Sharief Taraman, MD, DABPN, DABPM, FAAP

Large language models (LLMs) continue to exhibit noteworthy capabilities across a spectrum of areas, including emerging proficiencies across the health care continuum. Successful LLM implementation and adoption depend on digital readiness, modern infrastructure, a trained workforce, privacy, and an ethical regulatory landscape. These factors can vary significantly across health care ecosystems, dictating the choice of a particular LLM implementation pathway. This perspective discusses 3 LLM… Show more

Large language models (LLMs) continue to exhibit noteworthy capabilities across a spectrum of areas, including emerging proficiencies across the health care continuum. Successful LLM implementation and adoption depend on digital readiness, modern infrastructure, a trained workforce, privacy, and an ethical regulatory landscape. These factors can vary significantly across health care ecosystems, dictating the choice of a particular LLM implementation pathway. This perspective discusses 3 LLM implementation pathways—training from scratch pathway (TSP), fine-tuned pathway (FTP), and out-of-the-box pathway (OBP)—as potential onboarding points for health systems while facilitating equitable adoption. The choice of a particular pathway is governed by needs as well as affordability. Therefore, the risks, benefits, and economics of these pathways across 4 major cloud service providers (Amazon, Microsoft, Google, and Oracle) are presented. While cost comparisons, such as on-demand and spot pricing across the cloud service providers for the 3 pathways, are presented for completeness, the usefulness of managed services and cloud enterprise tools is elucidated. Managed services can complement the traditional workforce and expertise, while enterprise tools, such as federated learning, can overcome sample size challenges when implementing LLMs using health care data. Of the 3 pathways, TSP is expected to be the most resource-intensive regarding infrastructure and workforce while providing maximum customization, enhanced transparency, and performance. Because TSP trains the LLM using enterprise health care data, it is expected to harness the digital signatures of the population served by the health care system with the potential to impact outcomes. The use of pretrained models in FTP is a limitation. It may impact its performance because the training data used in the pretrained model may have hidden bias and may not necessarily be health care–related. However, FTP provides a… Show less

Artificial intelligence (AI) broadly describes a branch of computer science focused on developing machines capable of performing tasks typically associated with human intelligence. Those who connect AI with the world of science fiction may meet its growing rise with hesitancy or outright skepticism. However, AI is becoming increasingly pervasive in our society, from algorithms helping to sift through airline fares to substituting words in emails and SMS text messages based on user choices. Data… Show more

Artificial intelligence (AI) broadly describes a branch of computer science focused on developing machines capable of performing tasks typically associated with human intelligence. Those who connect AI with the world of science fiction may meet its growing rise with hesitancy or outright skepticism. However, AI is becoming increasingly pervasive in our society, from algorithms helping to sift through airline fares to substituting words in emails and SMS text messages based on user choices. Data collection is ongoing and is being leveraged by software platforms to analyze patterns and make predictions across multiple industries. Health care is gradually becoming part of this technological transformation, as advancements in computational power and storage converge with the rapid expansion of digitized medical information. Given the growing and inevitable integration of AI into health care systems, it is our viewpoint that pediatricians urgently require training and orientation to the uses, promises, and pitfalls of AI in medicine. AI is unlikely to solve the full array of complex challenges confronting pediatricians today; however, if used responsibly, it holds great potential to improve many aspects of care for providers, children, and families. Our aim in this viewpoint is to provide clinicians with a targeted introduction to the field of AI in pediatrics, including key promises, pitfalls, and clinical applications, so they can play a more active role in shaping the future impact of AI in medicine. Show less

This was a prospective case-control study of concussed patients presenting to the Emergency Department within 72 hours of injury, and age/sex-matched controls. Pupillary measurements were gathered using NeurOptics’ PLR 3000; evaluation included a symptom checklist and neurocognitive assessment. Data were analyzed using descriptive statistics and logistic/linear regression models.
Results
126 participants were enrolled—28 concussed patients and 29 controls ages 5-11, and 36 concussed… Show more

This was a prospective case-control study of concussed patients presenting to the Emergency Department within 72 hours of injury, and age/sex-matched controls. Pupillary measurements were gathered using NeurOptics’ PLR 3000; evaluation included a symptom checklist and neurocognitive assessment. Data were analyzed using descriptive statistics and logistic/linear regression models.
Results
126 participants were enrolled—28 concussed patients and 29 controls ages 5-11, and 36 concussed patients and 33 controls ages 12-18. One statistically significant difference in pupillometry between concussed and control participants was found, left minimum pupil diameter in patients aged 12-18 (p=0.02). Logistic regression models demonstrating odds of a concussion revealed one significant association, average dilation velocity of the left pupil in 5-11-year-olds (p=0.04). Logistic regression models predicting symptom improvement showed one significant (negative) association: maximum constriction velocity of left pupil in 5-11-years-olds (p=0.04). Logistic regression models for 12-18-year-olds predicting neurocognitive improvement demonstrated association between time to 75% in the left pupil and visual memory, visual motor processing speed, and reaction time (p=0.003, p=0.04, p=0.04).
Conclusion
The limited statistically significant associations found in this study suggest that pupillometry may not be useful in pediatrics in the acute post-injury setting for either the diagnosis of concussion or to stratify risk for prolonged recovery. Further large studies in the acutely injured population are needed. Show less

Methods
A de-identified aggregate data analysis of the initial 124 Canvas Dx prescriptions that fulfilled postmarket authorization was conducted. The analysis (PR015) was determined exempt by the Advarra IRB.
Results
Prescriptions were generated from 15 states. Children had a median age of 39.6 months, were 27.4% female, and had a 51.6% ASD prevalence rate. Compared to the clinical reference standard, Canvas Dx had a negative predictive value (NPV) of 95.2% (20/21; 95% CI, 76.2-99.9)… Show more

