Contrast enhancement for hypoechoic lesions.

We propose a novel method for contrast enhancement in hypoechoic lesions. Our approach is based on the concept of distance on image manifolds to obtain a credibility function which captures the delineation of the lesion; if the lesion is sharply defined, the credibility function will also give a sharp response, whereas it varies smoothly at locations where no reliable information about the lesion's perimeter exists due to shortcomings of image acquisition, or because the lesion's border itself is only vaguely defined. The credibility function is used to either increase contrast of the lesion with respect to the adjacent tissue or to stain the lesion by a color overlay.

Leukocyte segmentation and classification in blood-smear images.

The detection and classification of leukocytes in blood smear images is a routine task in medical diagnosis. In this paper we present a fully automated approach to leukocyte segmentation that is robust with respect to cell appearance and image quality. A set of features is used to describe cytoplasm and nucleus properties.