Multi Disease Detection System using Machine Learning

Abstract

In this modern era, early detection of critical diseases remains a significant challenge in healthcare. Many existing models either focus on detecting multiple common diseases orpredict a single disease at a time using algorithms such as Support Vector Machine, K Nearest Neighbors, Decision Tree, Navie Bayes and Logistic Regression. This limitation can result in delayed diagnose and treatment, negatively affecting patient outcomes and survival rates. This research focuses on developing aMulti DiseaseDetectionSystemusing advanced Machine Learning algorithms such as Random Forest and Convolutional Neural Networks (CNNs) to predict six high-mortality diseases: heart, kidney, liver, breast cancer, pneumonia, and brain tumor diseases. These diseases rank among the top global causes of death according to WHO statistics and require timely diagnosis for effective medical intervention. This system aims to reduce disease-related mortality, minimize treatment delays, and lower healthcare costs. Additionally, the platform offers personalized diet, food, exercise, and doctorrecommendations, providing a comprehensive approach to health management. Overall, this Multi-Disease Detection System leverages advanced machine learning techniques to improve early diagnosis, enhance treatment outcomes, and empower users with personalized health recommendations, setting the foundation for more efficient and accessible healthcare solutions.

Introduction

This paper presents a Multi-Disease Detection System that uses Machine Learning (ML) and Deep Learning (DL) to detect six major diseases—heart disease, kidney disease, liver disease, breast cancer, pneumonia, and brain tumor—from both clinical data and medical images. The system also generates personalized health reports, diet recommendations, exercise suggestions, and doctor recommendations to support preventive healthcare and early diagnosis.

Background

Cardiovascular diseases, cancer, and respiratory illnesses remain leading causes of death worldwide. Traditional diagnostic systems generally focus on detecting a single disease and rely on conventional machine learning algorithms such as SVM, KNN, Decision Trees, Naïve Bayes, and Logistic Regression. These approaches are limited in handling multiple diseases simultaneously, highlighting the need for a comprehensive, AI-powered diagnostic system.

Literature Review

Previous studies have shown that:

• Random Forest performs exceptionally well for heart, kidney, liver, and breast cancer prediction.
• Convolutional Neural Networks (CNNs) achieve high accuracy in image-based disease detection such as pneumonia and brain tumors.
• Ensemble learning, transfer learning, and optimization algorithms further improve prediction accuracy.
• Despite significant progress, challenges remain in model generalization, explainability, and clinical deployment.

Proposed System

The proposed Multi-Disease Detection System consists of several integrated modules:

• Web Interface for user input and result display.
• Disease Detection Module using Random Forest for clinical data and CNNs for medical images.
• Report Generation Module that automatically creates downloadable diagnostic reports.
• Recommendation Engine providing personalized diet, exercise, food, and doctor recommendations.
• BMI and Calorie Calculator for customized meal planning and preventive healthcare.

This architecture supports both disease diagnosis and long-term health management through personalized recommendations.

Datasets

The system is trained using publicly available Kaggle datasets:

• Heart Disease (303 records)
• Liver Disease (583 records)
• Kidney Disease (400 records)
• Breast Cancer (569 records)
• Pneumonia (5,189 chest X-rays)
• Brain Tumor (3,679 MRI images)
• Diet Recommendation dataset containing over 522,000 recipes for personalized nutrition planning.

Machine Learning Methodology

The datasets undergo preprocessing, including:

• Missing value handling
• Feature encoding
• Data normalization

The system uses:

• Random Forest for structured clinical datasets.
• Convolutional Neural Networks (CNNs) for medical image classification.

Model performance is evaluated using accuracy, precision, recall, and F1-score to ensure reliable disease prediction.

Workflow

The system operates as follows:

1. Users enter clinical information or upload medical images.
2. ML/DL models predict the presence of six diseases.
3. A diagnostic report is automatically generated.
4. Personalized recommendations for diet, exercise, and medical consultation are provided.
5. BMI and calorie calculations generate customized meal plans.

Results

The proposed system achieved high prediction accuracy:

The high accuracy for heart disease, kidney disease, pneumonia, breast cancer, and brain tumor demonstrates the effectiveness of Random Forest and CNN models. Liver disease achieved comparatively lower accuracy due to the complexity and variability of liver function data.

