Multi-Disease Prediction Using Machine Learning: A Web-Based Diagnostic Support System for Diabetes, Heart Disease, and Parkinson\'s Disease

Abstract

Chronic non-communicable diseases such as diabetes mellitus, cardiovascular disease, and Parkinson’s disease are significant causes of morbidity and mortality. Screening is limited by the accessibility to medical experts and the cost of appropriate diagnostic tests. In this paper, we present a diagnostic support system based on a unified web platform that classifies patients according to the risks of developing three diseases based on regularly collected clinical or audio data using classical supervised learning algorithms. For the diabetes mellitus classification task (Pima Indians dataset, 8 features) and Parkinson’s disease detection (Oxford voice recording dataset, 22 features), Support Vector Machine classifiers with linear kernels were trained. A Logistic Regression classifier was used for predicting occurrences of heart disease (UCI Cleveland dataset, 13 features). The training results in the form of the classifiers were serialised with Pickle and implemented as a web application with the Flask framework and MySQL database software that differiates between the administrators’ and patients’ interfaces. With the stratified 80/20 holdout validation method, the system demonstrated accuracy of 77.3%, 85.2%, and 87.2% for the diabetes mellitus, heart disease, and Parkinson’s disease detection tasks, respectively. The paper describes mathematical algorithms used for machine learning in detail and presents computational specifics of the Support Vector Machine implementation for the Pima Indians dataset as an example. Additionally, the accuracy, precision, recall, and F1-score metrics were calculated based on the confusion matrices for each of the tasks.

Introduction

Artificial Intelligence (AI) and Machine Learning (ML) have significantly transformed the healthcare sector by enabling accurate disease diagnosis, early prediction, and improved clinical decision-making. The increasing availability of electronic health records, medical images, laboratory reports, and data from health sensors provides valuable information that can be analyzed using machine learning algorithms. These predictive models identify hidden patterns within patient data, allowing healthcare professionals to detect diseases at an early stage and recommend timely interventions. Data preprocessing, including cleaning, normalization, and feature selection, plays a crucial role in improving the accuracy and reliability of these prediction models.

Machine learning has been widely applied to predict chronic diseases such as diabetes, heart disease, and Parkinson’s disease. Early detection of diabetes helps prevent severe complications, while heart disease prediction assists clinicians in identifying cardiovascular risk factors and initiating appropriate treatment. Similarly, predictive models can identify Parkinson’s disease at an early stage, improving patients' quality of life through timely medical intervention. Compared with traditional diagnostic methods, which often require expensive laboratory tests and expert interpretation, machine learning provides a faster, more affordable, and accessible alternative. Supervised learning algorithms such as Support Vector Machine (SVM), Logistic Regression (LR), Decision Tree, and Random Forest are commonly employed because of their high prediction accuracy. Furthermore, integrating these models into web-based applications improves accessibility by allowing users and healthcare professionals to obtain predictions remotely through internet browsers.

The literature review highlights several recent developments in multi-disease prediction systems. Researchers have proposed hybrid machine learning frameworks, ensemble learning models, multi-label classification techniques, deep learning approaches, telemedicine-based systems, and intelligent healthcare platforms to improve disease prediction accuracy. These studies demonstrate that combining multiple algorithms, optimized feature extraction, and advanced AI techniques enhances diagnostic performance, reduces classification errors, and supports healthcare professionals in making informed clinical decisions.

The proposed study presents a Multi-Disease Health Prediction Platform capable of predicting diabetes, heart disease, and Parkinson’s disease using machine learning algorithms. The system utilizes publicly available datasets obtained from the UCI Machine Learning Repository and Kaggle. The diabetes dataset includes attributes such as pregnancies, glucose level, blood pressure, insulin, BMI, diabetes pedigree function, and age. The heart disease dataset contains clinical variables including age, chest pain type, cholesterol level, blood pressure, ECG results, heart rate, and exercise-induced angina. The Parkinson’s disease dataset consists of biomedical voice measurements such as jitter, shimmer, harmonic-to-noise ratio, and other speech-related features. These datasets are preprocessed to remove missing values, eliminate redundant information, normalize data, and prepare them for machine learning.

