From the Department of Radiology (F.B., L.H., S.M.N.), Department of Anesthesiology, Division of Operative Intensive Care Medicine (F.B.), Department of Cardiology (P.S.), and Department of Rheumatology (K.B.B., D.P., A.Z.), Charité-Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin and Humboldt Universität zu Berlin, 12203 Berlin, Germany; Department of Radiology and Nuclear Medicine, German Heart Center, Technical University of Munich, Munich, Germany (K.K.B.); Department of Diagnostic and Interventional Radiology, Technical University of Munich, Munich, Germany (F.B., K.K.B., M.R.M., L.C.A.); Artificial Intelligence in Medicine (AIM) Program, Mass General Brigham, Harvard Medical School, Boston, Mass (H.J.W.L.A.); Departments of Radiation Oncology and Radiology, Dana-Farber Cancer Institute and Brigham and Women's Hospital, Boston, Mass (H.J.W.L.A.); and Department of Radiology and Nuclear Medicine, CARIM & GROW, Maastricht University, Maastricht, the Netherlands (H.J.W.L.A.).
Radiol Artif Intell. 2024 Sep;6(5):e230502. doi: 10.1148/ryai.230502.
Purpose To develop and evaluate a publicly available deep learning model for segmenting and classifying cardiac implantable electronic devices (CIEDs) on Digital Imaging and Communications in Medicine (DICOM) and smartphone-based chest radiographs. Materials and Methods This institutional review board-approved retrospective study included patients with implantable pacemakers, cardioverter defibrillators, cardiac resynchronization therapy devices, and cardiac monitors who underwent chest radiography between January 2012 and January 2022. A U-Net model with a ResNet-50 backbone was created to classify CIEDs on DICOM and smartphone images. Using 2321 chest radiographs in 897 patients (median age, 76 years [range, 18-96 years]; 625 male, 272 female), CIEDs were categorized into four manufacturers, 27 models, and one "other" category. Five smartphones were used to acquire 11 072 images. Performance was reported using the Dice coefficient on the validation set for segmentation or balanced accuracy on the test set for manufacturer and model classification, respectively. Results The segmentation tool achieved a mean Dice coefficient of 0.936 (IQR: 0.890-0.958). The model had an accuracy of 94.36% (95% CI: 90.93%, 96.84%; 251 of 266) for CIED manufacturer classification and 84.21% (95% CI: 79.31%, 88.30%; 224 of 266) for CIED model classification. Conclusion The proposed deep learning model, trained on both traditional DICOM and smartphone images, showed high accuracy for segmentation and classification of CIEDs on chest radiographs. Conventional Radiography, Segmentation . © RSNA, 2024 See also the commentary by Júdice de Mattos Farina and Celi in this issue.
目的 开发并评估一种可公开获取的深度学习模型，用于对数字成像和通信在医学（DICOM）以及基于智能手机的胸部 X 光片上的心脏植入式电子设备（CIED）进行分割和分类。
材料与方法 本研究经机构审查委员会批准，回顾性纳入了 2012 年 1 月至 2022 年 1 月期间接受胸部 X 光检查的植入式起搏器、除颤器、心脏再同步治疗设备和心脏监测器的患者。创建了一个具有 ResNet-50 骨干的 U-Net 模型，用于对 DICOM 和智能手机图像上的 CIED 进行分类。该研究纳入了 897 名患者的 2321 张胸部 X 光片（中位年龄为 76 岁[范围，18-96 岁]；625 名男性，272 名女性），将 CIED 分为四个制造商、27 个型号和一个“其他”类别。使用五部智能手机获取了 11072 张图像。使用验证集上的 Dice 系数报告分割性能，或使用测试集上的制造商和模型分类的平衡准确率报告性能。
结果 分割工具的平均 Dice 系数为 0.936（IQR：0.890-0.958）。该模型对 CIED 制造商分类的准确率为 94.36%（95%CI：90.93%，96.84%；266 个中的 251 个），对 CIED 型号分类的准确率为 84.21%（95%CI：79.31%，88.30%；266 个中的 224 个）。
结论 该深度学习模型在传统 DICOM 和智能手机图像上进行训练，对胸部 X 光片上的 CIED 分割和分类具有较高的准确性。
