对头颈部鳞状细胞癌（HNSCC）的诊断和治疗管理是通过常规的头颈部计算机断层扫描（CT）来确定肿瘤和淋巴结特征的。包膜外扩展（ECE）是HNSCC患者生存结果的强烈预测因素。检测ECE的发生是至关重要的，因为它改变了患者的分期和治疗计划。目前，临床ECE检测依赖于临床医生进行视觉识别和病理确认。然而，在大多数当前基于机器学习的ECE诊断研究中，需要手动对淋巴结区域进行注释。 本文提出了一种基于梯度映射引导解释网络（GMGENet）框架，可以在不需要注释淋巴结区域信息的情况下自动进行ECE识别。该框架采用了梯度加权类激活映射（Grad-CAM）技术，引导深度学习算法关注与ECE密切相关的区域。所提出的框架包括一个特征提取器和一个分类器。在联合训练过程中，特征提取器提取出具有信息量的感兴趣体积（VOIs），无需标记的淋巴结区域信息，而分类器学习将提取的VOIs分类为ECE阳性和阴性。 在评估中，采用交叉验证方法对所提出的方法进行了充分的训练和测试。GMGENet在测试中实现了92.2％的准确度和89.3％的曲线下面积（AUC）。GMGENetV2在测试中实现了90.3％的准确度和91.7％的AUC。将结果与不同的现有模型进行了比较，并通过Grad-CAM技术生成了ECE概率热图，进一步确认和解释了结果与真实组织病理学结果的相关性。 所提出的深度网络能够学习有意义的模式，以识别ECE，而无需提供淋巴结区域轮廓。引入的ECE热图将有助于该模型的临床实施，并向放射科医师展示未知的特征。本研究的结果有望推动可解释人工智能辅助的ECE检测的实施。 © 2023年美国医学物理学协会。

Diagnosis and treatment management for head and neck squamous cell carcinoma (HNSCC) is guided by routine diagnostic head and neck computed tomography (CT) scans to identify tumor and lymph node features. The extracapsular extension (ECE) is a strong predictor of patients' survival outcomes with HNSCC. It is essential to detect the occurrence of ECE as it changes staging and treatment planning for patients. Current clinical ECE detection relies on visual identification and pathologic confirmation conducted by clinicians. However, manual annotation of the lymph node region is a required data preprocessing step in most of the current machine learning-based ECE diagnosis studies.In this paper, we propose a Gradient Mapping Guided Explainable Network (GMGENet) framework to perform ECE identification automatically without requiring annotated lymph node region information.The gradient-weighted class activation mapping (Grad-CAM) technique is applied to guide the deep learning algorithm to focus on the regions that are highly related to ECE. The proposed framework includes an extractor and a classifier. In a joint training process, informative volumes of interest (VOIs) are extracted by the extractor without labeled lymph node region information, and the classifier learns the pattern to classify the extracted VOIs into ECE positive and negative.In evaluation, the proposed methods are well-trained and tested using cross-validation. GMGENet achieved test accuracy and area under the curve (AUC) of 92.2% and 89.3%, respectively. GMGENetV2 achieved 90.3% accuracy and 91.7% AUC in the test. The results were compared with different existing models and further confirmed and explained by generating ECE probability heatmaps via a Grad-CAM technique. The presence or absence of ECE has been analyzed and correlated with ground truth histopathological findings.The proposed deep network can learn meaningful patterns to identify ECE without providing lymph node contours. The introduced ECE heatmaps will contribute to the clinical implementations of the proposed model and reveal unknown features to radiologists. The outcome of this study is expected to promote the implementation of explainable artificial intelligence-assiste ECE detection.© 2023 American Association of Physicists in Medicine.