performance. For sensors, we information. Additionally, we strongly recommend that studies conduct
recommend accuracy and precision with respect to a gold standard to be comprehensive testing and validation processes to minimize the risk of
reported as evaluation metrics. Moreover, statistical analysis should be liability.
performed if the kiosk’s performance is compared to that of a medical This review has examined the role of kiosks in the healthcare industry
procedure. Studies can estimate metrics such as Cohen’s kappa in this case. based on the existing literature. Five thousand five hundred thirty-seven
Another important factor is to analyze the performance of the kiosks studies were identified from three databases, out of which 36 were selected
should in terms of ease of use. We have proposed five metrics to quantify the and analyzed. The most common purpose of health kiosks was found to be
user experience of the kiosks in Table 1. Quantifying ease-of-use metrics cardiovascular disease screening, representing 56% of studies, with blood
needs to be an integral part of the kiosk design pipeline. Designers need to pressure being the most used vital sign being used in 29 of these studies.
iterate to maximize these metrics to ensure that the kiosk is fast and easy to Overall, 43% of the studies were from the US, UK, and India. Most studies
use for the users and efficient for the developer. had considerable limitations relating to study participant selection or data
Further, integration of point-of-care testing (POCT) into kiosks’ reporting, with only two studies found to have minimal limitations across all
workflows is recommended to enhance diagnostic capabilities. Point-of- criteria. The usability was good for all papers reviewed, and in general,
care tests have revolutionized diagnostic testing by enabling rapid and increasing the number of sensors or vital signs was found to increase the
accurate diagnoses of specific conditions. All the kiosks discussed in this total time required to use the kiosk. Many studies used kiosks with pro-
review rely on measuring physiological signals for diagnosis. Future studies mising technologies like machine learning for the detection of dental disease
should explore the utilization of POCT as an additional test when physio- and cardiovascular disease. Emerging technologies such as rPPG can also
logical signals indicate potential anomalies. Although the inclusion of transform the way health kiosks are used by making the process of data
POCT may introduce new challenges, such as integrating the tests with collection faster, more efficient, and contactless. Based on increasing pub-
kiosk systems, adhering to more stringent sanitization protocols, and lished evidence, it seems that health kiosks can help address major chal-
managing the disposal of testing kits, these can be addressed through lenges in our healthcare infrastructure, such as the shortage of healthcare
meticulous planning and design. workers, overcrowding in emergency rooms, and protecting healthcare
When measuring vital signs, the kiosks should use the gold standards workers from unnecessary exposure to infectious diseases. It is hoped that
directly or as ground truth. Unfortunately, most of these gold standards are healthcare kiosks can also greatly improve healthcare quality by increasing
invasive, complex to perform, and require trained professionals. For these accessibility and efficiency by providing clinical support, reducing the need
reasons, we recommend some practical gold standards to be used with to travel, and overcoming geographic barriers. However, despite these
kiosks. Sphygmomanometer (blood pressure cuff) is recommended to promising applications, several key implementation challenges (technical,
measure blood pressure from the patient’s upper arm. The technique used is financial, and regulatory) currently prevent health kiosks from achieving
the same as mentioned above for ABPM. Measuring the blood pressure not widespread adoption as part of the standard healthcare infrastructure.

Communications Medicine | (2025)5:28 11

Content courtesy of Springer Nature, terms of use apply. Rights reserved
https://doi.org/10.1038/s43856-025-00738-5 Article

