Prediction and Diagnosis of Thoracic Diseases using Rough Set and Machine Learning

• Published in: 2023 10th International Conference on Computing for Sustainable Global Development (INDIACom) pp. 206 - 213
• Main Authors: Hota, Radhanath, Dash, Sachikanta, Mishra, Sujogya, Pradhan, Sipali, Pattnaik, P. K., Pradhan, Geetanjali
• Format: Conference Proceeding
• Language: English
• Published: Bharati Vidyapeeth, New Delhi 15-03-2023
• Subjects: AdaBoost Model; Cardiac arrest; Cardiac Diseases; Computational modeling; Dimensionality reduction; Feature extraction; Logistic Regression; Medical tests; Predictive models; Rough sets; RST; UCI dataset

Thoracic diseases, which include all cardiac disorder such as cardiac arrest, pneumonia, and cardiac Asthma, are the maj or health disorder that needs significant scientific analysis for diagnosis and prediction. Thoracic diseases with similar symptoms are often difficult to diagnose without supporting medical tests. In this study, a random data set of several diseases are considered for accurate diagnosis and prediction, collected from different medical sources. Applying logistic regression on those data, the data set was reduced to six distinct groups, applying Rough Set Theory (RST) on it, results are related to thoracic diseases. The cardiac arrest comes under the domain of thoracic diseases, which has fatal consequencesnces. A universal data set taken from the University of California, Irvine Medical Centre (UCI) is being considered in this work for further analysis. To achieve significant results accuracy AdaBoost machine-learning model is used