Abstract：
OBJECTIVE: The purpose of our study is to present a method combining radiomics with deep learning and clinical data for improved differential diagnosis of sclerosing adenosis (SA)and breast cancer (BC).
METHODS: A total of 97 patients with SA and 100 patients with BC were included in this study. The best model for classification was selected from among four different convolutional neural network (CNN) models, including Vgg16, Resnet18, Resnet50, and Desenet121. The intra-/inter-class correlation coefficient and least absolute shrinkage and selection operator method were used for radiomics feature selection. The clinical features selected were patient age and nodule size. The overall accuracy, sensitivity, specificity, Youden index, positive predictive value, negative predictive value, and area under curve (AUC) value were calculated for comparison of diagnostic efficacy.
RESULTS: All the CNN models combined with radiomics and clinical data were significantly superior to CNN models only. The Desenet121+radiomics+clinical data model showed the best classification performance with an accuracy of 86.80%, sensitivity of 87.60%, specificity of 86.20% and AUC of 0.915, which was better than that of the CNN model only, which had an accuracy of 85.23%, sensitivity of 85.48%, specificity of 85.02%, and AUC of 0.870. In comparison, the diagnostic accuracy, sensitivity, specificity, and AUC value for breast radiologists were 72.08%, 100%, 43.30%, and 0.716, respectively.
CONCLUSIONS: A combination of the CNN-radiomics model and clinical data could be a helpful auxiliary diagnostic tool for distinguishing between SA and BC.
METHODS: A total of 97 patients with SA and 100 patients with BC were included in this study. The best model for classification was selected from among four different convolutional neural network (CNN) models, including Vgg16, Resnet18, Resnet50, and Desenet121. The intra-/inter-class correlation coefficient and least absolute shrinkage and selection operator method were used for radiomics feature selection. The clinical features selected were patient age and nodule size. The overall accuracy, sensitivity, specificity, Youden index, positive predictive value, negative predictive value, and area under curve (AUC) value were calculated for comparison of diagnostic efficacy.
RESULTS: All the CNN models combined with radiomics and clinical data were significantly superior to CNN models only. The Desenet121+radiomics+clinical data model showed the best classification performance with an accuracy of 86.80%, sensitivity of 87.60%, specificity of 86.20% and AUC of 0.915, which was better than that of the CNN model only, which had an accuracy of 85.23%, sensitivity of 85.48%, specificity of 85.02%, and AUC of 0.870. In comparison, the diagnostic accuracy, sensitivity, specificity, and AUC value for breast radiologists were 72.08%, 100%, 43.30%, and 0.716, respectively.
CONCLUSIONS: A combination of the CNN-radiomics model and clinical data could be a helpful auxiliary diagnostic tool for distinguishing between SA and BC.
摘要：
目的：我们研究的目的是提出一种将影像组学与深度学习和临床数据相结合的方法，以改善硬化性腺病（SA）和乳腺癌（BC）的鉴别诊断。
方法：本研究共纳入了97例SA患者和100例BC患者。从四种不同的卷积神经网络（CNN）模型中选择了最佳的分类模型，包括Vgg16、Resnet18、Resnet50和Desenet121。类内/类间相关系数和最小绝对收缩和选择算子方法用于影像组学特征选择。选择的临床特征是患者年龄和结节大小。整体精度,灵敏度,特异性，尤登指数,正预测值，负预测值，计算曲线下面积(AUC)值,以比较诊断效能.
结果：所有结合影像组学和临床数据的CNN模型均明显优于仅CNN模型。Desenet121+影像组学+临床数据模型显示出最佳的分类性能，准确率为86.80%，灵敏度为87.60%，特异性为86.20%，AUC为0.915，优于仅CNN模型，准确率为85.23%，灵敏度为85.48%，特异性为85.02%，AUC为0.870。相比之下,诊断的准确性，灵敏度,特异性，乳腺放射科医生的AUC值为72.08%，100%,43.30%,和0.716。
结论：CNN-影像组学模型和临床数据的结合可能是区分SA和BC的有用辅助诊断工具。
方法：本研究共纳入了97例SA患者和100例BC患者。从四种不同的卷积神经网络（CNN）模型中选择了最佳的分类模型，包括Vgg16、Resnet18、Resnet50和Desenet121。类内/类间相关系数和最小绝对收缩和选择算子方法用于影像组学特征选择。选择的临床特征是患者年龄和结节大小。整体精度,灵敏度,特异性，尤登指数,正预测值，负预测值，计算曲线下面积(AUC)值,以比较诊断效能.
结果：所有结合影像组学和临床数据的CNN模型均明显优于仅CNN模型。Desenet121+影像组学+临床数据模型显示出最佳的分类性能，准确率为86.80%，灵敏度为87.60%，特异性为86.20%，AUC为0.915，优于仅CNN模型，准确率为85.23%，灵敏度为85.48%，特异性为85.02%，AUC为0.870。相比之下,诊断的准确性，灵敏度,特异性，乳腺放射科医生的AUC值为72.08%，100%,43.30%,和0.716。
结论：CNN-影像组学模型和临床数据的结合可能是区分SA和BC的有用辅助诊断工具。
