Detection of Prostate Abnormality within the Peripheral Zone Using Grey Level Distribution

Development of CAD systems for detection of prostate cancer has been a recent topic of research and remains a challenging task. In this paper, we propose a novel method of prostate cancer detection within the peripheral zone. The key idea is to assume that every grey level could be associated with malignant or normal tissues by using a weighted probability. Based on the weighting, we use specific metrics to determine abnormality. We show experimental results to illustrate the performance of this method in comparison to some previous studies. Initial results show that our method achieved 81% correct classification result and 9% and 10% false positive and false negative results, respectively (sensitivity/specificity: 0.85/0.72).