Gupta Vibha, Petursson Petur, Rawshani Aidin, Boren Jan, Ramunddal Truls, Bhatt Deepak L, Omerovic Elmir, Angerås Oskar, Smith Gustav, Sattar Naveed, Andersson Erik, Redfors Björn, Hilgendorf Lukas, Bergström Göran, Pirazzi Carlo, Skoglund Kristofer, Rawshani Araz
Department of Molecular and Clinical Medicine, University of Gothenburg Institute of Medicine, Gothenburg, Sweden
Department of Cardiology, Sahlgrenska universitetssjukhuset, Goteborg, Sweden.
Open Heart. 2025 Jan 11;12(1):e002998. doi: 10.1136/openhrt-2024-002998.
We examined whether end-to-end deep-learning models could detect moderate (≥50%) or severe (≥70%) stenosis in the left anterior descending artery (LAD), right coronary artery (RCA) or left circumflex artery (LCX) in iodine contrast-enhanced ECG-gated coronary CT angiography (CCTA) scans.
From a database of 6293 CCTA scans, we used pre-existing curved multiplanar reformations (CMR) images of the LAD, RCA and LCX arteries to create end-to-end deep-learning models for the detection of moderate or severe stenoses. We preprocessed the images by exploiting domain knowledge and employed a transfer learning approach using EfficientNet, ResNet, DenseNet and Inception-ResNet, with a class-weighted strategy optimised through cross-validation. Heatmaps were generated to indicate critical areas identified by the models, aiding clinicians in understanding the model's decision-making process.
Among the 900 CMR cases, 279 involved the LAD artery, 259 the RCA artery and 253 the LCX artery. EfficientNet models outperformed others, with EfficientNetB3 and EfficientNetB0 demonstrating the highest accuracy for LAD, EfficientNetB2 for RCA and EfficientNetB0 for LCX. The area under the curve for receiver operating characteristic (AUROC) reached 0.95 for moderate and 0.94 for severe stenosis in the LAD. For the RCA, the AUROC was 0.92 for both moderate and severe stenosis detection. The LCX achieved an AUROC of 0.88 for the detection of moderate stenoses, though the calibration curve exhibited significant overestimation. Calibration curves matched probabilities for the LAD but showed discrepancies for the RCA. Heatmap visualisations confirmed the models' precision in delineating stenotic lesions. Decision curve analysis and net reclassification index assessments reinforced the efficacy of EfficientNet models, confirming their superior diagnostic capabilities.
Our end-to-end deep-learning model demonstrates, for the LAD artery, excellent discriminatory ability and calibration during internal validation, despite a small dataset used to train the network. The model reliably produces precise, highly interpretable images.
我们研究了端到端深度学习模型能否在碘对比剂增强的心电图门控冠状动脉CT血管造影（CCTA）扫描中检测左前降支（LAD）、右冠状动脉（RCA）或左旋支（LCX）中的中度（≥50%）或重度（≥70%）狭窄。
从6293例CCTA扫描数据库中，我们使用LAD、RCA和LCX动脉预先存在的曲面多平面重建（CMR）图像来创建用于检测中度或重度狭窄的端到端深度学习模型。我们通过利用领域知识对图像进行预处理，并采用迁移学习方法，使用EfficientNet、ResNet、DenseNet和Inception-ResNet，并通过交叉验证优化了类加权策略。生成热图以指示模型识别的关键区域，帮助临床医生理解模型的决策过程。
在900例CMR病例中，279例涉及LAD动脉，259例涉及RCA动脉，253例涉及LCX动脉。EfficientNet模型表现优于其他模型，EfficientNetB3和EfficientNetB0在LAD中显示出最高准确率，EfficientNetB2在RCA中最高，EfficientNetB0在LCX中最高。LAD中中度狭窄的受试者操作特征曲线下面积（AUROC）达到0.95，重度狭窄为0.94。对于RCA，中度和重度狭窄检测的AUROC均为0.92。LCX检测中度狭窄的AUROC为0.88，尽管校准曲线显示有显著高估。校准曲线与LAD的概率匹配，但与RCA存在差异。热图可视化证实了模型在描绘狭窄病变方面的准确性。决策曲线分析和净重新分类指数评估强化了EfficientNet模型的有效性，证实了它们卓越的诊断能力。
我们的端到端深度学习模型在内部验证中，对于LAD动脉显示出优异的鉴别能力和校准能力，尽管用于训练网络的数据集较小。该模型可靠地生成精确、高度可解释的图像。
