Presurgical Upgrade Prediction of DCIS to Invasive Ductal Carcinoma Using Time-dependent Deep Learning Models with DCE MRI.

"Just Accepted" papers have undergone full peer review and have been accepted for publication in Radiology: Artificial Intelligence. This article will undergo copyediting, layout, and proof review before it is published in its final version. Please note that during production of the final copyedited article, errors may be discovered which could affect the content. Purpose To determine whether time-dependent deep learning models can outperform single timepoint models in predicting preoperative upgrade of ductal carcinoma in situ (DCIS) to invasive malignancy on dynamic contrastenhanced (DCE) breast MRI without lesion segmentation prerequisite. Materials and Methods In this exploratory study, 154 cases of biopsy-proven DCIS (25 upgraded at surgery and 129 not upgraded) were selected consecutively from a retrospective cohort of preoperative DCE MRI in women with an average age of 58.6 years at time of diagnosis from 2012 to 2022. Binary classification was implemented with convolutional neural network-long short-term memory (CNN-LSTM) architectures benchmarked against traditional CNNs without manual segmentation of the lesions. Combinatorial performance analysis of ResNet50 versus VGG16-based models was performed with each contrast phase. Binary classification area under the receiver operating characteristic curve (AUC) was reported. Results VGG16-based models consistently provided better hold-out test AUCs than ResNet50 in CNN and CNNLSTM studies (multiphase test AUC: 0.67 versus 0.59, respectively, for CNN models; P = .04 and 0.73 versus 0.62 for CNN-LSTM models; P = .008). The time-dependent model (CNN-LSTM) provided a better multiphase test AUC over single-timepoint (CNN) models (0.73 versus 0.67, P = .04). Conclusion Compared with single-timepoint architectures, sequential deep learning algorithms using preoperative DCE MRI improved prediction of DCIS lesions upgraded to invasive malignancy without the need for lesion segmentation. ©RSNA, 2024.