Contour-aware multi-label chest X-ray organ segmentation

• Published in: International journal for computer assisted radiology and surgery Vol. 15; no. 3; pp. 425 - 436
• Main Authors: Kholiavchenko, M., Sirazitdinov, I., Kubrak, K., Badrutdinova, R., Kuleev, R., Yuan, Y., Vrtovec, T., Ibragimov, B.
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.03.2020; Springer Nature B.V
• Subjects: Algorithms; Architecture; Artificial neural networks; Augmentation; Chest; Coders; Computer Imaging; Computer Science; Contours; Databases, Factual; Feature extraction; Health Informatics; Humans; Image Processing, Computer-Assisted - methods; Image segmentation; Imaging; Lung - diagnostic imaging; Lung Diseases - diagnostic imaging; Masks; Medical imaging; Medicine; Medicine & Public Health; Neural Networks, Computer; Organs; Original Article; Pattern Recognition and Graphics; Performance evaluation; Radiography, Thoracic - methods; Radiology; Shape; Surgery; Tomography, X-Ray Computed - methods; Vision; Image segmentation; JSRT database; Convolutional neural networks; Deep learning architectures; Chest X-ray (CXR) images
• ISSN: 1861-6410; 1861-6429; 1861-6429
• DOI: 10.1007/s11548-019-02115-9

Purpose Segmentation of organs from chest X-ray images is an essential task for an accurate and reliable diagnosis of lung diseases and chest organ morphometry. In this study, we investigated the benefits of augmenting state-of-the-art deep convolutional neural networks (CNNs) for image segmentation with organ contour information and evaluated the performance of such augmentation on segmentation of lung fields, heart, and clavicles from chest X-ray images. Methods Three state-of-the-art CNNs were augmented, namely the UNet and LinkNet architecture with the ResNeXt feature extraction backbone, and the Tiramisu architecture with the DenseNet. All CNN architectures were trained on ground-truth segmentation masks and additionally on the corresponding contours. The contribution of such contour-based augmentation was evaluated against the contour-free architectures, and 20 existing algorithms for lung field segmentation. Results The proposed contour-aware segmentation improved the segmentation performance, and when compared against existing algorithms on the same publicly available database of 247 chest X-ray images, the UNet architecture with the ResNeXt50 encoder combined with the contour-aware approach resulted in the best overall segmentation performance, achieving a Jaccard overlap coefficient of 0.971, 0.933, and 0.903 for the lung fields, heart, and clavicles, respectively. Conclusion In this study, we proposed to augment CNN architectures for CXR segmentation with organ contour information and were able to significantly improve segmentation accuracy and outperform all existing solution using a public chest X-ray database.