基于泰国数据训练的深度学习模型，可同时识别胸部疾病并定位病灶。
From Classification to Localization and Clinical Validation: Large-Scale Development of a Deep Learning System for Thoracic Disease Detection on Chest Radiographs in Thailand
- 采用密集连接网络与注意力机制，单模型实现多病种分类与弱监督定位
- 在1.9万张泰国本地测试图上平均准确率超99%，跨13家医院泛化能力良好
- 放射科医生使用满意度高，定位精度达77.9%，适合临床辅助诊断场景
胸片（CXR）是胸部影像最常用手段，但泰国及东南亚地区放射科医生严重短缺。本地化深度学习模型在泰国人群中的应用显著提升准确性。本文介绍Inspectra CXR v5的开发与全面验证，该模型基于874,858张曼谷诗里拉吉医院的前后位胸片及其配对报告，采用DenseNet-121主干网络结合注意力对比模块（ACM）和概率类别激活图（PCAM）聚合层，实现多标签疾病分类与弱监督病灶定位。在19,871例由放射科医生验证的内部测试集中，平均AUROC达0.994（平均敏感度92.4%，特异度98.6%），覆盖九种重要临床疾病。在来自泰国13家医院的独立泛化集（5,992例）上，平均AUROC为0.970，显示良好跨机构迁移能力。在4,549例放射科医生标注的病灶定位评估中，模型达到77.9%的平均病灶定位分数（LLF），每张图像误报0.59个非病灶区域。五名胸腔放射科医生的可用性评估显示，系统分类一致性达93.6%，定位一致性达94.7%，平均系统可用性量表（SUS）得分为89。结果表明，基于本地数据构建的具备定位能力的胸片分析系统可实现高精度、强泛化性，并获得临床医生信任。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。