Volume 9, Issue 3, March – 2024 International Journal of Innovative Science and Research Technology

ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24MAR804

AI-Based Medical Chatbot for Disease Prediction

Ashish Zagade1; Vedant Killedar2; Onkar Mane3; Ganesh Nitalikar4; Smita Bhosale5
1,2,3,4,5
Department of Computer Engineering, Sinhgad Institute of Technology and Science, Pune

Abstract:- This research paper presents the development Moreover, Prakhar Srivastava et al. [2] worked on
and implementation of an AI-based medical chatbot for "Automatized Medical Chatbot (Medibot)" showcased the
disease prediction. Leveraging machine learning and feasibility of developing chatbots tailored to specific
artificial intelligence technologies, the chatbot utilizes medical domains. Presented at the International Conference
natural language processing (NLP) to understand user on Power Electronics & IoT Applications in Renewable
queries and provide accurate information, guidance, and Energy and its Control (PARC) in 2020, their research
assistance for various infectious diseases. Motivated by underscores the importance of personalized healthcare
the global spread of infectious diseases and the need for solutions in improving patient outcomes and healthcare
accessible healthcare support, the paper outlines the accessibility.
objectives, algorithm design, dataset description, and
application of the chatbot. The algorithm involves In addition, this research study sheds light on the
receiving user input, extracting symptoms, classifying transformative potential of chatbots in reshaping traditional
diseases, and suggesting prevention measures. The healthcare delivery models. By leveraging AI algorithms
dataset, structured in JSON format, facilitates training and real-time data, these chatbots can provide timely
and pattern recognition. The chatbot interface, medical assistance and information to users, thereby
accessible across platforms, offers information on addressing the growing need for accessible healthcare
symptoms, prevention measures, hospital bed support.
availability, and medication options. In conclusion, the
research highlights the potential of AI-based chatbots in It highlights the role of machine learning techniques in
revolutionizing healthcare accessibility and personalized enhancing the predictive capabilities of medical chatbots.
diagnosis, thereby bridging the gap between users and By analyzing symptom data and patient information, these
healthcare systems. chatbots can assist in early disease detection and
recommend appropriate treatment options.
Keywords:- AI-Based Chatbot, Disease Prediction, Machine
Learning, Natural Language Processing (NLP), Healthcare Building upon the insights and advancements
Accessibility, Infectious Diseases, Personalized Diagnosis, highlighted in these seminal works, this research paper aims
Healthcare Support. to further explore the potential of AI-based medical chatbots
for disease prediction. By examining the algorithm design,
I. INTRODUCTION dataset description, and application of these chatbots, this
paper seeks to contribute to the ongoing discourse
In recent years, the integration of machine learning and surrounding the integration of AI technologies in healthcare
artificial intelligence (AI) in healthcare has sparked delivery. Through a comprehensive analysis of the existing
considerable interest and innovation. Leveraging these literature and empirical findings, this research endeavors to
advanced technologies, researchers have developed a provide valuable insights into the future prospects of AI-
diverse array of applications aimed at enhancing healthcare powered chatbots in revolutionizing healthcare accessibility
accessibility, diagnosis, and treatment. Among these, AI- and personalized diagnosis.
based medical chatbots have emerged as promising tools for
disease prediction and personalized healthcare support. Fig-1 illustrates the system architecture that outlines
the key components involved in AI-Based Medical Chatbot
Inspired by the groundbreaking work of researchers, for Disease Prediction System.
Lekha Athota et al. [1] the potential of AI-powered chatbots
in revolutionizing healthcare delivery becomes evident.
Through the utilization of natural language processing
(NLP) techniques, these chatbots are capable of
understanding user queries and providing accurate
information, guidance, and assistance for various medical
conditions.

