DETECTION OF HEART DISEASES USING ARTIFICIAL INTELLIGENCE AND NOVEL DIGITAL STETHOSCOPE

ABSTRACT:-

            According to World Health Organization (WHO), the number of patients are increasing who died from heart failure or heart diseases. More people die annually from heart diseases than from any other disease. These murmur detection and heart sound detection techniques are most important to predict heart diseases. Heart sound detection is very useful for doctors to diagnose heart diseases. In other words, these studies are a substructure of clinical decision support systems. In India, the rate of dying people is very high due to improper diagnosis of heart diseases. Therefore, this paper develops a mobile application and system to detect cardiac abnormalities like Murmurs. Most people who have heart valve defects can be detected by using this technique. But the heart sound contains different background noises called Murmur. We have designed a unique system to record the heart sound of a patient and store it on the server. To detect heart diseases, heart sound is pre-processed via algorithms like audio slicing and segmentation and classified by pre-trained data using Convolution Neural Networks (CNN). This sound is passed to the CNN classifier and the classifier determines if the patient has heart disease or not. This system is very useful to detect heart diseases. This system is very efficient for home care for patients who have heart problems. And it has more advantages than an ECG system or other heart disease detection machine. This system will be a great advantage  to the healthcare industry for diagnosis of heart diseases in initial stages of clinical procedure.

OBJECTIVES OF THE PROJECT:-

The main objectives of this system are:

• Detect the early stage of Heart diseases like Murmur.
• Record Heart sound of patients and detect if the patient has heart disease or not. 
• To check the overall performance of a deep learning algorithm to look at murmurs and manage the heart sound of a patient recorded from the digital Stethoscope.
• To develop a machine learning and CNN based solution for detection of murmur qualities.
• To detect the heart sound and diagnose it at the early stage of  the heart disease that can save the patient’s lives.
• Save the lives of people.

MODULES:

• User Register/Login: User have to register/login themself to check heart condition and also user can monitor.
• Heart rate monitoring: QRS can extract heart rate value for monitoring and disease classification By fetching real time heart rate from sensor through arduino and raspberry pi.
• Heart rate analysis: Through QRS library heart rate values get extracted for analysis from AD8232.

HARDWARE AND SOFTWARE REQUIREMENTS:-

HARDWARE:-

• Processor: Intel Core i3 or more.
• RAM: 4GB or more.
• Hard disk: 250 GB or more.

SOFTWARE:-

• Operating System : Windows 10, 7, 8.
• Python 
• Anaconda
• Spyder, Jupyter notebook, Flask.
• MYSQL