Hopfield Neural Network for the segmentation of Near Infrared Fluorescent images for diagnosing prostate cancer

Although prostate cancer can be a slow-growing cancer, thousands of men die of the disease each year. Over the last decade, the nature of diagnostic healthcare has changed rapidly owing to an explosion in the availability of patient data, which are used as input data to Computed-Aided Diagnosis systems. The aim of this research is to develop a prototype system for the detection and classification of prostate tumors using Near-infrared and Mid-infrared spectrums of prostate pathological images. This optical imaging technique is a potent tool in cancer investigation that relies on stimulating endogenous chromophores or applying contrast agents able to target cancer cells. Here, we present a segmentation method of images obtained using Prostate Specific Membrane Antigen (PSMA) targeted Near Infrared Fluorescence (NIRF) optical imaging probes for intraoperative visualization of prostate cancer. An Artificial Neural Network classifies the pixels into distinguished clusters. Preliminary results demonstrate that the proposed segmentation method can enhance the existing clinical practice in identifying prostate area in the NIRF image, shape and volume analysis could be conducted using the segmentation result for further investigations.