Research gaps including a lack of performance testing, user experience 15. Moher, D. et al. Preferred reporting items for systematic review and
evaluation, clinical intervention, development standardization, and inade- meta-analysis protocols (PRISMA-P) 2015 statement. Syst. Rev. 4, 1
quate sanitization protocols remain a hurdle. Nevertheless, the long-term (2015).
view of health kiosks remains optimistic, as a means of improving access to 16. Haddaway, N. R., Collins, A. M., Coughlin, D. & Kirk, S. The role of
healthcare and enabling better screening and detection of common chronic google scholar in evidence reviews and its applicability to grey
and infectious diseases. literature searching. PloS One 10, e0138237 (2015).
17. Martín-Martín, A., Orduna-Malea, E., Thelwall, M. & López-Cózar, E.
Data availability D. Google scholar, web of science, and scopus: A systematic
The protocol for this systematic review is publicly available through the comparison of citations in 252 subject categories. J. Informetr. 12,
International Prospective Register of Systematic Reviews (PROSPERO), 1160–1177 (2018).
with the registration number CRD42022351687. The original contributions 18. Ouzzani, M., Hammady, H., Fedorowicz, Z. & Elmagarmid, A. Rayyan
presented in the study are included in the article/supplementary material. —a web and mobile app for systematic reviews. Syst. Rev. 5, 1–10
The comprehensive search strategy is outlined in Supplementary Note 1 (2016).
(Supplementary Information). Supplementary Data 1 provides a summary 19. Huang, H.-W. et al. Cost-effective solution of remote
of key data extracted from the selected studies, while Supplementary Data 3 photoplethysmography capable of real-time, multi-subject monitoring
presents the complete extracted data from all studies included in this review. with social distancing. In 2022 IEEE Sensors, 1–4 (IEEE, 2022).
Complete lists of all the screened publications for this study are provided as 20. Wong, K. L., Chin, J. W., Chan, T. T., Prae-Arporn, K. & So, R. H. Vitals:
an Excel file in Supplementary Data 2. Supplementary Data 3 provides the Camera-based physiological monitoring and health management
data extracted from the studies included in this review. Data about study platform. In 2022 IEEE/ACM Conference on Connected Health:
limitation analysis is provided in Supplementary Data 4 (overall assessment) Applications, Systems and Engineering Technologies (CHASE),
and Supplementary Data 5 (detailed assessment along with the adjudication 140–141 (IEEE, 2022).
process). Further inquiries can be directed to the corresponding author/s. 21. Green, B. B. et al. Clinic, home, and kiosk blood pressure
measurements for diagnosing hypertension: a randomized diagnostic
Received: 27 August 2023; Accepted: 8 January 2025; study. J. Gen. Intern. Med. 37, 2948–2956 (2022).
22. Vengadeshwaran, V. & Leo, R. A. A. Design and development of self-
service telemedicine kiosk for remote towns. In 2021 International
References Conference on Green Energy, Computing and Sustainable
1. Boniol, M. et al. The global health workforce stock and distribution in Technology (GECOST), 1–5 (IEEE, 2021).
2020 and 2030: a threat to equity and ‘universal’ health coverage? 23. Gómez-García, C. A., Askar-Rodriguez, M. & Velasco-Medina, J.
BMJ Glob. Health 7, e009316 (2022). Platform for healthcare promotion and cardiovascular disease
2. Unruh, L. et al. A comparison of 2020 health policy responses to the prevention. IEEE J. Biomed. Health Inform. 25, 2758–2767 (2021).
covid-19 pandemic in canada, ireland, the united kingdom and the 24. Pacheco, P., Santos, F., Coimbra, J., Oliveira, E. & Rodrigues, N. F.
united states of america. Health Policy 126, 427–437 (2022). Designing effective user interface experiences for a self-service kiosk
