www.ijcspub.

org © 2023 IJCSPUB | Volume 13, Issue 2 May 2023 | ISSN: 2250-1770

Overview of AI & NLP based Medical Chatbot

Systems
1
Kunal Vaidya, 2Pawan Raut, 3Kalpesh Kolhe, 4Aditya Zode, 5Shivam Sonone,
6
Prof. Harshada M. Raghuwanshi
1,2,3,4, 5
U.G. Students, Department of Information Technology, Sipna C.O.E.T., Amravati, Maharashtra, India
6
Assistant Professor, Department of Information Technology, Sipna C.O.E.T., Amravati, Maharashtra, India

Abstract: This essay analyses and contrasts the material that has already been written about the numerous chatbots that provide
virtual medical aid as well as the supporting systems. Although there are many research contributions in the literature, in this article
we have critically and thoroughly examined recent research and review papers that are relevant to AI-based health care chatbots.
The present techniques are divided into groups based on the fundamental ideas included into their mechanisms. The idea put out by
the concerned writers, the experimentation methodology, and the performance assessment criteria are highlighted. Also noted are
the researchers' assertions. The flaws that have been found are highlighted together with our findings from the thorough literature
review. The comparative analysis of various healthcare chatbot approaches, which is a requirement for resolving distant health
issues, would be greatly aided by this work. Finally, we have suggested developing our own chatbot for the healthcare industry. We
combined AI, NLP, and machine learning. Before consulting a medical professional or visiting a doctor, one can quickly and simply
diagnose the sickness or ailment, and the information can be delivered remotely. This chatbot may be conveniently accessed at any
time and location and has the inherent ability to lower healthcare costs. Additionally, our suggested strategy is accurate and
consistent.

Index Terms - Artificial Intelligence, Chatbots, Health Care Systems, Machine Learning, Natural Language Programming

All industries are developing and incorporating the newest trends. A chatbot that aids in recognising client issues and remote
troubleshooting are two of the approaches currently being used. In the end, this lowers communication barriers and enhances the
client experience [1]. A chatbot is a piece of intelligent software that can interact with users in a way that is similar to how a human
might. The chatbots are used for customer service, social media marketing, and instant messaging on a variety of platforms and
websites. Artificial intelligence and Natural Language Programming have long been an element of chatbot technology. have
programmes that analyse and recognise spoken speech [2].
The medical industry is no different. Healthcare systems are one of the many areas in which chatbots may be employed. An
interactive medical chatbot engages the user in a one-on-one discussion and questionnaire about their health issues. NLP and AI are
used by these chatbots [3]. The system is given all the data about illnesses and their symptoms, and with the aid of NLP, it will be
able to comprehend the user inquiry and produce the appropriate response. It is necessary to either construct or supply a database
with all the information on medical conditions and disorders. Data from a medical database are retrieved by the chatbot [4]. By posing
numerous questions, the chatbot mimics the interaction between a doctor and patient. These inquiries are based on the patient's prior
information and pre-filled conditions, and a potential diagnosis is formed based on the answers [5]. These bots interact with potential
patients who are browsing the website, assisting them in finding experts, scheduling appointments, and gaining access to the proper
care. The nicest part about a healthcare chatbot is that these types of chatbots can be accessible from anywhere, even in the most
distant place, and they offer patients 24/7 online healthcare help. It aids in the generation of health data and instantly transmits report
information to medical management [6].
The overview of the current methodologies is the primary topic of this study. Though there are many review articles on healthcare
chatbot systems that have been published in the literature, we have critically analysed and systematically summarised the most
current, significant papers. Additionally, this research offers a potential remedy for healthcare chatbot systems. The structure of this
essay is as follows. The systematic presentation of the literature review is done in Section 2, which includes a synopsis of the relevant
work as well as the main points of the connected methods. The reason for the formulation of the problem and the discovered problem
is provided in Section 3. The operation of the suggested technique is thoroughly explained in Section 4. The conclusion is contained
in Section 5.

IJCSP23B1281 International Journal of Current Science (IJCSPUB) www.ijcspub.org 355

www.ijcspub.org © 2023 IJCSPUB | Volume 13, Issue 2 May 2023 | ISSN: 2250-1770
I. LITERATURE REVIEW
The available literature that is pertinent to healthcare chatbot systems and the mechanics underlying them is critically analysed
in this section. Although there are many research contributions in the literature, we have only examined the most recent, significant,
and useful study and review publications here. Based on the fundamental ideas used in the mechanisms, the existing techniques are
divided into several categories. The authors' concepts, the platform they employed for their experiments, and the effectiveness of
their systems are all highlighted. Additionally, their claims are emphasised. Finally, a summary of the conclusions relating to the
research articles that were read and examined is provided. The section's conclusion includes the cause of the detected problem.

