发表机构
Perceptra Co., Ltd.; Radiology Department, Faculty of Medicine Siriraj Hospital, Mahidol University; Institute of Field Robotics, King Mongkut’s University of Technology Thonburi（Perceptra有限公司; 玛希隆大学诗里拉吉医院医学院放射科; 泰国国王科技大学吞武里分校野外机器人研究所）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
泰国因放射科医生短缺，胸部X光片解读受限。本文开发的Inspectra CXR版本5深度学习系统，结合DenseNet - 121等架构，能进行疾病分类与病变定位。经多组测试及评估，该系统在泰国不同医院表现良好，赢得放射科医生信任，为胸部疾病检测提供有效方案。
AI 中文摘要
胸部X光摄影（CXR）仍是最广泛使用的胸部成像方式，但在泰国及东南亚，专家解读因放射科医生严重短缺而受限。深度学习模型本地化适应泰国数据可大幅提高对泰国人群的准确性。本文介绍了Inspectra CXR版本5中胸部X光片分析模型的开发与全面验证，该模型能在单一模型中进行多标签胸部疾病分类和弱监督病变定位。其架构将DenseNet - 121主干与Attend - and - Compare模块（ACM）及概率类激活映射（PCAM）聚合层相结合，同时产生每个病症的分类分数和热图。该模型基于曼谷诗里拉吉医院的874,858张 frontal chest radiographs及配对放射科报告开发。在19,871例经放射科医生验证的域内测试集上，九个临床重要病症的平均AUROC为0.994（平均敏感度92.4%，特异度98.6%）。在来自泰国13家医院的5,992例独立泛化集上，平均AUROC为0.970。对于定位，在4,549例经放射科医生标注的病例上评估，模型在每张图像0.59个非病变定位时的平均病变定位分数（LLF）为77.9%。在与五位胸部放射科医生的可用性评估中，系统的分类一致性为93.6%，定位一致性为94.7%，平均系统可用性量表（SUS）得分为89。这些结果表明，本地开发的、具备定位能力的CXR系统可提供高精度，在不同的泰国医院中实现泛化，并赢得执业放射科医生的信任。
英文摘要
Chest radiography (CXR) remains the most widely used thoracic imaging modality, yet expert interpretation is constrained by a severe shortage of radiologists in Thailand and across Southeast Asia. Local adaptation of deep learning models to Thai data has been shown to substantially improve accuracy on Thai populations. Here we present the development and comprehensive validation of the chest radiograph analysis model in Inspectra CXR version 5, a deep learning system that performs multi-label thoracic disease classification and weakly supervised lesion localization within a single model. The architecture couples a DenseNet-121 backbone with Attend-and-Compare Modules (ACM) and a Probabilistic Class Activation Map (PCAM) aggregation layer, producing a per-condition classification score and heatmap simultaneously. The model was developed on 874,858 frontal chest radiographs with paired radiologist reports from Siriraj Hospital, Bangkok. On a held-out, radiologist-verified in-domain test set of 19,871 cases, it achieved a mean AUROC of 0.994 (mean sensitivity 92.4%, specificity 98.6%) across nine clinically important conditions. On an independent generalization set of 5,992 cases from 13 hospitals across Thailand, the mean AUROC was 0.970, indicating robust transfer across sites. For localization, evaluated on 4,549 radiologist-annotated cases, the model attained a mean lesion-localization fraction (LLF) of 77.9% at 0.59 non-lesion localizations per image. In a usability evaluation with five thoracic radiologists, the system reached a classification concordance of 93.6%, a localization concordance of 94.7%, and a mean System Usability Scale (SUS) score of 89. These results indicate that a locally developed, localization-capable CXR system can deliver high accuracy, generalize across heterogeneous Thai hospitals, and earn the trust of practicing radiologists.