3. Wang, Z. et al. Validation of a chinese version of the analysing and to reduce emergency department crowding. In 2020 IEEE 8th
developing adaptability and performance in teams to enhance resilience International Conference on Serious Games and Applications for
scale in nurses in china. J. Nurs. Manag. 30, 1324–1336 (2022). Health (SeGAH), 1–8 (IEEE, 2020).
4. Richardson, D. B. Increase in patient mortality at 10 days associated with 25. Ganesh, D. et al. Autoimpilo: smart automated health machine using
emergency department overcrowding. Med. J. Aust. 184, 213–216 (2006). iot to improve telemedicine and telehealth. In 2020 International
5. Asplin, B. R. et al. A conceptual model of emergency department Conference on Smart Technologies in Computing, Electrical and
crowding. Ann. Emerg. Med. 42, 173–180 (2003). Electronics (ICSTCEE), 487–493 (IEEE, 2020).
6. Raven, J., Wurie, H. & Witter, S. Health workers’ experiences of coping 26. Green, B. B. et al. Blood pressure checks and diagnosing
with the ebola epidemic in sierra leone’s health system: a qualitative hypertension (bp-check): Design and methods of a randomized
study. BMC Health Serv. Res. 18, 251 (2018). controlled diagnostic study comparing clinic, home, kiosk, and 24-
7. Morning Consult. United States. National Tracking Poll 210101. hour ambulatory bp monitoring. Contemp. Clin. trials 79, 1–13 (2019).
Retrieved from the Library of Congress: https://www.loc.gov/item/ 27. Tompson, A. et al. Mixed-methods feasibility study of blood pressure
lcwaN0016127/. (Accessed: 2024-07-01). self-screening for hypertension detection. BMJ Open 9, e027986
8. Straits Research. Interactive kiosk market share, growth to 2030. (2019).
https://straitsresearch.com/report/interactive-kiosks-market. 28. Pasquadibisceglie, V., Zaza, G. & Castellano, G. A personal healthcare
(Accessed: 2024-07-01). system for contact-less estimation of cardiovascular parameters. In
9. SkyQuestt. Healthcare kiosk market size, trends and forecast to 2028. 2018 AEIT International Annual Conference, 1–6 (IEEE, 2018).
https://skyquestt.com/report/healthcare-kiosk-market. (Accessed: 29. Rizal, A., Lin, Y.-C. & Lin, Y.-H. Contactless vital signs measurement
2024-07-01). for self-service healthcare kiosk in intelligent building. In 2018 3rd
10. Transparency Market Research. Health kiosk market worth US $2 Bn International Conference on Intelligent Green Building and Smart Grid
by 2027. https://www.transparencymarketresearch.com/health- (IGBSG), 1–4 (IEEE, 2018).
kiosk-market.html. (Accessed: 2024-07-01). 30. Pap, I. A., Oniga, S., Orha, I. & Alexan, A. Iot-based ehealth data
11. Jones, R. The role of health kiosks in 2009: literature and informant acquisition system. In 2018 IEEE International Conference on
review. Int. J. Environ. Res. public health 6, 1818–1855 (2009). Automation, Quality and Testing, Robotics (AQTR), 1–5 (IEEE, 2018).
12. Joshi, A. & Trout, K. The role of health information kiosks in diverse 31. Ng, G., Tan, S. W. & Tan, N. C. Health outcomes of patients with
settings: a systematic review. Health Inf. Lib. J. 31, 254–273 (2014). chronic disease managed with a healthcare kiosk in primary care:
13. Letafat-Nejad, M., Ebrahimi, P., Maleki, M. & Aryankhesal, A. protocol for a pilot randomised controlled trial. BMJ Open 8, e020265
Utilization of integrated health kiosks: A systematic review. Med. J. (2018).
Islam. Repub. Iran. 34, 114 (2020). 32. Sarkar, S. & Saha, R. A futuristic iot based approach for providing
14. Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G. & Group, P. Preferred healthcare support through e-diagnostic system in india. In 2017
reporting items for systematic reviews and meta-analyses: the prisma Second International Conference on Electrical, Computer and
statement. Ann. Intern. Med. 151, 264–269 (2009). Communication Technologies (ICECCT), 1–7 (IEEE, 2017).

