Learning-based Detection and Tracking in Medical Imaging: A Robust Information-Fusion Approach

Medical image processing tools are playing an increasingly important role in assisting the clinicians in diagnosis, therapy planning and image-guided interventions. Accurate, robust and fast tracking of deformable anatomical objects such as the heart, is a crucial task in medical image analysis. One of the main challenges is to maintain an anatomically consistent representation of target appearance that is robust enough to cope with inherent changes due to target movement, imaging device movement, varying imaging conditions and is consistent with the domain expert clinical knowledge. To address these challenges this chapter presents a robust learning-based fusion framework that relies on anatomically indexed componentbased object models that integrate several sources of information to determine the temporal trajectory of the deformable target. Large annotated imaging databases are exploited to encode the domain knowledge in shape models and motion models and to learn discriminative image classifiers for the target appearance. The framework robustly fuses the prior information with traditional tracking approaches based on template matching and registration. We demonstrate various medical image analysis applications with focus on cardiology such as 2D auto left heart, catheter detection and tracking, 3D cardiac chambers surface tracking, and 4D complex cardiac structure tracking, in multiple modalities including Ultrasound (US), Cardiac Computed Tomography (CT), Magnetic Resonance Imaging (MRI), and X-ray fluoroscopy.