Methods
A de-identified aggregate data analysis of the initial 124 Canvas Dx prescriptions that fulfilled postmarket authorization was conducted. The analysis (PR015) was determined exempt by the Advarra IRB.
Results
Prescriptions were generated from 15 states. Children had a median age of 39.6 months, were 27.4% female, and had a 51.6% ASD prevalence rate. Compared to the clinical reference standard, Canvas Dx had a negative predictive value (NPV) of 95.2% (20/21; 95% CI, 76.2-99.9) and a positive predictive value (PPV) of 94.4% (51/54; 95% CI, 84.6-98.8), with 60.5% (75/124; 95% CI, 51.3-69.1) receiving a determinate (positive or negative for ASD) result. The median age of children who received a positive output was 35.5 months.
Conclusions
When deployed in real-world settings, Canvas Dx provided highly accurate positive and negative outputs for ASD that closely aligned with the specialist reference standard. Children were provided a positive for ASD output over 1 year earlier than the current median age of ASD diagnosis in the United States. Confirmatory analysis on larger data sets is required; however, this finding is promising. Stakeholders, including clinicians, payors, and policymakers should consider Canvas Dx as a highly accurate and rapidly deployable replacement for existing non–FDA-authorized research tools repurposed for clinical care—both in the specialist and primary care settings. Show less

The impacts of digital health technologies are being felt across the spectrum of medical specialties, including pediatrics. Innovations ranging from mobile medical apps and software, health information technology, wearable devices, telehealth and personalized medicine are increasingly influencing how pediatric care is conceptualized and delivered. While many opportunities and barriers to clinical adoption mirror those described in adult populations, others are unique to the pediatric health… Show more

The impacts of digital health technologies are being felt across the spectrum of medical specialties, including pediatrics. Innovations ranging from mobile medical apps and software, health information technology, wearable devices, telehealth and personalized medicine are increasingly influencing how pediatric care is conceptualized and delivered. While many opportunities and barriers to clinical adoption mirror those described in adult populations, others are unique to the pediatric health space. Successful digital pediatric entrepreneurship will require a rich understanding of both the overarching and pediatric specific regulatory and clinical context. Lessons learned from digital pediatric entrepreneurship could also inform other special populations such as geriatrics which heavily rely on caregiver engagement. This chapter provides an overview of current challenges facing the pediatric workforce, including staff shortages, the growing mental and behavioral health crisis, and lack of access and equity for certain groups of children. Potential roles digital innovations could play in meeting these challenges are explored and illustrated through topical case studies. For the most part, digital pediatric innovation has been slower off the ground than innovations in the adult digital health space. Clinical, regulatory, and economic barriers all contribute to delays in the pace of pediatric innovation. These barriers are explored in detail in the second half of the chapter together with incentives and frameworks to stimulate future pediatric entrepreneurship and drive clinical adoption. Show less

Background
We investigated the association between chronic pediatric neurological conditions and the severity of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2).

Methods
This matched retrospective case-control study includes patients (n = 71,656) with chronic complex neurological disorders under 18 years of age, with laboratory-confirmed diagnosis of COVID-19 or a diagnostic code indicating infection or exposure to SARS-CoV-2, from 103 health systems in the United… Show more

Background
We investigated the association between chronic pediatric neurological conditions and the severity of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2).

Methods
This matched retrospective case-control study includes patients (n = 71,656) with chronic complex neurological disorders under 18 years of age, with laboratory-confirmed diagnosis of COVID-19 or a diagnostic code indicating infection or exposure to SARS-CoV-2, from 103 health systems in the United States. The primary outcome was the severity of coronavirus disease 2019 (COVID-19), which was classified as severe (invasive oxygen therapy or death), moderate (noninvasive oxygen therapy), or mild/asymptomatic (no oxygen therapy). A cumulative link mixed effects model was used for this study.

Results
In this study, a cumulative link mixed effects model (random intercepts for health systems and patients) showed that the following classes of chronic neurological disorders were associated with higher odds of severe COVID-19: muscular dystrophies and myopathies (OR = 3.22; 95% confidence interval [CI]: 2.73 to 3.84), chronic central nervous system disorders (OR = 2.82; 95% CI: 2.67 to 2.97), cerebral palsy (OR = 1.97; 95% CI: 1.85 to 2.10), congenital neurological disorders (OR = 1.86; 95% CI: 1.75 to 1.96), epilepsy (OR = 1.35; 95% CI: 1.26 to 1.44), and intellectual developmental disorders (OR = 1.09; 95% CI: 1.003 to 1.19). Movement disorders were associated with lower odds of severe COVID-19 (OR = 0.90; 95% CI: 0.81 to 0.99).

Conclusions
Pediatric patients with chronic neurological disorders are at higher odds of severe COVID-19. Movement disorders were associated with lower odds of severe COVID-19. Show less

Tremors affect pediatric and adult populations, with roughly 3% of people worldwide experiencing essential tremors. Treatments include medication, deep brain stimulation, occupational/physical therapy, or adaptive equipment. This unblinded experimental pre-test–post-test study was performed (April–September 2021) at Children’s Health of Orange County, evaluating the effectiveness of Move-D, a novel orthotic brace, on pediatric tremors. Ten participants (14–19 years old) experiencing upper… Show more

Tremors affect pediatric and adult populations, with roughly 3% of people worldwide experiencing essential tremors. Treatments include medication, deep brain stimulation, occupational/physical therapy, or adaptive equipment. This unblinded experimental pre-test–post-test study was performed (April–September 2021) at Children’s Health of Orange County, evaluating the effectiveness of Move-D, a novel orthotic brace, on pediatric tremors. Ten participants (14–19 years old) experiencing upper extremity tremors (5 essential, 2 dystonic, 1 coarse, 1 postural, and 1 unspecified) were enrolled. Participants completed a usability survey and performance was measured utilizing the Bruininks–Oseretsky Test of Motor Proficiency, second edition, with and without the brace, using one-sided t-tests of mean differences. Move-D improved age-equivalent scores for fine motor precision by 20.5 months and upper limb coordination by 15.1 months. Manual coordination percentile rankings increased by 2.9%. Manual dexterity performance was unaffected. The usability survey revealed that 7/10 participants agreed or strongly agreed that they could move their arm freely while wearing the brace, the brace reduced their tremors, and they felt comfortable wearing the brace at home. Through standardized testing and findings from the usability survey, Move-D showed an improvement of functional abilities in a pediatric population with tremors. Show less

