DeepIGeoS: A Deep Interactive Geodesic Framework for Medical Image Segmentation

Accurate medical image segmentation is essential for diagnosis, surgical planning and many other applications. Convolutional Neural Networks (CNNs) have shown to be state-of-the-art automatic segmentation methods while the result still needs to be refined to become accurate and robust enough for clinical use. We propose a deep learning-based interactive segmentation method in order to improve the segmentation obtained by an automatic CNN as well as reduce user interactions during refinement for better results. We use one CNN to obtain an initial segmentation automatically, on which user interactions are added to indicate mis-segmentations. Another CNN takes as input the user interactions with the initial segmentation and gives a refined result. We propose a new way to combine user interactions with CNNs through geodesic distance maps, and propose a resolution-preserving network that can give better dense prediction. In addition, we integrate user interactions as hard constraints into back-propagatable Conditional Random Fields. We validated the proposed framework in the application of placenta segmentation from fetal MRI and clavicle segmentation from chest radiographs. Experimental results show our method achieves a large improvement from automatic CNNs, and obtains comparable accuracy with fewer user interventions and less time compared with traditional interactive methods.