Communications Medicine | (2025)5:28 12

Content courtesy of Springer Nature, terms of use apply. Rights reserved
https://doi.org/10.1038/s43856-025-00738-5 Article

33. Silva, J., Brandão, P. & Prior, R. Usability assessment of a health kiosk. 51. Yao, Y. et al. Multiple vital-sign-based infection screening
In 2017 IEEE Symposium on Computers and Communications (ISCC), outperforms thermography independent of the classification
260–265 (IEEE, 2017). algorithm. IEEE Trans. Biomed. Eng. 63, 1025–1033 (2015).
34. Bahadin, J. et al. Follow-up consultation through a healthcare kiosk 52. MAGWILI, G. V. et al. Raspberry pi-based medical expert system for
for patients with stable chronic disease in a primary care setting: a pre-diagnosis of mosquito-borne diseases. In 2018 IEEE 10th
prospective study. J. Gen. Intern. Med. 32, 534–539 (2017). International Conference on Humanoid, Nanotechnology, Information
35. Ng, G., Tan, N., Bahadin, J., Shum, E. & Tan, S. W. Development Technology, Communication and Control, Environment and
of an automated healthcare kiosk for the management of chronic Management (HNICEM), 1–6 (IEEE, 2018).
disease patients in the primary care setting. J. Med. Syst. 40, 1–11 53. Pentti, L. et al. Can health kiosks be used to identify oral health care
(2016). needs? a pilot study. Acta Odontol. Scand. 79, 370–376 (2021).
36. Bagula, A. et al. Cloud based patient prioritization as service in public 54. Kapoor, R. et al. Detecting common eye diseases using the first
health care. In 2016 ITU Kaleidoscope: ICTs for a Sustainable World teleophthalmology globechek kiosk in the united states: a pilot study.
(ITU WT) (IEEE, 2016). Asia-Pac. J. Ophthalmol. 9, 315–325 (2020).
37. Chong, H. & Gan, K. Development of automated triage system for 55. World Health Organization (WHO) and the World Bank. World report
emergency medical service. In 2016 International Conference on on disability. Geneva, Switzerland. https://www.who.int/publications/
Advances in Electrical, Electronic and Systems Engineering (ICAEES), i/item/9789241564182. (Accessed: 2024-07-01).
642–645 (IEEE, 2016). 56. Department of Justice, Washington D.C., United States. Americans
38. Lyu, Y. et al. Designing and optimizing a healthcare kiosk for the with Disabilities Act (ADA) Standards for Accessible Design. https://
community. Appl. Ergon. 47, 157–169 (2015). www.ada.gov/regs2010/2010ADAStandards/2010ADAStandards_
39. Ahn, H. S., Lee, M. H. & MacDonald, B. A. Healthcare robot systems prt.pdf (2010). (Accessed: 2024-07-01).
for a hospital environment: CareBot and ReceptionBot. In 2015 24th 57. European Telecommunications Standards Institute (ETSI). EN 301
IEEE International Symposium on Robot and Human Interactive 549 - v3.2.1 - Accessibility requirements for ICT products and
Communication (RO-MAN) (IEEE, 2015). services. https://www.etsi.org/deliver/etsi_en/301500_301599/
40. Padwal, R. S., Townsend, R. R., Trudeau, L., Hamilton, P. G. & Gelfer, 301549/03.02.01_60/en_301549v030201p.pdf (2021). (Accessed:
M. Comparison of an in-pharmacy automated blood pressure kiosk to 2024-07-01).
daytime ambulatory blood pressure in hypertensive subjects. J. Am. 58. Canadian Transportation Agency. Implementation guide regarding
Soc. Hypertens. 9, 123–129 (2015). automated self-service kiosks available in multiple formats. otc-cta.
41. Ahn, H. S. et al. Design of a kiosk type healthcare robot system for gc.ca/sites/default/files/kiosk-guide_en_0.pdf (2014). (Accessed:
older people in private and public places. In Simulation, Modeling, and 2024-07-01).
Programming for Autonomous Robots: 4th International Conference, 59. World Wide Web Consortium Recommendation. World Wide Web
SIMPAR 2014, Bergamo, Italy, October 20-23, 2014. Proceedings 4, Consortium Recommendation. Web Content Accessibility Guidelines
578–589 (Springer, 2014). (WCAG) 2.1. https://www.w3.org/TR/2018/REC-WCAG21-20180605/.
