VISVESVARAYA TECHNOLOGICAL UNIVERSITY
BELGAVI, KARNATAKA -590 018
Project Synopsis Submission
on
“Smart Multiple Disease Prediction System using Machine
Learning and Streamlit”
By
Team Code: CSPT2025-608
Swathi S 4MH22CS163
Aishwarya H N 4MH22CS008
Varshitha M T 4MH22CS176
Vidyashree N 4MH22CS180
DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING
MAHARAJA INSTITUTE OF TECHNOLOGY MYSORE
Belawadi, S.R. Patna Taluk, Mandya Dist-571477.
Accredited By:
2024-25
1. Aim:
The aim of this project is to develop a machine learning-based system that predicts the
likelihood of multiple diseases. Users can select a specific disease, input patient data, and
receive real-time predictions using trained ML models.
2. Introduction:
Healthcare is rapidly evolving with technological advancements, particularly through the
integration of machine learning models for disease prediction. Early diagnosis plays a crucial
role in preventing the progression of chronic illnesses, yet healthcare professionals often
struggle with using different applications for different diseases, leading to a fragmented and
time-consuming diagnostic process.
This project aims to overcome these limitations by developing a Smart Multi-Disease
Prediction System that uses machine learning to forecast multiple diseases — such as
Diabetes, Heart Disease, Chronic Kidney Disease, and Cancer — from a single [Link]
achieve this, the system implements various classification algorithms, including K-Nearest
Neighbor, Support Vector Machine, Decision Tree, Random Forest, Logistic Regression, and
Gaussian Naive Bayes. These models are trained on disease-specific datasets, with their
accuracy validated and compared to identify the best-performing model for each condition.
3. Motivation:
The motivation behind this project stems from the increasing prevalence of chronic
diseases and the growing need for early detection and timely intervention. In many
cases,healthcare professionals and patients are required to use separate applications for
different disease predictions-an approach that is not only inefficient but also fragmented.
This project aims to address these challenges by developing a Multi-Disease Prediction
System — a unified platform that integrates multiple machine learning models, each
trained to predict a specific disease. The motivation driving this project includes:
• Empowering users — allowing both patients and medical practitioners to
independently assess health risks through a single, user-friendly web application.
• Streamlining diagnostics — offering a centralized system where users can
quickly select a disease, input health data, and receive predictions, reducing the
hassle of switching between multiple applications.
• Enhancing healthcare efficiency — providing preliminary insights that support
medical professionals in making informed decisions even before clinical
consultation.
• Bridging technology and healthcare — leveraging machine learning algorithms
to improve diagnostic accuracy and promote data-driven healthcare solutions.
• Scalability and flexibility — creating a system that can be easily expanded to
include additional disease models in the future.
By integrating these machine learning models into one platform, this project strives to
simplify the diagnostic workflow, encourage early detection, and enhance healthcare
accessibility for all.
4. Problem Statement:
Early diagnosis is crucial for improving patient outcomes,but existing systems focus on
individual diseases,requiring separate applications or models for each [Link] makes
it difficult for healthcare professionals or patients to efficiently assess the risk of multiple
diseases.A unified system that integrates separate disease prediction models could provide
more efficient and accurate predictions for various diseases using a single platform.
5. Proposed Solution:
The Smart Multi-Disease Prediction System will integrate several machine learning
models, each trained to predict a specific disease based on patient input [Link] can select
a disease they want to check—such as diabetes, Parkinson's disease, or heart disease—enter
relevant details like age, gender, symptoms, and test results, and receive instant predictions
about the likelihood of having that condition.
The platform uses machine learning models trained on disease-specific datasets—for
example, Support Vector Machine (SVM) for predicting diabetes and Parkinson's disease,
and Logistic Regression for heart disease. This system simplifies disease detection by
offering a single platform, eliminating the need for multiple applications.
6. Expected Output:
1) Accurate Disease Prediction for Multiple Diseases:
The system is expected to accurately predict multiple diseases—Diabetes, Heart
Disease, Chronic Kidney Disease, and Cancer—using a single user interface. The
accuracy of predictions is validated using various machine learning algorithms, with the
best-performing models selected for each disease. The expected accuracy levels are as
follows:
• Diabetes: Highest accuracy of 98.75% using the Random
Forestalgorithm.
• Heart Disease: Highest accuracy of 86.67% using the Support Vector
Machine (SVM) algorithm.
• Chronic Kidney Disease: Highest accuracy of 99.16% using the
Random Forest algorithm.
• Cancer: Highest accuracy of 85.9% using the Random Forest algorithm.
These high accuracy levels ensure reliable predections,enabling early detection and
timely medical intervention.
2) Early Detection and Prevention of Life-Threatening Diseases:
The system is expected to facilitate the early detection of life-threatening diseases,
which is crucial for preventing their progression. Early intervention can significantly
improve patient outcomes and reduce the overall healthcare burden.
3) User-Friendly Web Application for Easy Access:
The system will be implemented as a web application using Streamlit, ensuring
that it is accessible and easy to use for both healthcare professionals and patients.
The expected features of the application include:
• A sidebar for navigating between different disease prediction modules.
• Input forms where users can enter relevant attributes (e.g., age, BMI,
blood pressure, symptoms) for each disease.
• A prediction button to trigger the disease prediction process and display
results.
This initiative interface ensures that users can interact with the system without
requiring technical expertise,making it suitable for a wide range of users.
4) Model Accuracy:
The predictions made by the models will have an accuracy rate above a specified
threshold (e.g., 85%), validated using real-world datasets. This ensures that the
system provides accurate and reliable disease predictions, supporting early diagnosis
and better healthcare outcomes.
7. References:
Paper-1: “Multi-Disease Prediction System Using Machine Learning” - Jairus Mathews ,
John Joseph,Ronald Reji,Abhijeet Kamthe,Prof. Rupali Deshmukh- 2023
[Link]
Disease_Prediction_System_Using_Machine_Learning
Paper-2: : “Design and Implement Multiple Disease Prediction System” – Chandradeep
Bhatt; Astha Mehra; Rahul Chauhan; Aman Sharma; Arsh Dogra; Akansha Tyagi 2023
World Conference on Communication & Computing (WCONF),
[Link]
sease_Prediction_System
Paper-3: “Machine Learning Based Heart Disease Prediction System” - [Link] Raja ,
[Link],[Link] Reddy, NageswaraRao Sirisala-2021
[Link]
sease_Prediction_System
Paper-4: “Diabetes prediction using supervised machine learning” - Muhammad Exell
Febrian, Fransiskus Xaverius Ferdinan, Rezki Yunanda-2023
[Link]