■本研究旨在开发一种使用多模态成像的深度学习影像组学模型,以区分良性和恶性乳腺肿瘤。
■多模态成像数据,包括超声检查(美国),乳房X线照相术(MG),磁共振成像(MRI),在2018年12月至2023年5月期间,回顾性收集了322例经组织病理学证实的乳腺肿瘤患者(112例乳腺良性肿瘤和210例乳腺恶性肿瘤).基于多模态成像,实验分为三个部分:传统的影像组学,深度学习影像组学,和特征融合。我们测试了七个分类器的性能,即,SVM,KNN,随机森林,额外的树木,XGBoost,LightGBM,和LR,在不同的特征模型上。通过使用集成和堆叠策略的特征融合,我们获得了良性和恶性乳腺肿瘤的最佳分类模型。
■就传统的影像组学而言,集成融合策略达到了最高的精度,AUC,和特异性,值为0.892、0.942[0.886-0.996],和0.956[0.873-1.000],分别。与美国的早期融合战略,MG,MRI达到最高灵敏度0.952[0.887-1.000]。就深度学习影像组学而言,堆叠融合策略达到了最高的精度,AUC,和灵敏度,值为0.937、0.947[0.887-1.000],和1.000[0.999-1.000],分别。US+MRI和US+MG的早期融合策略达到0.954[0.867-1.000]的最高特异性。在特征融合方面,后期融合策略的集合和堆叠方法达到了0.968的最高精度。此外,堆叠实现了最高的AUC和特异性,分别为0.997[0.990-1.000]和1.000[0.999-1.000],分别。在早期融合策略下,USMGMR的传统影像和深度特征达到了1.000[0.999-1.000]的最高灵敏度。
■这项研究证明了将深度学习和影像组学特征与多模态图像集成的潜力。作为一种单一的模态,基于影像组学特征的MRI比US或MG具有更高的准确性。与单模或放射学模型相比,US和MG模型通过迁移学习实现了更高的准确性。在早期融合策略下,US+MG+MR的传统影像和深度特征达到了最高的灵敏度,显示出更高的诊断性能,并为乳腺良恶性肿瘤的鉴别提供了更有价值的信息。
UNASSIGNED: This study aimed to develop a deep learning radiomic model using multimodal imaging to differentiate benign and malignant breast tumours.
UNASSIGNED: Multimodality imaging data, including ultrasonography (US), mammography (MG), and magnetic resonance imaging (MRI), from 322 patients (112 with benign breast tumours and 210 with malignant breast tumours) with histopathologically confirmed breast tumours were retrospectively collected between December 2018 and May 2023. Based on multimodal imaging, the experiment was divided into three parts: traditional radiomics, deep learning radiomics, and feature fusion. We tested the performance of seven classifiers, namely, SVM, KNN, random forest, extra trees, XGBoost, LightGBM, and LR, on different feature models. Through feature fusion using ensemble and stacking strategies, we obtained the optimal classification model for benign and malignant breast tumours.
UNASSIGNED: In terms of traditional radiomics, the ensemble fusion strategy achieved the highest accuracy, AUC, and specificity, with values of 0.892, 0.942 [0.886-0.996], and 0.956 [0.873-1.000], respectively. The early fusion strategy with US, MG, and MRI achieved the highest sensitivity of 0.952 [0.887-1.000]. In terms of deep learning radiomics, the stacking fusion strategy achieved the highest accuracy, AUC, and sensitivity, with values of 0.937, 0.947 [0.887-1.000], and 1.000 [0.999-1.000], respectively. The early fusion strategies of US+MRI and US+MG achieved the highest specificity of 0.954 [0.867-1.000]. In terms of feature fusion, the ensemble and stacking approaches of the late fusion strategy achieved the highest accuracy of 0.968. In addition, stacking achieved the highest AUC and specificity, which were 0.997 [0.990-1.000] and 1.000 [0.999-1.000], respectively. The traditional radiomic and depth features of US+MG + MR achieved the highest sensitivity of 1.000 [0.999-1.000] under the early fusion strategy.
UNASSIGNED: This study demonstrated the potential of integrating deep learning and radiomic features with multimodal images. As a single modality, MRI based on radiomic features achieved greater accuracy than US or MG. The US and MG models achieved higher accuracy with transfer learning than the single-mode or radiomic models. The traditional radiomic and depth features of US+MG + MR achieved the highest sensitivity under the early fusion strategy, showed higher diagnostic performance, and provided more valuable information for differentiation between benign and malignant breast tumours.