Leveraging anatomical constraints with uncertainty for pneumothorax segmentation  

作　　者：Han Yuan Chuan Hong Nguyen Tuan Anh Tran Xinxing Xu Nan Liu 

机构地区：[1]Centre for Quantitative Medicine,Duke-NUS Medical School,Singapore [2]Department of Biostatistics and Bioinformatics,Duke University,Durham,North Carolina,USA [3]Department of Diagnostic Radiology,Singapore General Hospital,Singapore [4]Institute of High Performance Computing,Agency for Science,Technology and Research,Singapore [5]Programme in Health Services and Systems Research,Duke-NUS Medical School,Singapore [6]Institute of Data Science,National University of Singapore,Singapore

出　　处：《Health Care Science》2024年第6期456-474,共19页科学医疗(英文)

基　　金：Duke-NUS Medical School,Singapore。

摘　　要：Background:Pneumothorax is a medical emergency caused by the abnormal accumulation of air in the pleural space—the potential space between the lungs and chest wall.On 2D chest radiographs,pneumothorax occurs within the thoracic cavity and outside of the mediastinum,and we refer to this area as“lung+space.”While deep learning(DL)has increasingly been utilized to segment pneumothorax lesions in chest radiographs,many existing DL models employ an end-to-end approach.These models directly map chest radiographs to clinician-annotated lesion areas,often neglecting the vital domain knowl-edge that pneumothorax is inherently location-sensitive.Methods:We propose a novel approach that incorporates the lung+space as a constraint during DL model training for pneumothorax segmentation on 2D chest radiographs.To circumvent the need for additional annotations and to prevent potential label leakage on the target task,our method utilizes external datasets and an auxiliary task of lung segmentation.This approach generates a specific constraint of lung+space for each chest radiograph.Furthermore,we have incorporated a discriminator to eliminate unreliable constraints caused by the domain shift between the auxiliary and target datasets.Results:Our results demonstrated considerable improvements,with average performance gains of 4.6%,3.6%,and 3.3%regarding intersection over union,dice similarity coefficient,and Hausdorff distance.These results were con-sistent across six baseline models built on three architectures(U-Net,LinkNet,or PSPNet)and two backbones(VGG-11 or MobileOne-S0).We further con-ducted an ablation study to evaluate the contribution of each component in the proposed method and undertook several robustness studies on hyper-parameter selection to validate the stability of our method.Conclusions:The integration of domain knowledge in DL models for medical applications has often been underemphasized.Our research underscores the significance of incorporating medical domain knowledge about the location-specific nature

关 键 词：constrained optimization deep transfer learning diagnostic radiology pneumothorax detection semantic segmentation 

分 类 号：R65[医药卫生—外科学]