Classifying normal and abnormal heart sounds using Artficial Intelliigence

Well, learning different types of murmur is a cumbersome process for a medical student. It usually requires hours of practice to be able to differentiate between a normal heartbeat or heartbeat with a murmur or extra added sounds.

This is my first project on Deep Learning. I have used the dataset of heart sounds from http://www.peterjbentley.com/heartchallenge/. The heart sounds in the data is neither sufficient nor neat. But I had to start somewhere, and I choose this as I found the idea of teaching a computer to detect a murmur fascinating.I've used Python library numpy/scipy and librosa.

Click here to how the AI works.

The datasets are not neat but still, the Support Vector Machine of SkLearn works pretty good in classifying the heart sound with or without murmur. I am still working on it to make it better at recognizing heart sounds for the real-world purpose.

The code goes here.

https://github.com/vivekkarn/classification-of-heart-sounds/blob/master/classifying-heart-sounds-using-LSTM.ipynb

Comments

Popular posts from this blog

A simple explanation of Artificial Intelligence for people not belonging to the tech sector.

AI replaces radiologist doctors on diagnosing pneumonia

The best way to deal with depression and anxiety