• Version
• Download 20
• File Size 335.01 KB
• File Count 1
• Create Date 26 May 2025
• Last Updated 26 May 2025

CLINICAL DIAGONSIS DECISION MAKING AND DRUG USAGE REGULATIONS

Authors:

Dr. A. Suresh Rao ¹, Waghmare Manisha ², Suroju Gowtham ³, Sai Reddy Srivalli ⁴, Shanigarapu Srinivas ⁵.

1 Head & Professor, Department of Computer and Science Engineering, TKR College of Engineering and Technology.

2,3,4,5UG Scholars, Department of Computer and Science Engineering, TKR College of Engineering and Technology, Medbowli, Meerpet.

ABSTRACT: 21st century, the emergence of new diseases with overlapping and additional symptoms has made disease prediction more complex. Early diagnosis plays a crucial role in reducing fatalities by enabling timely treatment. The existing systems rely on traditional machine learning models like Random Forest and Naïve Bayes for disease prediction but lack the efficiency and accuracy needed for precise diagnosis. This project proposes a Boosting-based disease prediction model using XGBoost and AdaBoost, along with comparative analysis against Light GBM, Extra Trees, K-Nearest Neighbors (KNN), Naïve Bayes, and Random Forest. The dataset consists of 1,200 datapoints, covering 24 diseases, with symptom descriptions in natural language. By leveraging Boosting techniques, the system enhances prediction accuracy through an ensemble approach. The system is implemented using Flask, providing an interactive web-based interface where users can input symptoms in textual format. The model then predicts the most probable disease and provides drug recommendations, precautionary measures, and dietary suggestions. The proposed approach demonstrates improved performance over traditional models, ensuring a more reliable disease diagnosis and prevention system.

KEYWORDS: Disease Prediction, Machine Learning, Boosting Algorithms, XGBoost, AdaBoost, Flask, Natural Language Processing, Light GBM , Extra Trees, Random Forest, Naïve Bayes, KNN, Symptom-Based Diagnosis