Bushra Kidwai and Nadesh RK created a diagnostic chatbot in 2020 utilising the most recent machine learning methods. A
decision tree technique is applied for symptom mapping and diagnosis [1]. The system is given all the data about illnesses and their
symptoms, and with the aid of NLP, it will be able to comprehend the user inquiry and produce the appropriate response. The
information is retrieved from a medical database that contains information on 150 disorders. By posing numerous questions, the
chatbot mimics the interaction between a doctor and patient. These inquiries are based on the patient's prior information and pre-
filled conditions, and a potential diagnosis is formed depending on the patient's responses. The system can be interactive, according
to the creators. Round-Trip Time is the performance assessment parameter that is employed.

A healthcare chatbot that uses NLP was proposed by Papiya Mahajan et al. for 2020. The chatbot system may offer voice or text
support [2]. According on user symptoms, the created chatbot is said to deliver ailment details. The N-gram algorithm, TF-IDF
algorithm, and Cosine similarity algorithm are the three algorithms employed in this technique to create a health care chatbot system.
The system also gives information on the doctor. People will therefore be aware of their health and be properly protected. The
authors assert that their chatbot offers individualised diagnoses based on symptoms. However, no performance evaluation metric is
offered to support the assertion.

Harsh Mendapara and colleagues created a healthcare chatbot system based on AI and NLP in 2021. Before seeing a doctor, a
person may use the chatbot to self-diagnose their sickness and learn more about their particular disease [3]. The authors utilised
chatterbot library to train a chatbot using AI. Additionally, they have taught the chatbot to recognise particular words and phrases
so that the user's purpose may be understood. The system is then provided the information that has been gathered. Personal
information provided by the user will be kept in the database. The chatbot will use NLP to enquire about symptoms. It will list
potential diseases and make medication recommendations for those with known illnesses. A clinic appointment can also be made
via the system. The resulting findings, according to the authors, are precise and quick.

Mr. Niraj A. Wanjari and colleagues created a chatbot based on AI and NLP in 2022. The authors assert that their chatbot
improves accessibility and lowers total healthcare costs[4]. The illness is identified by text-to-text communication between the
chatbot system and the patients. Additionally, the right course of action is advised. We discovered that the bot architecture is in its
infancy and is not fully formed. The paper is more of an assertion than a practical application. To support the authors' assertions,
no performance assessment criteria is offered. However, this project has a lot of room for expansion. It is possible to create a mobile
assistant that many people can utilise. This can be exact and will eventually somewhat shorten the time.

M.V. Patil, et al. created an AI-based healthcare conversation bot system in 2021[5]. The suggested medical chatbot deals with
giving medical advise while making decisions in accordance with the patient's request and using dialogues to draw analogies. For
this, a database is constructed. If a certain piece of information isn't in the database, a search engine will be used to find it and
present it to the user. Three separate outputs from experiments are utilised to demonstrate how well the suggested system recognises
illnesses and other common symptoms. To support the authors' assertions, no performance assessment criteria is offered, though.

Before contacting a doctor, Lekha Athota et al. created a chatbot in 2020 that can identify the illness and offer details[6]. N-
gram, TFIDF, and cosine similarity are used in the ranking and text similarity computation performance assessment parameters.
The information is kept in a database, and the queries are handled by a third party. The keywords are given so that the right response
to the query may be found. Users also have access to a web interface. The authors assert that by protecting user privacy, character
integrity, and accessing pertinent information in response to queries, their chatbot application has been improved in terms of security
and efficacy. To support the authors' assertions, no performance assessment criteria is offered, though.

II. MOTIVATION
Using artificial intelligence chatbots, it is now feasible to have natural language conversations with robots. We learned from the
literature that chatbots often perform the same duties as a standard search engine. The essential process flow remains the same even
if the chatbot only provided one output rather than several outputs or results—each time an input is entered, a fresh search is
conducted. Nothing to do with the earlier output. The goal of this study is to develop chatbots into search engines that can process
subsequent searches in relation to the results of prior searches.

III. AIM & OBJECTIVE

• Chatbot is a smart enough to response a particular suggestion from database stored in it.
• To make user friendly GUI based system which is easy to use by user and whenever user enter a text or ask a question.
• To recognize the requirement of user using NLP technology of text.
• CNN algorithm digitization work fast processes the finding answers from data sets on keyboards written in the text.

