Feature extraction from retina vascular images for classification

Classifying medical images is a tedious and complex task. Using machine learning algorithms to assist the process could be a huge help. There are many challenges to making machine learning algorithms work reliably on image data. First of all, you need a rather large image database with ground truth information (expert’s labeled data with diagnosis information). The second problem is preprocessing images, including merging modalities, unifying color maps, normalizing, and filtering. This part is essential and may impact the last part – feature extraction. This step is crucial because on how well you can extract informative features depends on how well machine learning algorithms will work. Dataset To demonstrate the classification procedure of medical images, the ophthalmology STARE (Structured Analysis of the Retina) image database was pulled from https://cecas.clemson.edu/~ahoover/stare/. The database consists of 400 images with 13 diagnostic cases along with preprocessed images. For classification problem, we have chosen only vessels images and only two classes: Normal and Choroidal Neovascularization (CNV). So the number of images was reduced to 99 where 25 were used as test data and 74 as training.

Continue reading