Deep Architectures for High-Resolution Multi-organ Chest X-ray Image Segmentation Gómez, Óscar Mesejo Santiago, Pablo Ibáñez, Óscar Valsecchi, Andrea Cordón García, Óscar Semantic segmentation Chest X-ray segmentation Convolutional neural networks Deep networks simplification This research was supported by the Spanish Ministerio de Economía y Competividad under the NEWSOCO project [Grant Number TIN2015-67661-P], including European Development Regional Funds (EDRF). This work was also supported by the Spanish Ministry of Science, Innovation and Universities, and European Regional Development Funds (ERDF) under grant EXASOCO (PGC2018-101216-B-I00). Mr. Gómez’s work was supported by Spanish MECD FPU Grant [Grant Number FPU14/02380]. Pablo Mesejo is funded by the European Commission H2020-MSCA-IF-2016 through the Skeleton-ID Marie Curie Individual Fellowship [Grant Number 746592]. We acknowledge the support of NVIDIA Corporation with the donation of the Titan X Pascal GPU used for this research. Chest X-ray images (CXRs) are the most common radiological examination tool for screening and diagnosis of cardiac and pulmonary diseases. The automatic segmentation of anatomical structures in CXRs is critical for many clinical applications. However, existing deep models work on severely down-sampled images (commonly pixels), reducing the quality of the contours of the resulting segmentation and negatively affecting the possibilities of such methods to be effectively used in a real environment. In this paper, we study multi-organ (clavicles, lungs, and hearts) segmentation, one of the most important problems in semantic understanding of CXRs. We completely avoid down-sampling in images up to (as in the JSRT dataset), and we diminish its impact in higher resolutions via network architecture simplification without a significant loss in the accuracy. To do so, we propose four different convolutional models by introducing structural changes to the baselines employed (U-Net and InvertedNet) as well as by integrating several techniques barely used by CXRs segmentation algorithms, such as instance normalization and atrous convolution. We also compare single-class and multi-class strategies to elucidate which approach is the most convenient for this problem. Our best proposal, X-Net+, outperforms nine state-of-the-art methods on clavicles and lungs obtaining a Dice similarity coefficient of 0.938 and 0.978, respectively, employing a tenfold cross-validation protocol. The same architecture yields comparable results to the state of the art in heart segmentation with a Dice value of 0.938. Finally, its reduced version, RX-Net+, obtains similar results but with a significant reduction in memory usage and training time. 2024-11-07T07:27:33Z 2024-11-07T07:27:33Z 2020 journal article Published version: Gómez, Óscar. Deep Architectures for High-Resolution Multi-organ Chest X-ray Image Segmentation. Neural Computing & Applications 32:20 (2020) 15949-15963. https://doi.org/10.1007/s00521-019-04532-y https://hdl.handle.net/10481/96710 10.1007/s00521-019-04532-y eng info:eu-repo/grantAgreement/EC/H2020/746592 http://creativecommons.org/licenses/by-nc-nd/4.0/ open access Attribution-NonCommercial-NoDerivatives 4.0 Internacional Springer Nature