Medical diagnosis prediction using Machine Learning
| dc.contributor.author | P Thejaswini 1NH22CE033 Miruthula S 1NH22CE029 CH L Sriram 1NH22CE008 | |
| dc.date.accessioned | 2026-02-06T05:45:20Z | |
| dc.date.available | 2026-02-06T05:45:20Z | |
| dc.date.issued | 2026 | |
| dc.description.abstract | Clinical knowledge, manual symptom interpretation, and laborious laboratory tests have historically been the mainstays of medical diagnosis. Recent developments in machine learning (ML) have made it possible to create intelligent diagnostic systems that can quickly and accurately analyze vast amounts of medical data. The goal of this work is to develop and assess a machine learning-based diagnostic framework that helps medical practitioners identify diseases early and accurately. In addition to investigating feature-engineering techniques that enhance predictive performance, the main goal is to use supervised learning techniques to classify patient conditions using structured clinical datasets. Starting with data collection and preprocessing, the suggested system integrates multiple stages. The most important factors influencing diagnostic accuracy are found using feature selection algorithms like mutual-information analysis and Recursive Feature Elimination (RFE). Several machine learning algorithms, such as Logistic Regression, Random Forests, Support Vector Machines, and Gradient Boosting, are used in the construction of the diagnostic model. The best classifier for the chosen disease category can be found through comparative analysis. To guarantee generalizability and avoid overfitting, stratified data splitting and cross-validation are used during model training. The model gives each patient a risk score in place of binary classification, allowing clinical workflows to prioritize patients. By anonymizing patient data and implementing secure data-handling procedures, emphasis has been placed on protecting data privacy and adhering to healthcare standards. Overall, this study shows how machine learning can improve medical diagnosis by increasing accuracy, decreasing manual labour, and facilitating early disease detection. A complete ML-based diagnostic pipeline from preprocessing to deployment, comparative assessment of various algorithms, interpretability mechanisms for clinician trust, and risk-based diagnostic outputs that support medical decision-making are among the work's notable contributions. The results demonstrate how incorporating machine learning into healthcare environments can greatly improve patient outcomes and strengthen diagnostic support systems. | |
| dc.identifier.uri | http://192.168.75.5:4000/handle/123456789/20928 | |
| dc.language.iso | en | |
| dc.publisher | New Horizon College of Engineering | |
| dc.title | Medical diagnosis prediction using Machine Learning | |
| dc.type | Learning Object |