A Comprehensive Analysis of Disease prediction Using Machine Learning-Medi Consult

Abstract

This is a thorough and timely investigation of any health-related issue is essential for disease prevention and treatment. The normal way of diagnosis may not be sufficient in the event of a serious illness. Medi Consult is an Android app that predicts illnesses using machine learning algorithms based on symptoms supplied by patients or users. Using the K Nearest Neighbor (KNN) technique, the system calculates the disease\'s likelihood. Medi Consult\'s use of linear regression and decision tree algorithms to forecast illnesses such as Diabetes, Malaria, Jaundice, and Dengue assists in accurate and fast disease diagnosis. It provides people with a handy approach to monitor their health and gain early identification of potentially life-threatening disorders. The system\'s accuracy and quickness in forecasting illnesses aid in prompt medical intervention, resulting in better health outcomes. Furthermore, Medi Consult continually learns and improves its accuracy with each diagnosis, guaranteeing that it is up to speed with the most recent medical research and breakthroughs. Overall, Medi Consult is a helpful tool for both healthcare professionals and patients, enabling early illness identification and treatment and potentially saving lives.

Introduction

Healthcare plays a vital role in human life and economic stability, and the integration of Artificial Intelligence (AI) and Machine Learning (ML) has significantly enhanced medical diagnostics. A disease prediction system acts as a virtual diagnostic assistant that analyzes patient symptoms and predicts possible diseases without physical contact, which is especially valuable during outbreaks such as COVID-19 and Ebola.

The literature review highlights that various ML algorithms—such as Logistic Regression, Decision Trees, Random Forest, Support Vector Machines (SVM), Neural Networks, and Convolutional Neural Networks (CNNs)—have been successfully applied in disease prediction, including diabetes, heart disease, Parkinson’s disease, breast cancer, metabolic syndrome, COVID-19, and skin cancer. In many studies, Random Forest and deep learning models achieved superior accuracy.

The proposed system primarily uses the K-Nearest Neighbor (KNN) algorithm for disease prediction. The system architecture includes symptom data input, data collection, preprocessing, feature extraction, classification, and disease prediction. KNN classifies diseases based on similarity measures (e.g., Euclidean distance) by identifying the k-closest neighbors in a labeled training dataset and assigning the majority class. The value of “k” influences accuracy and noise reduction.

Performance evaluation is conducted using metrics such as accuracy, precision, recall, F1-score, and AUC-ROC. Comparative analysis shows that KNN outperforms other algorithms, achieving 90% accuracy in standard evaluation and up to 93.5% accuracy with a Weighted KNN model, making it the most effective approach for the given dataset.

The system ultimately predicts diseases based on user-input symptoms and provides information about causes and prevention. Overall, the study demonstrates that machine learning—particularly Weighted KNN—can serve as a reliable, accurate, and efficient tool for disease prediction and healthcare decision support.

Conclusion

The publication described a strategy for predicting illness based on a patient\'s symptoms, age, and gender. The Weighted KNN model produced .The above-mentioned parameters resulted in the maximum accuracy of 93.5% for disease prediction. Almost all of the ML models produced good accuracy results. Because some models were parameter dependant, they were unable to predict the illness and had a poor accuracy rate. We could simply manage the medical resources necessary for therapy once the sickness was predicted. This concept would assist to reduce the expense of treating with the sickness while also improving the healing process.

References

[1] \"A Comparative Study of Machine Learning Algorithms for Disease Prediction\" by S. S. Mishra (2021)J. Clerk Maxwell, A Treatise on Electricity and Magnetism, 3rd ed., vol. 2. Oxford: Clarendon, 1892, pp.68–73. [2] \"Machine learning-based prediction of heart disease using electronic health records\" by N. Rajkomar (2018) [3] \"An Ensemble Learning Approach for Diabetes Prediction Using Electronic Health Records\" by A. Alaa (2018) [4] \"Prediction of Parkinson\'s disease using machine learning algorithms\" by M. A. Siddiqui et al. (2020) [5] \"Machine Learning Techniques for Breast Cancer Diagnosis Using Histopathology Images: A Review\" by A. Z. A. Al-Zubaidi et al. (2021) [6] .Cho, D., Kim, D., & Choi, S. (2020). Predicting the risk of metabolic syndrome using machine learning and artificial intelligence. Journal of Personalized Medicine, 10(3), 97 [7] Ghosh, S., Mitra, S., & Saha, S. (2021). Machine learning and deep learning-based COVID-19 diagnosis using chest X-ray and CT scan images: A review. Journal of X-Ray Science and Technology, 29(3), 483-495. [8] Nwankwo, E. O., Adebesin, O. A., Adetunji, A. E., Ogunmefun, O. T., & Ojo, S. A. (2021). Machine learning models for malaria prediction: A review. Heliyon, 7(7), e07459. [9] Zame, W. R., & Javaheri, A. (2020). Machine learning in healthcare: Overview, trends, and future directions. Journal of Medical Systems, 44(8), 1-18. [10] Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun, S. (2017). Dermatologist-level classification of skin cancer with deep neural networks. Nature, 542(7639), 115-118. [11] Galloway, M., & Hamilton, J. (2019). Machine learning in medical imaging. Biomedical Physics & Engineering Express, 5(1), 1-20. [12] Jang, S. M., Song, I., Cho, J. W., Kim, B. H., & Kim, H. (2020). Diagnosis of Alzheimer\'s disease using machine learning algorithms on MRI images: A systematic review. Healthcare, 8(1), 37 [13] Chen, C., Huang, J., Cheng, Z., Lu, X., & Yang, J. (2021). Applications of machine learning in the prediction of mental health disorders: A review. Journal of Medical Systems, 45(6), 1-11. [14] Gomathi, N., & Sumathi, P. (2021). An overview of machine learning algorithms for predicting disease outbreak. Health and Technology, 11(3), 757-768. [15] Huang, G., Liu, Z., Van Der Maaten, L., & Weinberger, K. Q. (2017). Densely connected convolutional networks. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 4700-4708). [16] Jiang, F., Jiang, Y., Zhi, H., Dong, Y., Li, H., Ma, S., ... & Wang, Y. (2017). Artificial intelligence in healthcare: Past, present and future. Seminars in Cancer Biology, 52, 228-243. [17] Liao, K., Xie, Y., Liu, Y., Chen, Y., & Huang, Y. (2020). Deep learning-based disease diagnosis with genetic data. IEEE Transactions on Biomedical Engineering, 67(1), 141-153. [18] Zhou, X., Zeng, X., Yang, J., Zhang, X., Wang, J., Liu, Z., ... & Gao, X. (2019). Deep learning: A review from a structure perspective. Neurocomputing, 335, 103-111. [19] Zhou, Y., Liu, Y., Liu, L., & Wu, J. (2021). A systematic review of machine learning-based diagnosis models for COVID-19. Frontiers in Public Health, 9, 1-10. [20] Chen, Z., Chu, C., Zheng, W., Zhang, Y., Han, Y., & Cai, Y. (2020). Early warning system for sepsis using machine learning: A systematic review. Journal of Healthcare Engineering, 2020.

Copyright

Copyright © 2026 Dr. Devendra, Mr. Munish Kumar, Shorya Yadav. 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.