Zhang Renzhi, Wei Wei, Li Rang, Li Jing, Zhou Zhuhuang, Ma Menghang, Zhao Rui, Zhao Xinming
Department of Diagnostic Radiology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China.
School of Electronics and Information, Xi'an Polytechnic University, Xi'an, China.
Front Oncol. 2022 Jan 28;11:733260. doi: 10.3389/fonc.2021.733260. eCollection 2021.
The probability of Breast Imaging Reporting and Data Systems (BI-RADS) 4 lesions being malignant is 2%-95%, which shows the difficulty to make a diagnosis. Radiomics models based on magnetic resonance imaging (MRI) can replace clinicopathological diagnosis with high performance. In the present study, we developed and tested a radiomics model based on MRI images that can predict the malignancy of BI-RADS 4 breast lesions.
We retrospective enrolled a total of 216 BI-RADS 4 patients MRI and clinical information. We extracted 3,474 radiomics features from dynamic contrast-enhanced (DCE), T-weighted images (TWI), and diffusion-weighted imaging (DWI) MRI images. Least absolute shrinkage and selection operator (LASSO) and logistic regression were used to select features and build radiomics models based on different sequence combinations. We built eight radiomics models which were based on DCE, DWI, TWI, DCE+DWI, DCE+TWI, DWI+TWI, and DCE+DWI+TWI and a clinical predictive model built based on the visual assessment of radiologists. A nomogram was constructed with the best radiomics signature combined with patient characteristics. The calibration curves for the radiomics signature and nomogram were conducted, combined with the Hosmer-Lemeshow test.
Pearson's correlation was used to eliminate 3,329 irrelevant features, and then LASSO and logistic regression were used to screen the remaining feature coefficients for each model we built. Finally, 12 related features were obtained in the model which had the best performance. These 12 features were used to build a radiomics model in combination with the actual clinical diagnosis of benign or malignant lesion labels we have obtained. The best model built by 12 features from the 3 sequences has an AUC value of 0.939 (95% CI, 0.884-0.994) and an accuracy of 0.931 in the testing cohort. The sensitivity, specificity, precision and Matthews correlation coefficient (MCC) of testing cohort are 0.932, 0.923, 0.982, and 0.791, respectively. The nomogram has also been verified to have calibration curves with good overlap.
Radiomics is beneficial in the malignancy prediction of BI-RADS 4 breast lesions. The radiomics predictive model built by the combination of DCE, DWI, and TWI sequences has great application potential.
乳腺影像报告和数据系统（BI-RADS）4类病变的恶性概率为2% - 95%，这表明诊断存在困难。基于磁共振成像（MRI）的放射组学模型能够以高性能替代临床病理诊断。在本研究中，我们开发并测试了一种基于MRI图像的放射组学模型，该模型可预测BI-RADS 4类乳腺病变的恶性程度。
我们回顾性纳入了总共216例BI-RADS 4类患者的MRI和临床信息。我们从动态对比增强（DCE）、T加权图像（TWI）和扩散加权成像（DWI）的MRI图像中提取了3474个放射组学特征。使用最小绝对收缩和选择算子（LASSO）及逻辑回归来选择特征，并基于不同的序列组合构建放射组学模型。我们构建了八个基于DCE、DWI、TWI、DCE + DWI、DCE + TWI、DWI + TWI以及DCE + DWI + TWI的放射组学模型，以及一个基于放射科医生视觉评估构建的临床预测模型。使用最佳的放射组学特征结合患者特征构建列线图。对放射组学特征和列线图进行校准曲线分析，并结合Hosmer-Lemeshow检验。
使用Pearson相关性分析消除了3329个无关特征，然后使用LASSO和逻辑回归对我们构建的每个模型的剩余特征系数进行筛选。最终，在性能最佳的模型中获得了12个相关特征。利用这12个特征结合我们已获得的良性或恶性病变标签的实际临床诊断构建了一个放射组学模型。由这3个序列中的12个特征构建的最佳模型在测试队列中的AUC值为0.939（95%CI，0.884 - 0.994），准确率为0.931。测试队列的敏感性、特异性、阳性预测值和马修斯相关系数（MCC）分别为0.932、0.923、0.982和0.791。列线图也已被验证具有良好重叠的校准曲线。
放射组学有助于BI-RADS 4类乳腺病变的恶性预测。由DCE、DWI和TWI序列组合构建的放射组学预测模型具有巨大的应用潜力。
