views:

121

answers:

2

Hi guyz,

I need some help with a algorithm I'm working with for my college course. The idea is use an artificial neural network to read a electrocardiogram and try to recognize some disturbs in the waves, that's ok, I've the neural network and I can test it, no problem, but I'd like to give the function to the user to open a eletrocardiogram (import a jpeg) and the program find the waves and convert it in to the arrays that will feed my ANN, but there's the problem. I did some code that read the image and transform it in a binary image, but I can't find a nice way for the program to locate the waves, since the exact position can vary from hospital to hospital, I need some suggestions of approachs I should use.

Thankz!!

A: 

You'd have to assume a certain minimal contrast between the "signal" (the waves) and the background of the image. An edge-finding algorithm might be useful in that case. You could isolate the wave from the background and plot the wave.

This post by Rick Barraza deals with vector fields in Silverlight. You might be able to adapt the concept to your particular problem.

Dave Swersky
+1  A: 

If you've got the wave values in a list, you can use a Fourier transform or FFT (fast Fourier transform) to determine the frequency content at any particular time value. Disturbances typically create additional high-frequency content (ie, sharp, steep waves) that you should be able to use to spot irregularities.

David Lively
Hi, thankz for the reply, but I think it's not about the Fourier transform (that will help too), but the question actually is how to read an image with a lot of text, logos and waves and find the waves and then make a rectangle over it, with the exact width and height. Then, after finding it, FT will be a hand.
Alaor
That sounds like a whole different (and much bigger) problem than what you described in your question. Are you sure that's what your instructor is really asking you to do!
Tom Bushell
The subject of the project is use a neural artificial network to identify disturbs in an ecg, that's ok, it's already working. This reading jpeg stuff is me who want to implement, the IA thing is done, it would be a plus.
Alaor