On Oct 24, 4:23 am, Anonymous wrote: > Hello everybody, > I'm in trouble coz I have to make a program which is able to diagnosis tuberculosis or bacterial infection but I really don?t know how to do. > I can send the pictures which must be processed, with some explanation about how to know which pictures is tuberculosis or bacterial infection. > If someone can help me to do it, I will really appreciate it. > Thank you -------------------------------------
You may get better advise on an algorithm to characterize an image or find something in the image in sci.image.processing than here. Post your image to some public image hosting web site. Then ask again.
I'm not sure why you're doing this - maybe it's a grad school thesis/ dissertation project, and that's fine - but this will never fly in the medical market. There is heavy, heavy bias from doctors against any software that claims to do diagnosis. I think basically they don't want to use any software that claims it's smarter than them and can take doctors out of the equation. They just won't buy it. They won't believe it, and if they did, they wouldn't buy something that would put them out of a job. The only strategy that works is to spin your product such that it gives them advice or tells them where they might want to take a closer look, but never claim that you'll make a diagnosis. You'll get totally ignored. (I worked in medical imaging, with radiologists and ophthalmologists, for about 7 or 8 years.) Regards, ImageAnalyst