42. Lee, M. H., Ahn, H. S., Wang, K. & MacDonald, B. A. Design of a (Accessed: 2024-07-01).
healthcare sensor managing system for vital sign measuring devices. 60. World Health Organization (WHO). Decontamination and
In Simulation, Modeling, and Programming for Autonomous Robots: reprocessing of medical devices for health-care facilities. Geneva,
4th International Conference, SIMPAR 2014, Bergamo, Italy, October Switzerland. https://www.who.int/publications/i/item/
20-23, 2014. Proceedings 4, 521–532 (Springer, 2014). 9789241549851 (2016). (Accessed: 2024-07-01).
43. Shibu, J. & Ramkumar, R. Atm based remote healthcare monitoring 61. Rutala, W. A. & Weber, D. J. Guideline for disinfection and sterilization
system. In 2014 International Conference on Electronics and in healthcare facilities, 2008. update: June 2024. https://www.cdc.
Communication Systems (ICECS), 1–6 (IEEE, 2014). gov/infection-control/media/pdfs/Guideline-Disinfection-H.pdf
44. Vaidya, A. S., Srinivas, M. B., Himabindu, P. & Jumaxanova, D. A smart (2024). (Accessed: 2024-07-01).
phone/tablet based mobile health care system for developing 62. Food and Drug Administration (FDA) Center for Devices and
countries. In 2013 35th Annual International Conference of the IEEE Radiological Health. Reprocessing Medical Devices in Health Care
Engineering in Medicine and Biology Society (EMBC) (IEEE, 2013). Settings: Validation methods and labeling. https://www.fda.gov/
45. Gleason-Comstock, J. A. et al. Consumer health information media/80265/download (2015). (Accessed: 2024-07-01).
technology in an adult public health primary care clinic: a heart health 63. Regulation (eu) 2017/745 of the european parliament and of the
education feasibility study. Patient Educ. Couns. 93, 464–471 (2013). council of 5 april 2017 on medical devices, amending directive 2001/
46. Safi’ie, M. A. et al. Contactless diagnosis using internet of things (iot) 83/ec, regulation (ec) no 178/2002 and regulation (ec) no 1223/2009
technology for covid-19 suspect patient. In 2022 1st International and repealing council directives 90/385/eec and 93/42/eec (text with
Conference on Smart Technology, Applied Informatics, and eea relevance.). http://data.europa.eu/eli/reg/2017/745/oj (2017).
Engineering (APICS), 138–141 (IEEE, 2022). (Accessed: 2024-07-01).
47. Brizio, A., Faure, V., Baudino, F., Wilmet, A. & Gonzalez, J.-P. Hospital- 64. International Electrotechnical Commission (IEC). IEC 60601-1:2024
based autonomous pre-clinical screening of covid-19: An emergency SER - Medical electrical equipment all parts. Geneva, Switzerland.
triage using a vital signs recording system, paris-ile de france region. https://webstore.iec.ch/en/publication/2603 (2016). (Accessed:
Health Serv. Manag. Res. 36, 291–298 (2020). 2024-07-01).
48. Ganesh, D. et al. Automatic health machine for covid-19 and other 65. Association for the Advancement of Medical Instrumentation (AAMI).
emergencies. In 2021 International Conference on COMmunication ANSI/AAMI ST58:2013/(r)2018 - Guideline for disinfection and sterilization
Systems & NETworkS (COMSNETS), 685–689 (IEEE, 2021). in healthcare facilities. https://www.aami.org/detail-pages/product/ansi-
49. Wang, Z.-H., Horng, G.-J., Hsu, T.-H., Chen, C.-C. & Jong, G.-J. A aami-st582013-r-2018-pdf-a152e000006j61xqaa (2018). (Accessed:
novel facial thermal feature extraction method for non-contact 2024-07-01).
healthcare system. IEEE Access 8, 86545–86553 (2020). 66. Pickering, T. G. et al. Recommendations for blood pressure
50. Khetan, D. et al. Innovative product for premise safety from covid 19: measurement in humans: an aha scientific statement from the council
Neelkavach kiosk. In 2020 IEEE Bombay Section Signature on high blood pressure research professional and public education
Conference (IBSSC), 251–255 (IEEE, 2020). subcommittee. J. Clin. Hypertens. 7, 102 (2005).

