Computer Vision News - June 2016

Every month, Computer Vision News reviews a challenge related to our field, be it in medical imaging, automotive, robotics or else. If you don’t find time to read challenges, but are interested in the new methods proposed by the scientific community to solve them, you can read our challenge summaries! This month we have chosen to review the Cervical Cytology Image Segmentation challenge , held under the auspices of the IEEE International Symposium on Biomedical Imaging ( ISBI 2015) . The website of the challenge is here . Background Cervical cancer is the fourth most common cancer in women, with more than half million new cases every year . More than 80% of cases occur in the less developed regions of the globe. It is considered the most common cancer in women in Eastern and Middle Africa. Cervical cancer mortality is as high as a quarter of a million deaths every year, accounting for 7.5% of all female cancer deaths. 87% of cervical cancer deaths occur in the less developed regions, where mortality is dramatically higher. With the exception of skin cancer, cervical cancer has the greatest potential for prevention and cure when detected early enough , thanks to a pre-cancerous condition that is recognized and treated in time when preventive measures are applied. The main screening strategy is based on Pap smear , a simple, quick and essentially painless procedure used to detect pre-cancerous and cancerous processes, which is carried out by collecting a sample of cells from the cervix and smearing them onto a glass slide for further examination under a microscope. This test is typically based on human visual analysis, generating high intra- and inter-observer variability and leading to a large variation in false negative results. In particular, when the sampled cells are numerous and overlapping, the quality of human observation is affected, hence the call for automated image analysis techniques aimed at ensuring greater consistency in these tests and reducing the occurrence of false negatives. The Cervical Cytology Image Segmentation challenge requires to suggest computational tools to automatically detect, segment and classify the cells present in multi-layer cervical cell volumes from a microscope slide. Cervical cancer cells 28 Computer Vision News Challenge Cervical Cytology Image Segmentation