Jump to content

jinp6301

High Rollers
  • Posts

    7,402
  • Joined

  • Last visited

Posts posted by jinp6301

  1. I'm guessing that ANN is Artificial Neural Network which needs multiple examples to train but I guessing that the convolution based approaches allow you to look at across the length scales and correlate that with past tumors length scales which would be very interesting.....

    As far as websites go I don't know I believe all I used was Mathworld....

    Thanks. Even that small tidbit helped with my understanding :)

  2. Thankfully convolution in the time domain is multiplication in the frequency domain.:) You doing discrete or continuous stuff?

    What's the project. I do a lot of convoluted convolutions :) Time domain and frequency domain.

    Using convolutions/cross correlation for pattern recognition (tumor recognition) in CT/MRI images. I think I got the general idea, but I cant see how this is better than using ANN/SVMs :(

    But if any of you guys have any websites/pages I should read up on, that would be great :)

  3. Same as you minus the biking part. And the working part. I don't really know heh. Submitted my dissertation this semester. Taking 3 classes and teaching 7 so busy. Very busy.

    Good to see you posting again.

    teaching 7 classes? wtf? does the school even let you teach 7 classes?

×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.