BACKGROUND: Prolonged delays between first caregiver concern and autism spectrum disorder (ASD) diagnosis have been reported, but associations between length of time to diagnosis (TTD) and health care resource utilization (HCRU) and costs have not been studied in a large sample of children with ASD.
RESULTS: 8,954 children met selection criteria: 4,205 aged 3 years or younger and 4,749 aged older than 3 years at diagnosis, with median TTD of 9.5 and 22.1 months, respectively. In the year… Show more

BACKGROUND: Prolonged delays between first caregiver concern and autism spectrum disorder (ASD) diagnosis have been reported, but associations between length of time to diagnosis (TTD) and health care resource utilization (HCRU) and costs have not been studied in a large sample of children with ASD.
RESULTS: 8,954 children met selection criteria: 4,205 aged 3 years or younger and 4,749 aged older than 3 years at diagnosis, with median TTD of 9.5 and 22.1 months, respectively. In the year preceding ASD diagnosis, children with longer TTD in both age cohorts experienced a greater number of all-cause and ASD-related health care visits compared with those with shorter TTD (mean and median number of office or home visits were approximately 1.5- and 2-fold greater in longer vs shorter TTD groups; P < 0.0001). The mean all-cause medical cost per child in the year preceding ASD diagnosis was approximately 2-fold higher for those with longer vs shorter TTD ($5,268 vs $2,525 in the younger and $5,570 vs $2,265 in the older cohort; P < 0.0001 for both). Mean ASD-related costs were also higher across age cohorts for those with longer vs shorter TTD ($2,355 vs $859 in the younger and $2,351 vs $1,144 in the older cohort; P < 0.0001 for both). CONCLUSIONS: In the year prior to diagnosis, children with longer TTD experienced more frequent health care visits and greater cost burden in their diagnostic journey compared with children with shorter TTD. Novel diagnostic approaches that could accelerate TTD may reduce costs and HCRU for commercially insured children. Show less

Abstract
Technological breakthroughs, together with the rapid growth of medical information and improved data connectivity, are creating dramatic shifts in the health care landscape, including the field of developmental and behavioral pediatrics. While medical information took an estimated 50 years to double in 1950, by 2020, it was projected to double every 73 days. Artificial intelligence (AI)–powered health technologies, once considered theoretical or research-exclusive concepts, are… Show more

Abstract
Technological breakthroughs, together with the rapid growth of medical information and improved data connectivity, are creating dramatic shifts in the health care landscape, including the field of developmental and behavioral pediatrics. While medical information took an estimated 50 years to double in 1950, by 2020, it was projected to double every 73 days. Artificial intelligence (AI)–powered health technologies, once considered theoretical or research-exclusive concepts, are increasingly being granted regulatory approval and integrated into clinical care. In the United States, the Food and Drug Administration has cleared or approved over 160 health-related AI-based devices to date. These trends are only likely to accelerate as economic investment in AI health care outstrips investment in other sectors. The exponential increase in peer-reviewed AI-focused health care publications year over year highlights the speed of growth in this sector. As health care moves toward an era of intelligent technology powered by rich medical information, pediatricians will increasingly be asked to engage with tools and systems underpinned by AI. However, medical students and practicing clinicians receive insufficient training and lack preparedness for transitioning into a more AI-informed future. This article provides a brief primer on AI in health care. Underlying AI principles and key performance metrics are described, and the clinical potential of AI-driven technology together with potential pitfalls is explored within the developmental and behavioral pediatric health context. Show less

A growing number of artificial intelligence-based medical devices are receiving clearance from the Food and Drug Administration (FDA). Debate has arisen about best practices for the regulation and safe oversight of such devices whose capabilities, if “unlocked”, include iterative learning and adaptation with exposure to new data. One regulatory mechanism proposed by the FDA is the predetermined change control plan (PCCP). This analysis provides what we believe would be the first example of how… Show more

A growing number of artificial intelligence-based medical devices are receiving clearance from the Food and Drug Administration (FDA). Debate has arisen about best practices for the regulation and safe oversight of such devices whose capabilities, if “unlocked”, include iterative learning and adaptation with exposure to new data. One regulatory mechanism proposed by the FDA is the predetermined change control plan (PCCP). This analysis provides what we believe would be the first example of how a PCCP has been leveraged to improve the performance of a de novo autism diagnostic device in practice. Following the PCCP's model update procedures included in the marketing authorization of the first generation of the device (“algorithm V1”), we conducted an algorithmic threshold optimization procedure to improve the device's ability to detect or rule out autism in children ages 18–72 months without changing the accuracy or intended use of the device. Decision threshold optimization was achieved using a repeated train/test validation procedure on a dataset of 722 children with concern for developmental delay, aged 18–72 months (28% autism, 22% neurotypical, 50% other developmental delay, mean age 3.6 years, 39% female). In 1000 repeats, 70% of samples were selected for threshold optimization and 30% for evaluation, ensuring that no training data appeared in the test set. Out-of-sample performance was estimated by evaluating the selected threshold pair on the test set and comparing the performance metrics of the new pair to the corresponding V1 metrics on the same test set. The device, with optimized decision thresholds, produced a determinate output for 66.5% (95% CI, 62.5–71.0) of children. Positive Predictive Value (PPV) and Negative Predictive Value (PPV) were 87.5% (95% CI, 82.5–96.7) and 95.6% (95% CI, 93.7–97.9) respectively. Threshold optimization improved the device's ability to accurately detect or rule out autism in a greater proportion of children... Show less

A total of 64 concussed participants and 35 control participants were recruited for this analysis. In concussed children aged 5 to 11 years, pupil constriction latency and mean pupil constriction velocity were found to be significantly different in distribution when compared with nonconcussed controls. Concussed patients were found to have a significantly slower pupil constriction velocity and mean pupil constriction velocity (3.01 mm/s, 4.6) than controls (3.38 mm/s, 5), P = .03 and P = .005… Show more

