Abstract
Mobile Solution for Early Detection of Heart Diseases using Artificial Intelligence and Novel Digital Stethoscope - written by Devang Sharma , Saurabh Sahu , Dr. Amol Pande published on 2020/05/28 download full article with reference data and citations
Highlights
Heart diseases have emerged as the primary cause of deaths worldwide [4]
Our study aims to understand whether a low-cost system can be developed to detect early signs of valvular heart disorders
The Michigan Dataset consisted of 23 audio files and is very small to train and validate a robust Convolution Neural Networks (CNN) model, this dataset was used to identify ideal parameter settings for a low-pass filter, audio splitting, and spectrogram conversion
Summary
Heart diseases have emerged as the primary cause of deaths worldwide [4]. A study of the Medical Council of India’s historical data conducted in 2017 shows that there were only. Illustrates the difference between the PCG of a normal and abnormal heart sound. CNN model is primarily trained with normal and abnormal heart sounds of three open-source repositories. H. Training Dataset – We primarily used 3 open datasets to train our CNN model: 1.) PASCAL’s Classifying Heart Sounds Challenge: It consists of data gathered from two sources: Dataset A – recorded via the iStethoscope Pro iPhone app from the general public via, and Dataset B – recorded using the digital stethoscope DigiScope from a clinical trial in hospitals. Dataset A consists of 176 audio files in WAV format out of which 31 are normal, 34 are murmur and remaining are other cardiac diseases. Validation set – a copy of 300 records from the training set [3]. 3.) University of Michigan: It consists of 23 audio files of approximately one-minute length taken from 4 different sections, i.e., Apex Area Supine, Apex Area - Left Decubitus, Aortic Area - Sitting; Listening with the bell of the stethoscope, Pulmonic Area Supine; Listening with the diaphragm of the stethoscope [1]
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
