Segmenting “Flares” in Ultrasound Images Using Prior Statistics

The common method neonatologists use nowadays to determine White Matter Damage (leukomalacia) is by visual inspection of ultrasound images of the neonatal brain. A need for a (semi)computerized way of delineating the affected regions, in order to make quantitative measurements as an aid to the subjective diagnosis, is felt. The use of active contours for this purpose is a classical approach [1, 2]. The performance of active contours for this purpose, however, is heavily deteriorated by the presence of speckle noise. In this article a new filter, incorporating prior statistics concerning medical features in these images, is proposed, that removes a significant amount of speckle noise in the healthy parts, while it makes regions affected by WMD more uniform, thus severely improving the performance of the active contour. The results of the active contour after applying the proposed technique are compared with the manual delineation of an expert. Furthermore the proposed technique is compared with two other popular speckle suppression techniques, namely the ones proposed by Lee and Frost. Keywords— speckle, medical ultrasound, texture, adaptive filter, neonatal brain, leukomalacia, active contour, segmentation