IJISRT24MAR804 www.ijisrt.com 763

Volume 9, Issue 3, March – 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24MAR804

 Literature Survey on Chatbot for Disease Prediction and

Treatment Recommendation using ML:
In the research by Mathew et al. [4]. Their study
focused on enhancing the predictive capabilities of medical
chatbots through machine learning techniques. By analyzing
symptom data and patient information, their chatbot
facilitated early disease detection and recommended
appropriate treatment options, thereby improving patient
outcomes and healthcare efficiency.

Collectively, these studies highlight the significant

advancements and potential applications of AI-based
medical chatbots in disease prediction and healthcare
delivery. By leveraging AI technologies, NLP algorithms,
and machine learning techniques, these chatbots offer
promising solutions to address various healthcare
challenges, including accessibility, efficiency, and
Fig 1 System Architecture personalized diagnosis. However, further research is needed
to explore the scalability, reliability, and effectiveness of
II. LITERATURE SURVEY these chatbots in real-world healthcare settings.
The literature on AI-based medical chatbots for disease  Summary of Literature Review
prediction encompasses a wide range of studies exploring The literature survey above, highlights the
the integration of artificial intelligence and machine learning transformative potential of integrating artificial intelligence
in healthcare delivery. Several seminal works have and machine learning in healthcare delivery. Seminal studies
contributed significantly to this burgeoning field, shedding have demonstrated the effectiveness of chatbots in providing
light on various aspects of chatbot development, accurate medical information, personalized healthcare
implementation, and effectiveness. support, and early disease detection. Leveraging natural
language processing (NLP) techniques and machine learning
 Literature Survey on Chatbot for Healthcare System algorithms, these chatbots offer promising solutions to
Using AI: address healthcare challenges such as accessibility,
The survey for research paper [1], Athota et al. Their efficiency, and personalized diagnosis. Further research is
research demonstrated the potential of AI-powered chatbots needed to explore the scalability and reliability of these
in providing accurate medical information and assistance to chatbots in real-world healthcare settings.
users. By leveraging natural language processing (NLP)
techniques, their chatbot exhibited the capability to III. FUTURE SCOPE & INCREMENTATIONS
understand user queries and respond with relevant
healthcare advice.  Enhanced Disease Prediction Models: Future iterations
of the AI-based medical chatbot can incorporate more
 Literature Survey on Automatized Medical Chatbot advanced machine learning algorithms to improve
(Medibot): disease prediction accuracy. Techniques such as
In research paper [2], Srivastava and Singh, introduced ensemble learning, deep learning, and reinforcement
the concept of an "Automatized Medical Chatbot (Medibot)" learning could be explored to refine the prediction
Their study highlighted the importance of personalized models further.
healthcare solutions in improving patient outcomes and
healthcare accessibility. Through the development of a  Personalized Healthcare Recommendations: The chatbot
specialized chatbot tailored to medical domains, they can be enhanced to provide personalized healthcare
showcased the feasibility of using AI technologies to recommendations based on individual user profiles,
address specific healthcare needs effectively. including demographic information, medical history, and
lifestyle factors. This would involve integrating
 Literature Survey on E-Health Bot to change the Face of additional data sources and refining the recommendation
Medicare: algorithms.
In the paper, Tanmay et al. [3] presented research on
"E-Health Bot to change the Face of Medicare". Their study  Integration of Real-time Data: Incorporating real-time
underscored the transformative potential of chatbots in data sources such as wearable devices, electronic health
reshaping traditional healthcare delivery models. By records (EHRs), and public health databases can enhance
leveraging AI algorithms and real-time data, their chatbot the chatbot's ability to provide up-to-date information
provided timely medical assistance and information to users, and personalized insights. This would require developing
thereby addressing the growing need for accessible robust data integration pipelines and ensuring data
healthcare support. privacy and security.

