Anatomically Constrained Neural Networks (ACNNs): Application to Cardiac Image Enhancement and Segmentation

• Published in: IEEE transactions on medical imaging Vol. 37; no. 2; pp. 384 - 395
• Main Authors: Oktay, Ozan, Ferrante, Enzo, Kamnitsas, Konstantinos, Heinrich, Mattias, Wenjia Bai, Caballero, Jose, Cook, Stuart A., de Marvao, Antonio, Dawes, Timothy, O'Regan, Declan P., Kainz, Bernhard, Glocker, Ben, Rueckert, Daniel
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.02.2018; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Artefacts; Artificial neural networks; Benchmarks; Biomarkers; Biomedical imaging; Cardiac Imaging Techniques - methods; Cardiomyopathies - diagnostic imaging; Computational modeling; convolutional neural network; Data processing; Databases, Factual; Heart; Heart - diagnostic imaging; Heart diseases; Humans; Image acquisition; Image analysis; Image enhancement; Image processing; Image resolution; Image segmentation; image super-resolution; Imaging, Three-Dimensional - methods; Magnetic Resonance Imaging; medical image segmentation; Model accuracy; Motion segmentation; Neural networks; Neural Networks, Computer; Performance enhancement; Regularization; Shape; Shape prior; State of the art; Three dimensional models; Training
• ISSN: 0278-0062; 1558-254X; 1558-254X
• DOI: 10.1109/TMI.2017.2743464

Incorporation of prior knowledge about organ shape and location is key to improve performance of image analysis approaches. In particular, priors can be useful in cases where images are corrupted and contain artefacts due to limitations in image acquisition. The highly constrained nature of anatomical objects can be well captured with learning-based techniques. However, in most recent and promising techniques such as CNN-based segmentation it is not obvious how to incorporate such prior knowledge. State-of-the-art methods operate as pixel-wise classifiers where the training objectives do not incorporate the structure and inter-dependencies of the output. To overcome this limitation, we propose a generic training strategy that incorporates anatomical prior knowledge into CNNs through a new regularisation model, which is trained end-to-end. The new framework encourages models to follow the global anatomical properties of the underlying anatomy (e.g. shape, label structure) via learnt non-linear representations of the shape. We show that the proposed approach can be easily adapted to different analysis tasks (e.g. image enhancement, segmentation) and improve the prediction accuracy of the state-of-the-art models. The applicability of our approach is shown on multi-modal cardiac data sets and public benchmarks. In addition, we demonstrate how the learnt deep models of 3-D shapes can be interpreted and used as biomarkers for classification of cardiac pathologies.