A total of 64 concussed participants and 35 control participants were recruited for this analysis. In concussed children aged 5 to 11 years, pupil constriction latency and mean pupil constriction velocity were found to be significantly different in distribution when compared with nonconcussed controls. Concussed patients were found to have a significantly slower pupil constriction velocity and mean pupil constriction velocity (3.01 mm/s, 4.6) than controls (3.38 mm/s, 5), P = .03 and P = .005, respectively. In children 12 to 18 years of age, pupil dilation velocity was significantly slower in patients with concussions (1.2 mm/s) than controls (1.3 mm/s), P = .01. Pupil constriction latency (P = .02), neurological pupil index (P = .02), and mean constriction velocity (P = .03) were significantly associated with PCSS in patients 12 to 18 years of age. In patients 5 to 11 years of age, initial pupil size (P = .009) and constriction velocity (P = .01) were significant predictors of PCSI score. In patients 12 to 18 years of age, an increase of log 1 mm/s in pupil constriction velocity was associated with a significant increase in total BESS scores (P = .03). Initial pupil size and time to 75% pupil size were significantly associated with a decrease in PCSI score in patients 5 to 11 years of age (P = .03 and P = .004, respectively).

Conclusion
Pupillary responses are significantly different in concussed patients compared with age- and gender-matched nonconcussed controls in the 5-11 and 12-18 years old pediatric populations seen in this study. This suggests that with further study, pupillometry may become an objective and reliable diagnostic tool for the assessment, management, and treatment of pediatric patients with or at risk for concussion. Show less

We report a case of neuroblastoma, a pediatric neuroendocrine tumor of the sympathetic nervous system, in a 3-year-old female with multiple cranial nerve involvement. A 3-year-old afebrile, lethargic female presented with bilateral eyelid droop, right head tilt, slurred speech, gagging, abnormal walking and no bowel movement. Neurological examination noted bilateral ptosis, dysarthria, left tongue deviation, proximal weakness in upper and lower extremities, areflexia in biceps and patellar… Show more

We report a case of neuroblastoma, a pediatric neuroendocrine tumor of the sympathetic nervous system, in a 3-year-old female with multiple cranial nerve involvement. A 3-year-old afebrile, lethargic female presented with bilateral eyelid droop, right head tilt, slurred speech, gagging, abnormal walking and no bowel movement. Neurological examination noted bilateral ptosis, dysarthria, left tongue deviation, proximal weakness in upper and lower extremities, areflexia in biceps and patellar tendons, dysmetria, and wide-based gait. MRI of the brain showed heterogeneous appearance of the clivus and MRI of the spine showed right adrenal mass and heterogeneous enhancement of multiple vertebrae, suggesting possible metastatic disease. Serum and cerebrospinal studies were unremarkable. Patient was treated with intravenous methylprednisolone and plasmapheresis for suspected paraneoplastic syndrome; however, she continued to clinically progress. Adrenal mass biopsy results and elevated urine VMA and HVA levels were consistent with the diagnosis of neuroblastoma. Neuroblastoma should be considered as a differential for a neurological presentation involving multiple cranial nerves in a child. Show less

Results: The study comprised 171 416 pediatric patients; the median age was 8 years (IQR, 2-14 years), and 50.28% were male. Of these patients, 17 065 (9.96%) had severe COVID-19. The random intercept model showed that the following cardiovascular conditions were associated with severe COVID-19: cardiac arrest (odds ratio [OR], 9.92; 95% CI, 6.93-14.20), cardiogenic shock (OR, 3.07; 95% CI, 1.90-4.96), heart surgery (OR, 3.04; 95% CI, 2.26-4.08), cardiopulmonary disease (OR, 1.91; 95% CI… Show more

Results: The study comprised 171 416 pediatric patients; the median age was 8 years (IQR, 2-14 years), and 50.28% were male. Of these patients, 17 065 (9.96%) had severe COVID-19. The random intercept model showed that the following cardiovascular conditions were associated with severe COVID-19: cardiac arrest (odds ratio [OR], 9.92; 95% CI, 6.93-14.20), cardiogenic shock (OR, 3.07; 95% CI, 1.90-4.96), heart surgery (OR, 3.04; 95% CI, 2.26-4.08), cardiopulmonary disease (OR, 1.91; 95% CI, 1.56-2.34), heart failure (OR, 1.82; 95% CI, 1.46-2.26), hypotension (OR, 1.57; 95% CI, 1.38-1.79), nontraumatic cerebral hemorrhage (OR, 1.54; 95% CI, 1.24-1.91), pericarditis (OR, 1.50; 95% CI, 1.17-1.94), simple biventricular defects (OR, 1.45; 95% CI, 1.29-1.62), venous embolism and thrombosis (OR, 1.39; 95% CI, 1.11-1.73), other hypertensive disorders (OR, 1.34; 95% CI, 1.09-1.63), complex biventricular defects (OR, 1.33; 95% CI, 1.14-1.54), and essential primary hypertension (OR, 1.22; 95% CI, 1.08-1.38). Furthermore, 194 of 258 patients (75.19%) with a history of cardiac arrest were younger than 12 years.
Conclusions and Relevance: The findings suggest that some previous or preexisting cardiovascular conditions are associated with increased severity of COVID-19 among pediatric patients in the US and that morbidity may be increased among individuals children younger than 12 years with previous cardiac arrest. Show less

Autism spectrum disorder can be reliably diagnosed at 18 months, yet significant diagnostic delays persist in the US. This double-blinded, multi-site, prospective, active comparator cohort study tested the accuracy of an AI-based Software as a Medical Device designed to aid primary care HCPs in diagnosing ASD. The Device combines behavioral features from three distinct inputs (a caregiver questionnaire, analysis of two short home videos, and an HCP questionnaire) in a gradient boosted decision… Show more