IJISRT24MAR804 www.ijisrt.com 764

Volume 9, Issue 3, March – 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24MAR804

 Multimodal Interaction: Expanding the chatbot's V. CONCLUSION

capabilities to support multimodal interaction, including
voice input, image recognition, and natural language This research paper underscores the transformative
understanding, can improve user experience and potential of artificial intelligence and machine learning in
accessibility. This would involve integrating speech healthcare delivery. Through studies conducted, it is evident
recognition, computer vision, and multimodal fusion that chatbots have the capability to provide accurate medical
techniques into the chatbot's architecture. information, personalized healthcare support, and early
disease detection. Leveraging natural language processing
 Continuous Learning and Adaptation: Implementing (NLP) techniques and machine learning algorithms, these
mechanisms for continuous learning and adaptation chatbots offer promising solutions to address healthcare
would enable the chatbot to improve over time based on challenges such as accessibility, efficiency, and personalized
user feedback and evolving medical knowledge. diagnosis.
Techniques such as online learning, active learning, and
transfer learning could be leveraged to facilitate Despite the advancements highlighted in these studies,
continuous improvement. further research is needed to explore the scalability and
reliability of AI-based chatbots in real-world healthcare
 Expanded Domain Coverage: While the current focus settings. Additionally, efforts should be made to address
may be on infectious diseases, the chatbot's domain concerns related to data privacy, security, and ethical
coverage could be expanded to include a wider range of considerations in the development and implementation of
medical conditions, preventive measures, and treatment medical chatbots.
options. This would involve curating additional datasets,
training models on diverse healthcare topics, and REFERENCES
expanding the chatbot's knowledge base.
[1]. Lekha Athota, Vinod Kumar Shukla, Nitin Pandey,
 Integration with Telemedicine Platforms: Integrating the Ajay Rana, ‘‘Chatbot for Healthcare System Using
chatbot with telemedicine platforms and electronic Artificial Intelligence,’’ 2020 8th International
health record systems can streamline healthcare delivery Conference on Reliability, Infocom Technologies and
processes and facilitate seamless communication Optimization (Trends and Future Directions) (ICRITO)
between patients and healthcare providers. This would Amity University, Noida, India. June 4-5, 2020
require interoperability standards and robust integration [2]. Prakhar Srivastava, Nishant Singh,”Automatized
APIs. Medical Chatbot (Medibot)”Int Conf on Power
Electronics & IoT Applications in Renewable Energy
 Ethical and Regulatory Considerations: As the chatbot and its Control (PARC) GLA University, Mathura, UP,
becomes more widely used in healthcare settings, India. Feb 28-29, 2020
addressing ethical and regulatory considerations [3]. Tushar Tanmay, Akanksha Bhardwaj, Shilpi
becomes crucial. Ensuring compliance with data Sharma‘‘E-Health Bot to change the Face of
protection regulations, maintaining patient Medicare’’ 2020 Research, Innovation, Knowledge
confidentiality, and mitigating bias in AI algorithms are Management and Technology Application for Business
important areas for future development and research. Sustainability (INBUSH)
[4]. Rohit Binu Mathew, Sandra Varghese, Sera Elsa Joy,
The AI-based medical chatbot can continue to evolve Swanthana Susan Alex,‘‘Chatbot for Disease
as a valuable tool for disease prediction, personalized Prediction and Treatment Recommendation using
healthcare support, and healthcare accessibility. Machine Learning” Proceedings of the Third
International Conference on Trends in Electronics and
IV. SUMMARY Informatics (ICOEI 2019) IEEE Xplore
[5]. Tae-Ho Hwang, JuHui Lee, Se-Min Hyun, KangYoon
This research paper proposed the potential of Lee*,‘‘Utilizes biological sensors for measuring body
integrating artificial intelligence and machine learning in temperature, oxygen saturation, pulse, ECG, etc.”IEEE
healthcare. Studies highlight the effectiveness of chatbots in Internet Things J.,vol.1,no.1, pp. 22–
providing accurate medical information and personalized 32, Feb. 2014.
healthcare support. Leveraging NLP techniques and
machine learning algorithms, these chatbots offer promising
solutions to address healthcare challenges such as
accessibility and early disease detection. Further research is
needed to explore the scalability and reliability of these
chatbots in real-world healthcare settings.

IJISRT24MAR804 www.ijisrt.com 765