Automatic segmentation variability estimation with segmentation priors

Segmentation of anatomical structures and pathologies in medical images is a fundamental technical problem in medical image processing. Segmentation is the state-of-practice in a variety of medical products and is gaining prominence in clinical practice, as more clinicians rely on it for a wide variety of clinical tasks, including diagnosis, treatment planning, treatment delivery, and treatment evaluation. Indeed, segmentation is state-of-practice and is a key component in an increasing number of commercial medical products applications such as multi-modal tumor volumetry and tumor tracking, patient-specific custom jig printing for orthopaedic surgery, non-invasive CT-based coronary heart disease detection, minimally invasive spinal pedicle screw insertion, and image analytics in oncology clinical trials to name only a few (Medical Image Analysis Software Market Reports, 2016a, 2016b).
Source: Medical Image Analysis - Category: Radiology Authors: Source Type: research