Nuclei Cell Semantic Segmentation Using Deep Learning Unet

Semantic Segmentation has benefited a lot from the advancement of Deep Learning technologies. Image Segmentation has been a boon for medical field in both application as well as research areas where identification and localization of infected/abnormal parts of the organs can be extracted. Study of cell images from infected organs with the help of high definition medical images taken from MRI, CT, etc to provide treatment as well as make it possible for researchers to develop vaccines for such diseases. This study is on to train a network model tuned using different hyper-parameters so that the network learns to automatically segment nuclei from the medical images available from standard database library. Segmentation is the key problem before the any other type of problems can be encountered like that of nuclei detection and classification. This study also looks at the performance of the network build through evaluation matrix.