Conclusion

The Multi-Disease Detection System using Machine Learning represents a transformative advancement in healthcare, enabling early and accurate diagnosis of six high-mortality diseases: heart disease, liver disease, kidney disease, pneumonia, breastcancer,andbraintumors.ByintegratingadvancedtechniquessuchasRandom Forest and ConvolutionalNeural Networks(CNNs), the systemachieves remarkable predictiveaccuracy,withmostdiseasessurpassing95%.However,liverdiseaseposes challengesduet otheinherentcomplexityofliverfunctiontestresults,highlightingan area for further improvement. Thesystem’sabilitytounifyclinicalparametersandmedicalimagingwithin a single framework, coupled with personalized food, exercise, healthcare anddiet recommendations, underscores its potential to revolutionize patient-centric care. Furthermore, the automated generation of detailed reports and actionable recommendationsfacilitatesinformeddecision-makingforpatientsandhealthcare providers alike. Future work will focus on real-time data integration, leveraging wearable devicesandIoTsystems for continuoushealth monitoring. Incorporatingexplainable AI frameworks will enhance transparency in model predictions, fostering trust and adoptionamongusers.Mobileapplicationdevelopmentandmultilingualsupportwill further enhance accessibility, particularly in remote and underserved regions. Advanced functionalities, including telemedicine integration and personalizedtreatmentrecommendations, havethepotentialtoelevatethesysteminto a comprehensive healthcare solution. Leveraging blockchain technology for secure patient data management and integrating genetic data for precision diagnostics are promising directions for future development. Additionally, combining models into a unifiedlarge languagemodel(LLM)framework could enhancesystem efficiencyand predictive performance. Inconclusion,the Multi-DiseaseDetectionSystem marksa significantstep towardrevolutionizinghealthcaredelivery .Byfosteringearly intervention, reducing treatment delays, and empowering patients through actionable insights, the system paves the way for more accessible, equitable, and effective healthcare solutions worldwide.

References

[1] Athira K & MohanM. (2024). Multi disease detection and diagnosis through ensemble learning. International Conference on Advancements in Power, Communication and Intelligent Systems (APCI),IEEE, 1–6. https://doi.org/10.1109/apci61480.2024.10616480 [2] Laxmi Deepthi G,Srinivasa Karthik L k,sai Rohan p, Sneha K,Niharika P& Hari Priya K .(2023). Multiple Disease Prediction System using MachineLearning and Streamlit. 5th International Conference on Smart Systems and Inventive Technology (ICSSIT),IEEE, 923–931. https://doi.org/10.1109/icssit55814.2023.10060903 [3] Harshit J, Sarthak A, Rishabh K, Rachna J, & Preeti N.(2021)Heart disease predictionusingmachinelearningalgorithms.IOPConferenceSeriesMaterials Science and Engineering, 1022(1), 012072. [4] https://doi.org/10.1088/1757-899x/1022/1/012072 [5] AhmadAyidA&HuseyinP.(2023).Predictionofheartdiseasebasedonmachine learning using Jellyfish optimization algorithm. Diagnostics,MDPI, 13(14), 2392. https://doi.org/10.3390/diagnostics13142392 [6] Srilatha T, Hajarathaiah K, Sai Ram Praneeth G, Srinivasarao B, Lakshmikath N, Nagamanohar P, Satish A, Murali Krishna E. (2023). Liver Disease Prediction and Classification using Machine LearningTechniques. (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 14(No. 2). [7] RahmanA K M S, ShamratFMJM, TasnimZ, RoyJ, Akhter Hossain S. (2019). A comparative study on liver disease prediction using supervised machine learning algorithms. INTERNATIONAL JOURNAL OFSCIENTIFIC&TECHNOLOGYRESEARCH,8(11),419–420. [8] https://www.ijstr.org [9] Adeba D A, & Melak Sitote T. (2022). Chronic kidney disease prediction using machine learning techniques. Journal of Big Data, 2–19. https://doi.org/10.1186/s40537-022-00657-5 [10] Pankaj C, Sandeep C, Prasun C, Gaurav K,Tulika C, Zbigniew L, Michal J,Lukasz J, Radomir G, Elzbieta J, & Vadim B.(2021). Prediction of Chronic kidney Disease - a Machine learning perspective. IEEE Access, 9, 17312– 17334. https://doi.org/10.1109/access.2021.3053763 [11] Lamia A, & Fawaz A. (2022). Detection of pneumonia infection by using deep learning on a mobile platform. Computational Intelligence andNeuroscience, 2022, 1–9. https://doi.org/10.1155/2022/7925668 [12] Sharma S, & Guleria K. (2023).ADeep Learning based model for the Detection of Pneumonia from Chest X-Ray Images using VGG-16 and Neural Networks. Procedia Computer Science, 218, 357–366. https://doi.org/10.1016/j.procs.2023.01.018 [13] H. Rajaguru & S. R. Sannasi Chakravarthy, \"Classification of Wisconsin Breast Cancer Data with Extreme Learning Machine and Osprey Optimization Algorithm,\"2023 Third International Conference on Smart Technologies, Communication and Robotics (STCR), pp. 1-4, https://doi.org/10.1109/STCR59085.2023.10397042 [14] Akmalbek Bobmirzaevich A, Mukhriddin M, & Taeg Keun W. (2023). Brain tumor detection based on deep learning approaches and magnetic resonance imaging. Cancers (Vol. 15, p. 4172). https://doi.org/10.3390/cancers15164172 [15] Soheila s, Sorayya R, Hamidreza k, Sharareh R & Niakan K. (2023). MRI-based brain tumordetectionusingconvolutional deeplearningmethodsand chosen machine learning techniques. BMC Medical Informatics and Decision Making, 23. https://doi.org/10.1186/s12911-023-02114-6

Copyright

Copyright © 2026 Hemanth Kumar Rao Mahendrakar, Jahnavi Dasaraju, Divya Sai Roopanagudi, Dinesh Naik Sugali, Dr. Ranjit Reddy Midde. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.