Heart Sounds Classification using Artificial Neural Network
According to the World Health Organization, cardiovascular diseases (CVDs) are the number one cause of death globally: more people die annually from CVDs than from any other cause. An estimated 17.1 million people died from CVDs in 2004, representing 29% of all global deaths. Of these deaths, an est
2025-06-28 16:32:52 - Adil Khan
Heart Sounds Classification using Artificial Neural Network
Project Area of Specialization Artificial IntelligenceProject SummaryAccording to the World Health Organization, cardiovascular diseases (CVDs) are the number one cause of death globally: more people die annually from CVDs than from any other cause. An estimated 17.1 million people died from CVDs in 2004, representing 29% of all global deaths. Of these deaths, an estimated 7.2 million were due to coronary heart disease.
The heart sound consists of four main parts: the first heart sound (S 1), the systolic period, the second heart sound (S2) and the diastolic period. The first heart sound (S 1) is produced by the closure of the Mitral (MV) and Tricuspid (TV) valves while for second heart sound 2) is caused by the closure of the aortic and pulmonic valves. The systolic period is a period between S 1 and S2 and for diastolic period, it is a period between S2 and S 1. In the case of abnormal heart sound, it often produces a sound called murmur. One of the types of heart disease is valvular heart disease. The murmur can happen in the disease which caused by valves that do not close tightly or blood that leaks backward in the valve.
Project Objectives- To classify heart sounds for the diagnosis of many cardiac abnormalities
- To apply Multilayer Neural Network for heart disease identification into four categories like normal, Murmur, Extra Heart Sound and Artifact
- To develop a protype for real time heart sound classification to assist health diagnosis.
Recently new developments using Digital Signal Processing (DSP) techniques results to quantify the heart sound characteristics. From the analysis, various heart valve-related diseases can be detected and classified. The current study considers heart sound cases of normal and 4 common types of valvular heart disease which are aortic stenosis, aortic regurgitation, mitral stenosis and mitral regurgitation.
Benefits of the ProjectAny method which can help to detect signs of heart disease could therefore have a significant impact on world health. Auscultation of the heart can provide clues to the diagnosis of many cardiac abnormalities. During heart auscultation, the observer listens and analyzes the heart sound components separately by using stethoscope.
Technical Details of Final Deliverable- Software Simulation results
- Comparative Study
- HW/SW integrated system
| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 75000 | |||
| Raspberry Pi Module | Equipment | 2 | 5600 | 11200 |
| Raspberry Pi Casings | Equipment | 2 | 500 | 1000 |
| Raspberry Pi Power Adoptors | Equipment | 2 | 600 | 1200 |
| Nexys 3 Spartan-6 FPGA Board | Equipment | 1 | 34000 | 34000 |
| Electronic Stethoscope | Equipment | 1 | 22600 | 22600 |
| OverHead Expenditure | Miscellaneous | 1 | 5000 | 5000 |