Extraction and analysis of nail-fold capillaries

Nail-fold capillaries exhibit distinctive features which help in early diagnosis of various anomalies like diabetes. The examination of microcirculation of nail-fold capillaries serves as the window to monitor one´s health condition as the changes in nail-fold microcirculation often indicate certain clinical disorders. This way of finding disorders is a less harmful technique. Separation of multiple layers of capillary in an image is performed using Speeded Up Robust Features (SURF). The capillary images have lot of smudge and blur in them which makes it difficult to extract the capillaries. This is tackled by using varied metric threshold in SURF features. Cross-correlation between a reference capillary image and the test image is used to separate one layer of capillaries. Extraction of the single capillary is performed via SURF points. This gives the density of capillaries in an image and also talks about the avascularity. The proposed techniques gave satisfactory results in the measurement of density of capillaries and avascularity and hence detecting the anomalies.