Symptoms Checker For Disease Detection Using Agentic AI

Read full text Summarise with VIRAT
Authors: S Tejaswi, R Manjeera, Y Karthik, G. Chandra Sekhar, T Mahitha Bhargavi
Cited by: 0
References: 8

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

The system I am discussing in this paper is designed to enable more manageable easy accurate diagnosis to people, particularly in areas where physicians are scarce.We refer to it as an agentic-based Symptoms Checker, i.e., it has a number of small agents, which perform various tasks in a parallel manner.On our test data, we vote with six distinct models, Decision Tree, Random Forest, SVM, Logistic Regression, Naive Bayes, and K-NN, and the final guess of what the disease is, gives the models a good 99.74% overall accuracy.The tool does not just do this: it also transfers the prediction to the agents that propose the appropriate specialist and search the nearby hospitals using a live search API.Sentence embeddings provide a conversational agent with cloud-based access to receive personal advice by the user and interact with the agent in a chat-like manner.The app was developed using Flask but included session-based authentication and SQLite database allowing us to store prediction history, quiz score and chat records securely.We find this combination of all-in-one agentic pipeline more trustworthy in our real world tests than just using a single model and enhances access to healthcare in general.Hopefully, this project will be used as a template when it comes to other projects of this kind in digital health.

Fields of Study & Subjects

Artificial Intelligence in Healthcare Artificial Intelligence in Healthcare and Education Innovation in Digital Healthcare Systems

Publication details

Journal / source
International journal of engineering science and advanced technology.
Volume
26
Issue
3
Pages
525–530
Publication date
1 March 2026
Language
EN