Segmenting the Uterus in Monocular Laparoscopic Images without Manual Input

Automatically segmenting organs in monocular laparoscopic images is an important and challenging research objective in computerassisted intervention. For the uterus this is difficult because of high interpatient variability in tissue appearance and low-contrast boundaries with the surrounding peritoneum. We present a framework to segment the uterus which is completely automatic, requires only a single monocular image, and does not require a 3D model. Our idea is to use a patientindependent uterus detector to roughly localize the organ, which is then used as a supervisor to train a patient-specific organ segmenter. The segmenter uses a physically-motivated organ boundary model designed specifically for illumination in laparoscopy, which is fast to compute and gives strong segmentation constraints. Our segmenter uses a lightweight CRF that is solved quickly and globally with a single graphcut. On a dataset of 220 images our method obtains a mean DICE score of 92.9%.