Distinguishing benign inverted papilloma (IP) tumors from those that have undergone malignant transformation to squamous cell carcinoma (IP-SCC) is important but challenging to do preoperatively. Magnetic resonance imaging (MRI) can help differentiate these two entities, however no established method exists that can automatically synthesize all potentially relevant MRI image features to distinguish IP and IP-SCC. We explored a deep learning approach, using 3-dimensional convolutional neural networks (CNNs), to address this challenge.Retrospective chart reviews were performed at two institutions to create a dataset of preoperative MRIs with corresponding surgical pathology reports. The MRI dataset included all available MRI sequences in the axial plane, which were used to train, validate, and test three CNN models. Saliency maps were generated to visualize areas of MRIs with greatest influence on predictions.A total of 90 patients with IP (n = 64) or IP-SCC (n = 26) tumors were identified, with a total of 446 images of distinct MRI sequences for IP (n = 329) or IP-SCC (n = 117). The best CNN model, All-Net, demonstrated a sensitivity of 66.7%, specificity of 81.5%, overall accuracy of 77.9%, and ROC-AUC of 0.80 ([0.682 - 0.898], 95% confidence interval) for test classification performance. The other two models, Small-All-Net and Elastic-All-Net, showed similar performances.A deep learning approach with 3-dimensional CNNs can distinguish IP and IP-SCC with moderate test classification performance. Although CNNs demonstrate promise to enhance the prediction of IP-SCC using MRIs, more data is needed before they can reach the predictive value already established by human MRI evaluation. This article is protected by copyright. All rights reserved.
View details for DOI 10.1002/alr.22958
View details for PubMedID 34989484