Autism spectrum disorder can be reliably diagnosed at 18 months, yet significant diagnostic delays persist in the US. This double-blinded, multi-site, prospective, active comparator cohort study tested the accuracy of an AI-based Software as a Medical Device designed to aid primary care HCPs in diagnosing ASD. The Device combines behavioral features from three distinct inputs (a caregiver questionnaire, analysis of two short home videos, and an HCP questionnaire) in a gradient boosted decision tree machine learning algorithm to produce either an ASD positive, ASD negative, or indeterminate output. This study compared Device outputs to diagnostic agreement by two or more independent specialists in a cohort of 18–72-month-olds with developmental delay concerns (425 study completers, 36% female, 29% ASD prevalence). Device output PPV for all study completers was 80.8% and NPV was 98.3%. For the 31.8% of participants who received a determinate output (ASD positive or negative) Device sensitivity was 98.4% and specificity was 78.9%. The Device’s indeterminate output acts as a risk control measure when inputs are insufficiently granular to make a determinate recommendation with confidence. If this risk control measure were removed, the sensitivity for all study completers would fall to 51.6%, and specificity would fall to 18.5%. Among participants for whom the Device abstained from providing a result, specialists identified that 91% had one or more complex neurodevelopmental disorders. No significant differences in Device performance were found across participants’ sex, race/ethnicity, income, or education level. For nearly a third of this primary care sample, the Device enabled timely diagnostic evaluation with a high degree of accuracy. The Device shows promise to significantly increase the number of children able to be diagnosed with ASD in a primary care setting, potentially facilitating earlier intervention and more efficient use of specialist resources. Show less

The Extension for Community Healthcare Outcomes (ECHO) Autism Program trains participating clinicians to screen, diagnose and care for children with autism spectrum disorder (ASD) in primary care settings. This study will assess the feasibility and impact of integrating an artificial intelligence (AI)-based ASD diagnosis aid (the ‘Device’) into the existing ECHO Autism STAT diagnosis model. The prescription-only Software as a Medical Device is designed for use in 18-72-month-olds at risk for… Show more

The Extension for Community Healthcare Outcomes (ECHO) Autism Program trains participating clinicians to screen, diagnose and care for children with autism spectrum disorder (ASD) in primary care settings. This study will assess the feasibility and impact of integrating an artificial intelligence (AI)-based ASD diagnosis aid (the ‘Device’) into the existing ECHO Autism STAT diagnosis model. The prescription-only Software as a Medical Device is designed for use in 18-72-month-olds at risk for developmental delay. The Device combines behavioral features from 3 distinct inputs (a caregiver questionnaire, two short home videos analyzed by trained video analysts, and a healthcare provider questionnaire) in a machine learning algorithm to produce either an ‘ASD positive’, ‘ASD negative’ or ‘no result (indeterminate)’ output. The Device is not a standalone diagnostic, however, healthcare providers can leverage the Device output, in conjunction with their clinical judgment, when formulating a diagnosis. Registered with ClinicalTrials.gov (Protocol Identifier: NCT05223374) Show less

Since the U.S. Centers for Disease Control and Prevention began tracking the prevalence of autism spectrum disorder (ASD) over twenty years ago, rates have tripled, with an estimated one in 44 children now receiving a diagnosis. Early ASD diagnosis and intervention during the critical neurodevelopmental window is recommended to enhance long-term outcomes [2-4]; yet many families experience diagnostic delays and challenges accessing services. Diagnostic barriers include long waits for specialist… Show more

Since the U.S. Centers for Disease Control and Prevention began tracking the prevalence of autism spectrum disorder (ASD) over twenty years ago, rates have tripled, with an estimated one in 44 children now receiving a diagnosis. Early ASD diagnosis and intervention during the critical neurodevelopmental window is recommended to enhance long-term outcomes [2-4]; yet many families experience diagnostic delays and challenges accessing services. Diagnostic barriers include long waits for specialist assessment, lengthy and fragmented evaluation processes, and limited primary care diagnostic capacity. Race, ethnicity, gender, geography, and socioeconomic status contribute to further delays for some populations. Even after an ASD diagnosis is received, health services may struggle to fund and deliver targeted and timely interventions to the rapidly growing number of children requiring treatment. Data driven approaches to scale, streamline and enhance the quality of diagnostic and therapeutic ASD care available to families are urgently required. This narrative literature review considers the practice change potential of one such approach: Artificial Intelligence (AI) applied to the field of ASD. After providing
a brief overview of AI in healthcare, we review a number of ASD specific AI-based approaches and consider their potential to augment current ASD diagnostic or treatment pathways. Key challenges associated with integrating AI based technologies into clinical practice are also considered. Show less

This special article uses a biosocial-ecological framework to discuss findings in the literature on racial, ethnic, and sociodemographic diagnostic disparities in autism spectrum disorder. We draw explanations from this framework on the complex and cumulative influences of social injustices across interpersonal and systemic levels.

The third edition of Manual of Traumatic Brain Injury offers a thorough revision of the popular evidence-based guide to understanding and managing all levels of traumatic brain injury. Comprehensive in scope and concise in format, this reference describes the spectrum of injury from mild to severe and the continuum of care from initial injury to management of chronic sequelae. Chapters are designed with a practical clinical focus for targeted retrieval of content by topic area and for… Show more

The third edition of Manual of Traumatic Brain Injury offers a thorough revision of the popular evidence-based guide to understanding and managing all levels of traumatic brain injury. Comprehensive in scope and concise in format, this reference describes the spectrum of injury from mild to severe and the continuum of care from initial injury to management of chronic sequelae. Chapters are designed with a practical clinical focus for targeted retrieval of content by topic area and for self-review.

The text is organized into five sections. Part I addresses fundamental concepts necessary for understanding the underpinning of clinical decision-making. Part II is dedicated to mild TBI, including sport-related concussion, with chapters covering topics from natural history to associated somatic disorders, post-concussion syndrome, and PTSD. Part III covers moderate to severe TBI and details prehospital emergency and ICU care, rehabilitation, treatment of related conditions, and postinjury outcomes. Part IV focuses on TBI-related complications, including posttraumatic seizures, spasticity, behavioral and sleep disturbances, and chronic traumatic encephalopathy (CTE). Part V reviews special considerations in selected populations such as pediatric TBI and TBI in the military, as well as medicolegal and ethical considerations in TBI, complementary and alternative medicine, and return to work considerations.