Feature selection is performed to identify the most relevant medical attributes that influence disease prediction while reducing computational complexity. Important features include age, gender, blood pressure, glucose level, BMI, cholesterol, heart rate, insulin level, skin thickness, and disease-specific clinical indicators. These selected features serve as independent variables for training the prediction models.

The experimental implementation was carried out using Python and the Flask web framework. The datasets were divided into training and testing sets using an 80:20 ratio. Two supervised machine learning algorithms—Support Vector Machine (SVM) and Logistic Regression (LR)—were trained and evaluated. Model performance was assessed using standard evaluation metrics including accuracy, precision, recall, F1-score, and confusion matrix. The trained models were integrated into a Flask-based web application that enables users to input their health information and receive disease prediction results through a user-friendly interface.

Experimental results demonstrate that the developed platform effectively predicts diabetes, heart disease, and Parkinson’s disease. Both SVM and Logistic Regression produced high prediction accuracy, although Support Vector Machine consistently outperformed Logistic Regression across most evaluation metrics. The web-based system provided fast prediction results, making it suitable for practical clinical decision support and early disease screening.

The discussion confirms that careful data preprocessing, feature selection, and normalization significantly improve prediction performance. The developed platform can assist healthcare professionals by providing rapid, reliable predictions based on patient health indicators, thereby supporting early diagnosis and informed clinical decision-making.

For future work, the authors propose integrating advanced deep learning techniques and Explainable Artificial Intelligence (XAI) to improve prediction accuracy while providing transparent explanations for model decisions, increasing clinicians’ trust in AI-based recommendations. Additionally, they recommend connecting the system with wearable devices and IoT-enabled health sensors to enable continuous real-time health monitoring, allowing early detection of disease progression and personalized healthcare management. Overall, the proposed multi-disease prediction platform demonstrates the potential of machine learning to improve healthcare delivery through accurate, accessible, and cost-effective disease prediction.

Conclusion

We present a compact but complete multi-disease prediction prototype based on scikit-learn classifiers and a Flask/MySQL web application. Using standard benchmarks, our system achieves 77.3% accuracy for diabetes, 85.3% for heart disease, and 87.2% for Parkinson’s disease, all delivered through a role-differentiated interface for patients and administrators. The mathematical details and worked numerical examples in this paper render the algorithms’ operations fully transparent. We believe that our prototype demonstrates the ability of classical ML methods and a basic technical stack to serve as a foundation for serious screening applications, with room for incorporating additional diseases and expanding the evidence base beyond benchmark data. We have demonstrated the process of designing and implementing a web-based system for multi-disease prediction using Python-based machine learning algorithms. Our prototype application, which includes a user-friendly interface for patients and administrators, has been evaluated using standard benchmark datasets for diabetes, heart disease, and Parkinson’s disease. The results of our experiments, which were implemented using Support Vector Machines, Logistic Regression, and other classical ML methods, suggest that acceptable prediction performance (in terms of accuracy, precision, recall, and F1 score) can be achieved using a relatively small amount of code (less than 500 lines of Python) and standard software tools. We have also produced detailed numerical examples to illustrate how the algorithms work, which can help improve transparency and facilitate replication. Finally, we have discussed possible approaches to build upon our prototype in the future, including the use of federated learning and deep learning architectures.