Communications Medicine | (2025)5:28 13

Content courtesy of Springer Nature, terms of use apply. Rights reserved
https://doi.org/10.1038/s43856-025-00738-5 Article

67. Sehulster, L. et al. Guidelines for environmental infection control in 85. AL-Khalidi, F. Q., Saatchi, R., Burke, D., Elphick, H. & Tan, S. Respiration
health-care facilities. Morbidity and mortality weekly report rate monitoring methods: A review. Pediatr. Pulmonol. 46, 523–529 (2011).
recommendations and reports RR52 (2003). 86. World Medical Association. Declaration of Helsinki: ethical principles
68. World Health Organization (WHO). Care, cleaning and disinfection of for medical research involving human subjects. JAMA 310,
pulse oximeters and patient monitors devices. https://www.who.int/ 2191–2194 (2013).
publications/m/item/care-cleaning-and-disinfection-of-pulse-oximeters- 87. Brooke, J. et al. SUS-a quick and dirty usability scale. Usability Eval.
and-patient-monitors-devices (2024). (Accessed: 2024-07-01). Ind. 189, 4–7 (1996).
69. Centers for Disease Control and Prevention (CDC). Considerations for
blood glucose monitoring and insulin administration. https://www. Acknowledgements
cdc.gov/injection-safety/hcp/infection-control/?CDC_AAref_Val Saksham Bhutani is grateful for the support from ETH Zürich and Khalifa
(2024). (Accessed: 2024-07-01). University (grant number RC2-2018-022).
70. Tullis, T. & Albert, B. Chapter 4 - performance metrics. In Tullis, T. &
Albert, B. (eds.) Measuring the User Experience (Second Edition), Author contributions
Interactive Technologies, 63–97 (Morgan Kaufmann, Boston, 2013), S.B. and M.E. screened and assessed articles. S.B., M.E., A.A., R.R.F., U.E.,
second edition edn. https://www.sciencedirect.com/science/article/ H.B., and C.M. conceived the study. C.M. and U.E. provided directions. S.B.
pii/B9780124157811000042. and M.E. wrote the first draft of the manuscript. M.E. designed and led the
71. Elliott, M. & Coventry, A. Critical care: the eight vital signs of patient study. A.A., H.B. and U.E. provided clinical feedback. All authors have read
monitoring. Br. J. Nurs. 21, 621–625 (2012). and agreed to the published version of the manuscript.
72. Mehraeen, E. et al. Technology in the era of covid-19: a systematic
review of current evidence. Infect. Disord.-Drug Targets 22, 51–60 Funding
(2022). Open access funding provided by Swiss Federal Institute of Technology
73. Shamsabadi, A. et al. Internet of things in the management of chronic Zurich.
diseases during the covid-19 pandemic: A systematic review. Health
Sci. Rep. 5, e557 (2022). Competing interests
74. Mehraeen, E. et al. Current treatments and therapeutic options for The authors declare no competing interests.
covid-19 patients: a systematic review. Infect. Disord.-Drug Targets
(Former. Curr. Drug Targets-Infect. Disord.) 22, 62–73 (2022). Additional information
75. Ogedegbe, G. & Pickering, T. Principles and techniques of blood Supplementary information The online version contains
pressure measurement. Cardiol. Clin. 28, 571–586 (2010). supplementary material available at
76. Nelson, B. W. & Allen, N. B. Accuracy of consumer wearable heart rate https://doi.org/10.1038/s43856-025-00738-5.
measurement during an ecologically valid 24-hour period:
Intraindividual validation study. JMIR MHealth UHealth 7, e10828 Correspondence and requests for materials should be addressed to
(2019). Carlo Menon or Mohamed Elgendi.
77. Plüddemann, A., Thompson, M., Heneghan, C. & Price, C. Pulse
oximetry in primary care: primary care diagnostic technology update. Peer review information Communications Medicine thanks Anna Cantrell,
Br. J. Gen. Pract. 61, 358–359 (2011). Karen Rheuban and Anthony Bokolo for their contribution to the peer review
78. Liu, H., Allen, J., Zheng, D. & Chen, F. Recent development of of this work. [Peer reviewer reports are available].
respiratory rate measurement technologies. Physiol. Meas. 40,
07TR01 (2019). Reprints and permissions information is available at
79. Sermet-Gaudelus, I., Chadelat, I. & Lenoir, G. Body temperature http://www.nature.com/reprints
measurement in daily practice. Arch. Pediatr. 12, 1292–1300 (2005).
80. Alpert, B. S. Validation of the pharma-smart ps-2000 public use blood Publisher’s note Springer Nature remains neutral with regard to
pressure monitor. Blood Press. Monit. 9, 19–23 (2004). jurisdictional claims in published maps and institutional affiliations.
81. European Commission. Regulation (EU) 2016/679 of the European
Parliament and of the Council of 27 April 2016 on the protection of Open Access This article is licensed under a Creative Commons
natural persons with regard to the processing of personal data and on Attribution 4.0 International License, which permits use, sharing,
the free movement of such data, and repealing Directive 95/46/EC adaptation, distribution and reproduction in any medium or format, as long
(General Data Protection Regulation) (Text with EEA relevance). as you give appropriate credit to the original author(s) and the source,
https://eur-lex.europa.eu/eli/reg/2016/679/oj (2016). (Accessed: provide a link to the Creative Commons licence, and indicate if changes
2024-07-01). were made. The images or other third party material in this article are
82. Centers for Medicare & Medicaid Services. The Health Insurance included in the article’s Creative Commons licence, unless indicated
Portability and Accountability Act of 1996 (HIPAA). http://www.cms. otherwise in a credit line to the material. If material is not included in the
hhs.gov/hipaa/ (1996). (Accessed: 2024-07-01). article’s Creative Commons licence and your intended use is not permitted
83. Stergiou, G. S. et al. A universal standard for the validation of blood by statutory regulation or exceeds the permitted use, you will need to
pressure measuring devices. Hypertension 71, 368–374 (2018). obtain permission directly from the copyright holder. To view a copy of this
84. Blok, S., Piek, M., Tulevski, I., Somsen, G. & Winter, M. The accuracy licence, visit http://creativecommons.org/licenses/by/4.0/.
of heartbeat detection using photoplethysmography technology in
cardiac patients. J. Electrocardiol. 67, 148–157 (2021). © The Author(s) 2025

Communications Medicine | (2025)5:28 14

Content courtesy of Springer Nature, terms of use apply. Rights reserved
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:

1. use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
2. use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
3. falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
4. use bots or other automated methods to access the content or redirect messages
5. override any security feature or exclusionary protocol; or
6. share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at

[email protected]