Each chapter includes boxed Key Points which underscore major clinical takeaways, Study Questions to facilitate self-assessment and further emphasize core chapter content, and an Additional Reading list for a deeper dive into chapter concepts. Significant updates incorporating recent advancements in the field, combined with the clinical acumen of its experienced contributors, make this third edition the essential manual for healthcare professionals caring for individuals with traumatic brain injury. Show less

Objective: The purpose of this report is to provide insight from pediatric stakeholders with a shared desire to facilitate a revision of the current United States regulatory pathways for the development of pediatric healthcare devices. Methods: On August 5, 2020, a group of innovators, engineers, professors and clinicians met to discuss challenges and opportunities for the development of new medical devices for pediatric health and the importance of creating a regulatory environment that… Show more

Objective: The purpose of this report is to provide insight from pediatric stakeholders with a shared desire to facilitate a revision of the current United States regulatory pathways for the development of pediatric healthcare devices. Methods: On August 5, 2020, a group of innovators, engineers, professors and clinicians met to discuss challenges and opportunities for the development of new medical devices for pediatric health and the importance of creating a regulatory environment that encourages and accelerates the research and development of such devices. On January 6, 2021, this group joined regulatory experts at a follow-up meeting. Results: One of the primary issues identified was the need to present decision-makers with opportunities that change the return-on-investment balance between adult and pediatric devices to promote investment in pediatric devices. Discussion/Conclusion: Several proposed strategies were discussed, and these strategies can be divided into two broad categories: 1. Removal of real and perceived barriers to pediatric device innovation; 2. Increasing incentives for pediatric device innovation. Show less

Background
A wide array of existing instruments are commonly used to assess childhood behavior and development for the evaluation of social, emotional and behavioral disorders such as Autism Spectrum Disorder (ASD), attention-deficit/hyperactivity disorder (ADHD), and anxiety. Many of these instruments either focus on one diagnostic category or encompass a broad set of childhood behaviors. We analyze a wide range of standardized behavioral instruments and identify a comprehensive, structured… Show more

Background
A wide array of existing instruments are commonly used to assess childhood behavior and development for the evaluation of social, emotional and behavioral disorders such as Autism Spectrum Disorder (ASD), attention-deficit/hyperactivity disorder (ADHD), and anxiety. Many of these instruments either focus on one diagnostic category or encompass a broad set of childhood behaviors. We analyze a wide range of standardized behavioral instruments and identify a comprehensive, structured semantic hierarchical grouping of child behavioral observational features. We use the hierarchy to create Rosetta: a new set of behavioral assessment questions, designed to be minimal yet comprehensive in its coverage of clinically relevant behaviors. We maintain a full mapping from every functional feature in every covered instrument to a corresponding question in Rosetta.

Results
In all, 209 Rosetta questions are shown to cover all the behavioral concepts targeted in the eight existing standardized instruments.

Conclusion
The resulting hierarchy can be used to create more concise instruments across various ages and conditions, as well as create more robust overlapping datasets for both clinical and research use. Show less

Highlights
•Patients with cardiovascular diseases (CVD) are at high risk of severe COVID-19.
•Machine learning algorithms may help to better identify the most-at-risk CVD patients.
•Individual base learners/algorithms provide good performances in this prediction task.
•Super learning will further improve the performance of base learners.
•The improvements through super learning may result in greater impact of clinical interventions.

Abstract
Background
Pediatric oncology patients have high rates of hospital readmission but there is a dearth of research into risk factors for unplanned 30‐day readmissions among this high‐risk population.

Aim
In this study, we built a statistical model to provide insight into risk factors of unplanned readmissions in this pediatric oncology.

Methods
We retrieved 32 667 encounters from 10 418 pediatric patients with a neoplastic condition from 16 hospitals in the… Show more

Abstract
Background
Pediatric oncology patients have high rates of hospital readmission but there is a dearth of research into risk factors for unplanned 30‐day readmissions among this high‐risk population.

Aim
In this study, we built a statistical model to provide insight into risk factors of unplanned readmissions in this pediatric oncology.

Methods
We retrieved 32 667 encounters from 10 418 pediatric patients with a neoplastic condition from 16 hospitals in the Cerner Health Facts Database and built a mixed‐effects model with patients nested within hospitals for inference on 75% of the data and reserved the remaining as an independent test dataset.

Results
The mixed‐effects model indicated that patients with acute lymphoid leukemia (in relapse), neuroblastoma, rhabdomyosarcoma, or bone/cartilage cancer have increased odds of readmission. The number of cancer medications taken by the patient and the administration of chemotherapy were associated with increased odds of readmission for all cancer types. Wilms Tumor had a significant interaction with administration of chemotherapy, indicating that the risk due to chemotherapy is exacerbated in patients with Wilms Tumor. A second two‐way interaction between recent history of chemotherapy treatment and infections was associated with increased odds of readmission. The area under the receiver operator characteristic curve (and corresponding 95% confidence interval) of the mixed‐effects model was 0.714 (0.702, 0.725) on the independent test dataset.

Conclusion
Readmission risk in oncology is modified by the specific type of cancer, current and past administration of chemotherapy, and increased health care utilization. Oncology‐specific models can provide decision support where model built on other or mixed population has failed. Show less

Background
Unplanned readmission is one of many measures of the quality of care of pediatric patients with neurological conditions. In this multicenter study, we searched for novel risk factors of readmission of patients with neurological conditions.

Methods
We retrieved hospitalization data of patients less than 18 years with one or more neurological conditions. This resulted in a total of 105,834 encounters from 18 hospitals. We included data on patient demographics, prior… Show more

Background
Unplanned readmission is one of many measures of the quality of care of pediatric patients with neurological conditions. In this multicenter study, we searched for novel risk factors of readmission of patients with neurological conditions.

