Department of Otorhinolaryngology, The First Affiliated Hospital, Nanjing Medical University, Nanjing, 210029, China.
Department of Otolaryngology-Head and Neck Surgery, Renmin Hospital of Wuhan University, 238 Jie-Fang Road, Wuhan, Hubei, 430060, P.R. China.
Cancer Imaging. 2023 Feb 9;23(1):14. doi: 10.1186/s40644-023-00530-5.
The purpose of this study was to explore whether incorporating the peritumoral region to train deep neural networks could improve the performance of the models for predicting the prognosis of NPC.
A total of 381 NPC patients who were divided into high- and low-risk groups according to progression-free survival were retrospectively included. Deeplab v3 and U-Net were trained to build segmentation models for the automatic segmentation of the tumor and suspicious lymph nodes. Five datasets were constructed by expanding 5, 10, 20, 40, and 60 pixels outward from the edge of the automatically segmented region. Inception-Resnet-V2, ECA-ResNet50t, EfficientNet-B3, and EfficientNet-B0 were trained with the original, segmented, and the five new constructed datasets to establish the classification models. The receiver operating characteristic curve was used to evaluate the performance of each model.
The Dice coefficients of Deeplab v3 and U-Net were 0.741(95%CI:0.722-0.760) and 0.737(95%CI:0.720-0.754), respectively. The average areas under the curve (aAUCs) of deep learning models for classification trained with the original and segmented images and with images expanded by 5, 10, 20, 40, and 60 pixels were 0.717 ± 0.043, 0.739 ± 0.016, 0.760 ± 0.010, 0.768 ± 0.018, 0.802 ± 0.013, 0.782 ± 0.039, and 0.753 ± 0.014, respectively. The models trained with the images expanded by 20 pixels obtained the best performance.
The peritumoral region NPC contains information related to prognosis, and the incorporation of this region could improve the performance of deep learning models for prognosis prediction.
本研究旨在探讨在训练深度神经网络时纳入肿瘤周围区域是否可以提高模型预测 NPC 预后的性能。
回顾性纳入 381 例 NPC 患者，根据无进展生存期分为高危组和低危组。训练 Deeplab v3 和 U-Net 以建立肿瘤和可疑淋巴结自动分割的分割模型。从自动分割区域的边缘向外扩展 5、10、20、40 和 60 个像素，构建了五个数据集。使用原始、分割和五个新构建的数据集训练 Inception-Resnet-V2、ECA-ResNet50t、EfficientNet-B3 和 EfficientNet-B0，建立分类模型。使用受试者工作特征曲线评估每个模型的性能。
Deeplab v3 和 U-Net 的 Dice 系数分别为 0.741(95%CI:0.722-0.760)和 0.737(95%CI:0.720-0.754)。用原始和分割图像以及扩展 5、10、20、40 和 60 个像素的图像训练的深度学习分类模型的平均曲线下面积(aAUCs)分别为 0.717±0.043、0.739±0.016、0.760±0.010、0.768±0.018、0.802±0.013、0.782±0.039 和 0.753±0.014。用扩展 20 个像素的图像训练的模型获得了最佳性能。
肿瘤周围区域 NPC 包含与预后相关的信息，纳入该区域可以提高预后预测的深度学习模型的性能。
