A data-centric deep learning approach to airway segmentation

Respiratory Medicine, Royal Free London NHS Foundation Trust, London, UK Corresponding author: Dr Joseph Jacob UCL Centre for Medical Image Computing 1st Floor, 90 High Holborn, London WC1V6LJ j.jacob@ucl.ac.uk Abstract The morphology and distribution of airway tree abnormalities enables diagnosis and disease characterisation across a variety of chronic respiratory conditions. In this regard, airway segmentation plays a critical role in the production of the outline of the entire airway tree to enable estimation of disease extent and severity. In this study, we propose a data-centric deep learning technique to segment the airway tree. The proposed technique utilises interpolation and image split to improve data usefulness and quality. Then, an ensemble learning strategy is implemented to aggregate the segmented airway trees at different scales. In terms of segmentation performance (dice similarity coefficient), our method outperforms the baseline model by 2.5% on average when a combined loss is used. Further, our proposed technique has a low GPU usage and high flexibility enabling it to be deployed on any 2D deep learning model. Introduction Abnormal dilatation of the airways is a key feature in the diagnosis of idiopathic pulmonary fibrosis (IPF) patients. Disease extent and severity in IPF can be assessed by visual analysis of high-resolution CT images by radiologists.