Methods
We retrieved hospitalization data of patients less than 18 years with one or more neurological conditions. This resulted in a total of 105,834 encounters from 18 hospitals. We included data on patient demographics, prior healthcare resource utilization, neurological conditions, number of other conditions/diagnoses, number of medications, and number of surgical procedures performed. We developed a random intercept logistic regression model using stepwise minimization of Akaike Information Criteria for variable selection.

Results
The most important neurological conditions associated with unplanned pediatric readmissions include hydrocephalus, inflammatory diseases of the central nervous system, sleep disorders, disease of myoneural junction and muscle, other central nervous system disorder, other spinal cord conditions (such as vascular myelopathies, and cord compression), and nerve, nerve root and plexus disorders. Current and prior healthcare resource utilization variables, number of medications, other diagnoses, and certain inpatient surgical procedures were associated with changes in odds of readmission. The area under the receiver operator characteristic curve (AUROC) on the independent test set is 0.733 (0.722, 0.743).

Conclusions
Pediatric patients with certain neurological conditions are more likely to be readmitted than others. However, current and prior healthcare resource utilization remain some of the strongest indicators of readmission within this population as in the general pediatric population. Show less

Highlights
• Agile data science in healthcare is becoming a necessity given the COVID-19 pandemic and urgency of patient treatment.
• The proposed agile data science approach is based on the Scrumban framework.
• Publicly available healthcare datasets and cloud-based infrastructure enable adoption of the agile framework.
• Collaboration between physicians and data scientists needed in order to implement the agile framework.

There is a shortage of medical informatics and data science platforms using cloud computing on electronic medical record (EMR) data, and with computing capacity for analyzing big data. We implemented, described, and applied a cloud computing solution utilizing the fast health interoperability resources (FHIR) standardization and state-of-the-art parallel distributed computing platform for advanced analytics.

Objectives: The rate of pediatric 7-day unplanned readmissions is often seen as a measure of quality of care, with high rates indicative of the need for improvement of quality of care. In this study, we used machine learning on electronic health records to study predictors of pediatric 7-day readmissions. We ranked predictors by clinical significance, as determined by the magnitude of the least absolute shrinkage and selection operator regression coefficients.

Methods: Data consisting of… Show more

Objectives: The rate of pediatric 7-day unplanned readmissions is often seen as a measure of quality of care, with high rates indicative of the need for improvement of quality of care. In this study, we used machine learning on electronic health records to study predictors of pediatric 7-day readmissions. We ranked predictors by clinical significance, as determined by the magnitude of the least absolute shrinkage and selection operator regression coefficients.

Methods: Data consisting of 50 241 inpatient and observation encounters at a single tertiary pediatric hospital were retrieved; 50% of these patients' data were used for building a least absolute shrinkage and selection operator regression model, whereas the other half of the data were used for evaluating model performance. The categories of variables included were demographics, social determinants of health, severity of illness and acuity, resource use, diagnoses, medications, psychosocial factors, and other variables such as primary care no show.

Results: Previous hospitalizations and readmissions, medications, multiple comorbidities, longer current and previous lengths of stay, certain diagnoses, and previous emergency department use were the most significant predictors modifying a patient's risk of 7-day pediatric readmission. The model achieved an area under the curve of 0.778 (95% confidence interval 0.763-0.793).

Conclusions: Predictors such as medications, previous and current health care resource use, history of readmissions, severity of illness and acuity, and certain psychosocial factors modified the risk of unplanned 7-day readmissions. These predictors are mostly unmodifiable, indicating that intervention plans on high-risk patients may be developed through discussions with patients and parents to identify underlying modifiable causal factors of readmissions. Show less

Traumatic brain injury causes significant morbidity in youth, and headache is the most common postconcussive symptom. No established guidelines exist for pediatric post-traumatic headache management. We aimed to characterize common clinical practices of child neurologists. Of 95 practitioners who completed our survey, most evaluate <50 pediatric concussion patients per year, and 38.9% of practitioners consistently use International Classification of Headache Disorders criteria to diagnose… Show more

Traumatic brain injury causes significant morbidity in youth, and headache is the most common postconcussive symptom. No established guidelines exist for pediatric post-traumatic headache management. We aimed to characterize common clinical practices of child neurologists. Of 95 practitioners who completed our survey, most evaluate <50 pediatric concussion patients per year, and 38.9% of practitioners consistently use International Classification of Headache Disorders criteria to diagnose post-traumatic headache. Most recommend nonsteroidal anti-inflammatory drugs as abortive therapy, though timing after injury and frequency of use varies, as does the time when providers begin prophylactic medications. Amitriptyline, topiramate, and vitamins/supplements are most commonly used for prophylaxis. Approach to rest and return to activities varies; one-third recommend rest for 1 to 3 days and then progressive return, consistent with current best practice. With no established guidelines for pediatric post-traumatic headache management, it is not surprising that practices vary considerably. Further studies are needed to define the best, evidence-based management for pediatric post-traumatic headache. Show less

Abstract
Early identification and intervention of speech and language delays in children contribute to better communication and literacy skills for school readiness and are protective against behavioral and mental health problems. Through collaboration between the data science and clinical teams at Cognoa, we designed Storytime, an interactive storytelling experience on a mobile device using a virtual avatar to mediate speech and language screening for children ages 4 to 6 years old. Our… Show more

Abstract
Early identification and intervention of speech and language delays in children contribute to better communication and literacy skills for school readiness and are protective against behavioral and mental health problems. Through collaboration between the data science and clinical teams at Cognoa, we designed Storytime, an interactive storytelling experience on a mobile device using a virtual avatar to mediate speech and language screening for children ages 4 to 6 years old. Our proof-of-concept study collects Storytime session footage from 71 pairs of parents and children including 57 typically developing children and 14 children with a current or prior history of communication impairments. Initial findings suggest that participating children verbally engaged with the video avatar without significant differences in performance across age, gender, and experimental location, leading to promising implications for using Storytime as a future tracking tool with automated feature analyses to detect speech and language delays. Show less

After sustaining a concussion or mild traumatic brain injury, headaches are one of the most common complaints. The pathophysiologic changes that occur in the setting of injury likely contribute to or cause posttraumatic headaches. Posttraumatic headaches often present as migraine or tension-type headaches. Unlike pain from other types of injuries, headaches following mild traumatic brain injury are more likely to persist. Preexisting conditions such as migraine and mood disorders may influence… Show more

