Lung Disease Detection and Classification using Spirometric Dataset

Project Code :TCMAPY1621

Objective

The primary objective of this project is to design and implement an efficient, accurate, and reliable machine learning-based system for the detection and classification of lung diseases using spirometric datasets. Specifically, the system aims to predict six distinct lung condition categories — Normal, Obstruction, Unclassified, Restriction, PRISm, and Mixed — by analyzing patient spirometric data. To achieve high classification performance, the project employs a comparative approach, training multiple machine learning models including Decision Tree, Random Forest, XGBoost, SVM, CatBoost, Gradient Boosting, and Logistic Regression.

Abstract

The project, titled "Lung Disease Detection and Classification Using Spirometric Data Set," aims to develop a robust machine learning system for predicting lung disease classes based on spirometric data. The system employs multiple algorithms, including Decision Tree, Random Forest, XGBoost, Support Vector Machine (SVM), CatBoost, Gradient Boosting, and Logistic Regression, to ensure high predictive accuracy. To optimize feature selection, the K-best method is used for dimension reduction, identifying the top 20-25 most relevant features from the dataset. This enhances model efficiency and reduces computational complexity. The system is designed to classify lung conditions into six distinct categories: Normal, Obstruction, Unclassified, Restriction, PRISm (Preserved Ratio Impaired Spirometry), and Mixed. By inputting spirometric data, the model analyzes patterns and predicts the corresponding lung disease class. Each algorithm is trained and evaluated to determine the best-performing model for accurate classification. This approach ensures reliable detection and classification of lung diseases, aiding clinicians in early diagnosis and treatment planning. The integration of advanced feature selection and diverse algorithms makes the system versatile and effective for real-world medical applications.

Keywords:

Lung Disease Detection, Spirometric Data, Machine Learning, Decision Tree, Random Forest,XGBoost, Support Vector Machine (SVM), CatBoost, Gradient Boosting, Logistic Regression,,K-best Feature Selection,Dimension Reduction

NOTE: Without the concern of our team, please don't submit to the college. This Abstract varies based on student requirements.