The objective of this work is a new approach to computer-based analysis of Feulgen-stained urinary bladder cell nuclei to identify neoplastic nuclei. 56 patients were taken under consideration. Among them 20 cases were controls and 36 cases were cancer of two grades of malignancy. The cytological smears were obtained by "bladder washing" technique. Image analysis was carried out by means of digital image processing system designed by the authors. At the first step of the process slides were automatically scanned and 80 images were registered. Then image processing began performing image normalization, objects (nuclei) extraction and measuring of features describing nuclei. A multistage classifier was constructed to identify positive and negative cases. The results of this study yielded a 90% correct classification rate in the control group, while an 80,5% rate was obtained among the cancer patients. The sensitivity of the method was 91% and the specificity was 84%.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.