After sustaining a concussion or mild traumatic brain injury, headaches are one of the most common complaints. The pathophysiologic changes that occur in the setting of injury likely contribute to or cause posttraumatic headaches. Posttraumatic headaches often present as migraine or tension-type headaches. Unlike pain from other types of injuries, headaches following mild traumatic brain injury are more likely to persist. Preexisting conditions such as migraine and mood disorders may influence posttraumatic headache and complicate management. Patients are at high risk to overuse abortive medications and develop medication overuse headache. Headache hygiene and early education are essential for effective management. Abortive medications include nonsteroidal anti-inflammatory drugs and triptans. Preventive medications include tricyclic antidepressants and antiepileptics. Patients who fail outpatient therapies may benefit from referral for intravenous medications in the emergency department. Patients with persistent posttraumatic headache may benefit from multimodal treatments including physical rehabilitation and pain-focused cognitive-behavioral therapies. Show less

Highlights
• Sibling response to initial AED is predictive of success with the same AED in the other sibling.
• It may help clinicians to select better AEDs for patients with epilepsy.
• Genetic factors may play an important role in epilepsy treatment response.

Manual of Traumatic Brain Injury: Assessment and Management, Second Edition is a comprehensive evidence-based guide to brain injury diagnosis, treatment, and recovery, delivered in a succinct format designed for targeted access to essential content.

This concise text, featuring internationally known contributors drawn from leading TBI programs, is organized into five sections. Part I discusses fundamental concepts needed to provide a context for clinical decision-making. Part II covers… Show more

Manual of Traumatic Brain Injury: Assessment and Management, Second Edition is a comprehensive evidence-based guide to brain injury diagnosis, treatment, and recovery, delivered in a succinct format designed for targeted access to essential content.

This concise text, featuring internationally known contributors drawn from leading TBI programs, is organized into five sections. Part I discusses fundamental concepts needed to provide a context for clinical decision-making. Part II covers mild TBI, from natural history to sportrelated concussion, postconcussion syndrome, and more. Part III focuses on moderate to severe TBI and contains chapters on prehospital, emergency and ICU care, rehabilitation, community reintegration, management of associated impairments, and postinjury outcomes. Part IV covers the complications and long-term sequelae that may arise in patients with TBI, including spasticity, movement disorders, posttraumatic seizures, hydrocephalus, behavioral and sleep disturbances, and chronic traumatic encephalopathy (CTE). Part V focuses on special considerations and resources, including issues specific to selected populations or injury environments (military, pediatric, workers' compensation and older patients), as well as return to work and medicolegal issues in TBI.

Comprehensively updated to reflect the current state of the art in this rapidly evolving field, this book is a must-have for neurologists, physiatrists, primary care physicians, mental health professionals, social workers, and other health care providers who treat TBI patients. Show less

Cocaine use is associated with ischemic stroke through unique mechanisms, including reversible vasospasm, drug-induced arteritis, enhanced platelet aggregation, cardioembolism, and hypertensive surges. To date, no study has described disability in patients with cocaine-related ischemic stroke. The present study compared risk factors, comorbidities, complications, laboratory findings, medications, and outcomes in patients with cocaine-related (n = 41) and non–cocaine-related (n = 221) ischemic… Show more

Cocaine use is associated with ischemic stroke through unique mechanisms, including reversible vasospasm, drug-induced arteritis, enhanced platelet aggregation, cardioembolism, and hypertensive surges. To date, no study has described disability in patients with cocaine-related ischemic stroke. The present study compared risk factors, comorbidities, complications, laboratory findings, medications, and outcomes in patients with cocaine-related (n = 41) and non–cocaine-related (n = 221) ischemic stroke (n = 147) and transient ischemic attack (n = 115) in 3 academic hospitals. The patients with cocaine-related stroke were younger (mean age, 51.9 years vs 59.1 years; P = .0008) and more likely to be smokers (95% vs 62.9%; P < .004). The prevalence of arrhythmias was significantly higher in the patients with cocaine-related stroke, and that of diabetes was significantly higher in those with non–cocaine-related strokes. The prevalence of hypertension and lipid profiles were similar in the 2 groups; however, those with cocaine-related stroke were less likely to receive statins. Antiplatelet use was similar in the 2 groups. Survivors of both groups had similar modified Rankin scores and lengths of hospital stay. In the older urban population, smoking and cocaine use may coexist with other cerebrovascular risk factors, and cocaine-related strokes have similar morbidities and mortality as non–cocaine-related strokes. Moreover, because the patients with cocaine-related stroke is younger, they have an earlier morbidity. New strategies for effective stroke prevention interventions are needed in this subgroup. Show less

Abstract: This paper reviews the spectrum of vasculitides that affect the brain, specifically focusing on primary angiitis of the central nervous system (CNS) and how they relate to stroke in the pediatric population. CNS vasculitis accounts for a substantial portion of pediatric stroke. The extent and severity of the stroke is variable. Hemiparesis and encephalopathy occur commonly, but are not specific to CNS vasculitis. The non-specific presentation and results of investigations make… Show more

Abstract: This paper reviews the spectrum of vasculitides that affect the brain, specifically focusing on primary angiitis of the central nervous system (CNS) and how they relate to stroke in the pediatric population. CNS vasculitis accounts for a substantial portion of pediatric stroke. The extent and severity of the stroke is variable. Hemiparesis and encephalopathy occur commonly, but are not specific to CNS vasculitis. The non-specific presentation and results of investigations make diagnosis difficult. Lack of controlled treatment trials complicates the management. Blood inflammatory markers, cerebrospinal fluid analysis, multiple imaging techniques including conventional angiography, and brain biopsy form the routine workup. Therapeutic modalities including anti-platelet agents, corticosteroids, cyclophosphamide, and other immunomodulatory agents have been used with the vertical line on the right is cutting into the text apparent success, but the evidence is mostly anecdotal. Show less