Please use this identifier to cite or link to this item: http://hdl.handle.net/10266/6093
Title: Development of Cardiovascular Disease Prediction System
Authors: Narain, Renu
Supervisor: Saxena, Sanjai
Goyal, Achal Kumar
Keywords: Quantum neural network;Heart disease;Heart attack;Artificial Intelligence;Algorithm;Prognostic system
Issue Date: 8-Apr-2021
Abstract: This work presents the intelligent Cardiovascular Disease (CVD) prediction system based on machine learning, which uses Quantum Neural Network for machine learning. Early medical diagnosis of Heart disease is very important and should be performed accurately and efficiently. Unfortunately, the physicians don’t have the enough time to analyze past history of patient in depth. This Intelligent system would enhance the medical care and reduce costs, by quick analysis of past data of patients with percentage of risk prediction. The accuracy of this intelligent system is significantly higher than other existing prognostic systems. All available Physical, Physiological, Clinical parameters have been considered in this study. The data of 815 Patients suffering with the symptoms of Heart disease has been collected from hospital and used for training and evaluation. Furthermore, the dataset of famous Framingham study consisting 5209 CVD patients’ data has been used for validation purpose. All the patients’ reports have been diagnosed and analysed by medical practitioner previously. This system uses the Quantum Neural Network for machine learning. The results obtained have high degree of sensitivity and specificity that matches with the expert’s opinion with 98.5% accuracy. Such an expert system would be very useful when incorporated with other systems to provide diagnostic and predictive medical opinions in a timely manner. This system will work as an aid to physician for prognosis of heart disease. Using this system, medical practitioners may plan better medication and treatment strategy. The overall accuracy of this intelligent heart disease prognostic system is 98.5%, which is significantly higher than other existing approaches.
URI: http://hdl.handle.net/10266/6093
Appears in Collections:Doctoral Theses@DBT

Files in This Item:
File Description SizeFormat 
Renu Narain PhD Thesis _Updated.pdfPh.D. Thesis5.03 MBAdobe PDFThumbnail
View/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.