Fuzzy Logic and Neural Network for Clustering Heart beat Data

Posted on 2011-10-08
Last Modified: 2013-11-25
Is it possible to combine Fuzzy Logic and Neural Network to classify the data of Human Heart
beat to find whether is healthy or not healthy ..

If , yes how ? in matlab
If ,no .. What do you suggest    
Question by:Puneet Arora
    LVL 26

    Expert Comment

    Neural Nets could work if you have large data sets of normal and several/all types
    of abnormal EKG's.

    Not sure how you would use Fuzzy Logic.
    LVL 37

    Expert Comment

    There's not much sense in combining fuzzy logic with neural networks. Neural networks basically give a 'fuzzy' result anyway. You'll just set some threshold to determine between healthy and not.

    In a test for unhealthy heartbeats, false positives are much better than false negatives. So don't pick the threshold that gives the most accuracy, pick one that gives good accuracy and minimizes false negatives.

    (It's much better to let someone think their heart is unhealthy when it isn't than to tell someone he is fine when that is not true).
    LVL 44

    Expert Comment

    Have you done a pub search?  I seem to recall that someone has already applied a neural net or genetic algorithm to EKG data.  This is a recent research article on the subject:

    Also, I think that I read about it in a prior article at least three years ago.  I don't think they used Matlab.

    I agree with Tommy that fuzzy logic probably wouldn't be a good approach if you are trying to save someone's life by automating EKG pattern recognition -- too many false positives.

    Author Comment

    by:Puneet Arora
    I agree  that Neural Network based classification of Heart beat data will give a result which is
    Fuzzy in sense .

    Can I then call it Neuro-Fuzzy Classification or simply a classification based on Neural Network ...

    How must I calculate the threshold ...for reducing false positive rate ..What do you suggest ..
    LVL 44

    Expert Comment

    I guess it depends on how you train it and how many you train.  For instance, you could have several differently trained programs running in parallel (multi-core systems) trying to identify different cardiac pathologies/anomalies.
    LVL 37

    Accepted Solution

    You don't need the word 'Fuzzy' at all.

    When you run the neural network on your test data it will kick back a number (usually scaled from 0 to 1). Close to 0 will mean healthy (or unhealthy) and close to 1 will mean the other. Generally there is a gray area in the middle where data points from both sets end up.

    Let's say that 0 is unhealthy and 1 is healthy and between .4 and .6 there are both healthy and unhealthy heart data points. Classify all these as unhealthy (so set the threshold at .6).

    When you see the results of your data, it should all make sense.

    Featured Post

    What Is Threat Intelligence?

    Threat intelligence is often discussed, but rarely understood. Starting with a precise definition, along with clear business goals, is essential.

    Join & Write a Comment

    Go is an acronym of golang, is a programming language developed Google in 2007. Go is a new language that is mostly in the C family, with significant input from Pascal/Modula/Oberon family. Hence Go arisen as low-level language with fast compilation…
    Basic understanding on "OO- Object Orientation" is needed for designing a logical solution to solve a problem. Basic OOAD is a prerequisite for a coder to ensure that they follow the basic design of OO. This would help developers to understand the b…
    This tutorial covers a step-by-step guide to install VisualVM launcher in eclipse.
    The viewer will be introduced to the member functions push_back and pop_back of the vector class. The video will teach the difference between the two as well as how to use each one along with its functionality.

    729 members asked questions and received personalized solutions in the past 7 days.

    Join the community of 500,000 technology professionals and ask your questions.

    Join & Ask a Question

    Need Help in Real-Time?

    Connect with top rated Experts

    18 Experts available now in Live!

    Get 1:1 Help Now