IJCSP23B1281 International Journal of Current Science (IJCSPUB) www.ijcspub.org 356

 www.ijcspub.org © 2023 IJCSPUB | Volume 13, Issue 2 May 2023 | ISSN: 2250-1770
IV. PROPOSED METHODOLOGY
This approach starts with taking the text input from the user in the android app. We promote user to enter the question for the
chatbot and then we read the entered value in the edit text. Then send the value to the interface of our hosted Program – O. Get the
reply from the Server, Server gives reply in the JSON format. We need to parse it to get the bots response. If bots response is null
then set the response to the error string and show it to the user. Repeat the same task for the continuous chatting with the chatbot.

The following steps explain the implementation details to accomplish the above-mentioned tasks:
1. Read the user input on the Click event of the send button.
2. Send the user input to the program – o hosted server using the HttpUrlConnection. Here we are dealing with the internet stuff
so we need to call this method in the AsyncTask of the android system. AsyncTask runs in the background of the app.
3. Wait till getting the reply from the server. OnPostExcute method of the AsyncTask will get the JSON formatted reply. Then
send it to JSON parser to extract the bot‟s response.
4. After extracting the bot‟s response, we need to display it to the user. So, append the response to text view.
5. Then repeat the process for the next user input.

V. COMPARISON WITH EXISTING METHOD

Table1: Comparison with Existing Methods
EXISTING METHODS PROPOSED APPROACH
UI [Framework] is modarant Advance feature i.e. easy to handle
In this we have to train the data If there is problem in database management system data
against UI
Database is deleted and new database is created Old database is replaced by new database
Text to speech was not there Text to speech
In this conversation was not save Conversation are save
Book an appointment Call system

VI. ADVANTAGES
 24 Hours Availability
 Instant Answers
 Endless Patience
 Programmability
 Personalization
VII. CONCLUSION
Smart solutions are important for the success of any business. From providing 24/7 customer service, improving current marketing
activities, saving time spent on engaging with users to improving internal processes, chatbots can yield the much-needed competitive
advantage. If you are looking to develop a chatbot, the best thing to do is to approach a company that will understand your business
needs to develop a chatbot that helps you achieve your business goals.

REFERENCES
[1] Bushra Kidwai, Nadesh RK, “Design and Development of Diagnostic Chabot for supporting Primary Health Care Systems”, International
Conference on Computational Intelligence and Data Science (ICCIDS 2019)
[2] Papiya Mahajan, Rinku Wankhade, Anup Jawade, Pragati Dange, Aishwarya Bhoge, “Healthcare Chatbot using Natural Language Processing”,
International Research Journal of Engineering and Technology (IRJET), Volume 07, Issue 11, Nov 2020
[3] Harsh Mendapara, Suhas Digole, Manthan Thakur, Anas Dange, “AI Based Healthcare Chatbot System by Using Natural Language Processing”,
International Journal of Scientific Research and Engineering Development, Volume 4, Issue 2, Mar- Apr 2021
[4] Mr. Niraj A. Wanjari, et.al., “AI Healthcare Chatbot Using Natural Language Processing”, International Research Journal of Modernization in
Engineering Technology and Science, Volume 04, Issue 05, May-2022
[5] M.V. Patil, Subhawna, Priya Shree, Puneet Singh, “AI based healthcare chat bot system”, International Journal of Scientific & Engineering
Research, Volume 12, Issue 7, July-2021
[6] Lekha Athota, Vinod Kumar Shukla, Vinod Kumar Shukla, Ajay Rana, “Chatbot for Healthcare System Using Artificial Intelligence”, 2020 8th
International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO) Amity University,
Noida, India. June 4-5, 2020
[7] Vikramsingh R. Parihar, Heartbeat and Temperature Monitoring System for Remote Patients using Arduino, International Journal of Advanced
Engineering Research and Science (IJAERS), Volume 4, Issue 5, PP 55-58, May 2017

IJCSP23B1281 International Journal of Current Science (IJCSPUB) www.ijcspub.org 357

www.ijcspub.org © 2023 IJCSPUB | Volume 13, Issue 2 May 2023 | ISSN: 2250-1770
Author Biography

Prof. Harshada M. Raghuwanshi completed M.E. and B.E in Computer Science &
Engineering from Sant Gadge Baba Amravati University, India. She has granted
work in her U.G. project B.E level entitled Air quality monitoring system using
sensor. She has worked in Atos Syntel as Associate Consultant. She is now working
as an Assistant Professor in Sipna College of Engineering & Technology, Amravati.
She has completed 8-month internship in Python Web development in Ctronics
Company as well as Internship in Java with Oracle (ISO 9001:2015 certified
Company). She also had completed Mobile App development Internship with
Orange Bytes. Her area of interests is Digital Image Processing and Embedded Systems. She has published
numerous research papers in reputed journals. She is an author of 1 book which is successfully published
worldwide and is translated into 7 different languages like French, German, Spanish, etc.

IJCSP23B1281 International Journal of Current Science (IJCSPUB) www.ijcspub.org 358