References

[1] Gupta and A. Singh, “An Optimal Multi-Disease Prediction Framework Using Hybrid Machine Learning Techniques,” Kuwait Journal of Science, vol. 49, no. 2, 2022. Available: https://doi.org/10.48129/kjs.splml.19321 [2] A. S. Prakaash, K. Sivakumar, B. Surendiran, S. Jagatheswari, and K. Kalaiarasi, “Design and Development of Modified Ensemble Learning with Weighted RBM Features for Enhanced Multi-Disease Prediction Model,” Biomedical Signal Processing and Control, vol. 77, 2022. Available: https://doi.org/10.1007/s00354-022-00190-2 [3] B. Li, Y. Zhang, and X. Wu, “DLKN-MLC: A Disease Prediction Model via Multi-Label Learning,” Applied Intelligence, vol. 53, no. 4, pp. 4200–4218, 2023. Available: https://doi.org/10.3390/ijerph19159771 [4] S. Uddin, S. Wang, H. Lu, A. Khan, F. Hajati, and M. Khushi, “Comorbidity and Multimorbidity Prediction of Major Chronic Diseases Using Machine Learning and Network Analytics,” Expert Systems with Applications, vol. 218, 2023. Available: 10.1016/j.eswa.2022.117761 [5] J. Al Nahian, A. K. M. Masum, S. Abujar, and M. J. Mia, “Common Human Diseases Prediction Using Machine Learning Based on Survey Data,” arXiv preprint, 2022. Available: https://doi.org/10.48550/arXiv.2209.10750 [6] M. Asrarulhaq Khadir, A. Mohd, M. Ali, and P. A. Khan, “Multiple Disease Prediction System Using Machine Learning,” International Journal of Creative Research Thoughts, vol. 11, no. 5, 2023. Available: IJCRT Journal [7] D. R. Unnithan and J. R. Jeba, “A Novel Framework for Multiple Disease Prediction in Telemedicine Systems Using Deep Learning,” Biomedical Signal Processing and Control, vol. 91, 2024. Available: 10.1080/00051144.2024.2301889 [8] A. Pradhan, J. S. V. G. Krishna, B. Prasanna Kumar, G. Tabita, V. V. R. Sastry, and K. Sudhakar, “Integrated Disease Forecasting: Leveraging Deep Learning and Machine Learning for Multi-Disease Prediction,” Journal of Computational Intelligence and Neuroscience, 2024. Available: 10.1186/s12935-024-03295-4 [9] A. Yadav, A. Pandey, G. Srivastava, and H. Dwivedi, “Multi-Disease Prediction Using Machine Learning Algorithm,” International Journal of Advanced Research in Computer Science, vol. 15, no. 2, 2024. Available: IJARCS Journal [10] K. Mali, P. Rasane, M. Patil, G. Suryawanshi, and A. Chaphadkar, “Multiple Disease Prediction Using Machine Learning,” SSRN Electronic Journal, 2024. Available: http://dx.doi.org/10.2139/ssrn.4747365 [11] N. Reddy M and S. Satheeskumaran, “An Efficient Multi-Disease Prediction Model Using Advanced Optimization Aided Weighted CNN with Dilated GRU,” Intelligent Decision Technologies, vol. 18, no. 4, 2024. Available: http://dx.doi.org/10.2139/ssrn.4747365 [12] Y. Lin, M. Li, Z. Zhu, Y. Feng, L. Xiao, and Z. Chen, “Research on Disease Prediction Model Construction Based on Computer AI Deep Learning Technology,” arXiv preprint, 2024. Available: 10.1109/ICIPCA61593.2024.10709274 [13] J. Singh, N. K. Chahar, Lalman, and V. Kumar, “Diagnosify: Multidisease Forecast – An Integrated Machine Learning Approach for Disease Prediction and Early Intervention in Healthcare,” SSRN Electronic Journal, 2025. Available: http://dx.doi.org/10.2139/ssrn.5239503 [14] H. Belhad, A. Bourbia, and S. Boughanja, “Chronic Diseases Prediction Using Machine Learning and Deep Learning Methods,” arXiv preprint, 2025. Available: https://doi.org/10.48550/arXiv.2505.00189 [15] M. K. Bechir and F. Atasoy, “Enhancing Multi-Disease Prediction with Machine Learning: A Comparative Analysis and Hyperparameter Optimization Approach,” Gazi University Journal of Science Part C: Design and Technology, vol. 13, no. 1, 2025. Available:10.30855/gmbd.2021.03.09.

Copyright

Copyright © 2026 Vedamurthy D R, Dr. Anup Ritti , Ayesha Bibi , N Nandini . 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.