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

File Description SizeFormat 
FINAL_VERSION.pdfAccepted version4.53 MBAdobe PDFDownload
SUPPLEMENTARY.pdfSupporting information73.72 kBAdobe PDFDownload
08051114.pdfPublished version8.9 MBAdobe PDFDownload
Title: Anatomically Constrained Neural Networks (ACNN): Application to Cardiac Image Enhancement and Segmentation
Author(s): Oktay, O
Ferrante, E
Kamnitsas, K
Heinrich, M
Bai, W
Caballero, J
Cook, S
De Marvao, A
Dawes, T
O'Regan, D
Kainz, B
Glocker, B
Rueckert, D
Item Type: Journal Article
Abstract: 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 datasets and public benchmarks. Additionally, we demonstrate how the learnt deep models of 3D shapes can be interpreted and used as biomarkers for classification of cardiac pathologies.
Publication Date: 26-Sep-2017
Date of Acceptance: 15-Aug-2017
URI: http://hdl.handle.net/10044/1/50440
DOI: https://dx.doi.org/10.1109/TMI.2017.2743464
ISSN: 0278-0062
Publisher: IEEE
Journal / Book Title: IEEE Transactions on Medical Imaging
Copyright Statement: This work is licensed under a Creative Commons Attribution 3.0 License. For more information, see http://creativecommons.org/licenses/by/3.0/.
Sponsor/Funder: British Heart Foundation
Engineering & Physical Science Research Council (EPSRC)
Imperial College Healthcare NHS Trust- BRC Funding
Funder's Grant Number: PG/12/27/29489
EP/P001009/1
RDC04
Keywords: 08 Information And Computing Sciences
09 Engineering
Nuclear Medicine & Medical Imaging
Publication Status: Published online
Appears in Collections:Faculty of Engineering
Computing
Clinical Sciences
Imaging Sciences
Faculty of Medicine



Items in Spiral are protected by copyright, with all rights reserved, unless otherwise indicated.

Creative Commons