Combination of Deep Learning Grad-CAM and Radiomics for Automatic Localization and Diagnosis of Architectural Distortion on DBT

Xiao Chen; Yang Zhang; Jiejie Zhou; Yong Pan; Hanghui Xu; Ying Shen; Guoquan Cao; Min-Ying Su; Meihao Wang

doi:10.1016/j.acra.2024.10.031

Combination of Deep Learning Grad-CAM and Radiomics for Automatic Localization and Diagnosis of Architectural Distortion on DBT

Acad Radiol. 2024 Nov 3:S1076-6332(24)00792-X. doi: 10.1016/j.acra.2024.10.031. Online ahead of print.

Authors

Xiao Chen¹, Yang Zhang², Jiejie Zhou³, Yong Pan¹, Hanghui Xu⁴, Ying Shen¹, Guoquan Cao¹, Min-Ying Su⁵, Meihao Wang⁶

Affiliations

¹ Department of Radiology, First Affiliated Hospital of Wenzhou Medical University, Wenzhou, China (X.C., J.Z., Y.P., Y.S., G.C., M.W.).
² Department of Radiological Sciences, University of California, Irvine, CA (Y.Z., J.Z., M-Y.S.); Department of Radiation Oncology, University of California, Irvine, CA (Y.Z.).
³ Department of Radiology, First Affiliated Hospital of Wenzhou Medical University, Wenzhou, China (X.C., J.Z., Y.P., Y.S., G.C., M.W.); Department of Radiological Sciences, University of California, Irvine, CA (Y.Z., J.Z., M-Y.S.).
⁴ Zhuji People's Hospital of Zhejiang Province, China (H.X.).
⁵ Department of Radiological Sciences, University of California, Irvine, CA (Y.Z., J.Z., M-Y.S.); Department of Medical Imaging and Radiological Sciences, Kaohsiung Medical University, Kaohsiung, Taiwan (M-Y.S.). Electronic address: [email protected].
⁶ Department of Radiology, First Affiliated Hospital of Wenzhou Medical University, Wenzhou, China (X.C., J.Z., Y.P., Y.S., G.C., M.W.); Key Laboratory of Intelligent Medical Imaging of Wenzhou, Key Laboratory of Alzheimer's Disease of Zhejiang Province, Institute of Aging, Wenzhou Medical University, Wenzhou, China (M.W.).

PMID: 39496537
DOI: 10.1016/j.acra.2024.10.031

Abstract

Rationale and objectives: Detection and diagnosis of architectural distortion (AD) on digital breast tomosynthesis (DBT) is challenging. This study applied artificial intelligence (AI) using deep learning (DL) algorithms to detect AD, followed by radiomics for classification.

Materials and methods: 500 cases with AD on DBT reports were identified; the earlier 292 cases for training, and the later 208 cases for testing. The DL Gradient-weighted Class Activation Mapping (Grad-CAM) was applied to automatically localize abnormalities and generate a region of interest (ROI), which was put into the radiomics model to estimate the malignancy probability for constructing ROC curves. Radiologists delineated ROI manually for comparison. Cases were categorized into pure AD and AD associated with other features, including mass, regional high-density, and calcifications. The ROC curves were compared using the DeLong test.

Results: The overall malignancy rate was 57% (285/500). Of them, 267 cases were classified as pure AD, and the malignancy rate (106/267 = 39.7%) was significantly lower compared to AD cases associated with other features (179/233 = 76.8%, p < 0.01). In the testing set, the diagnostic AUC was 0.82 when using the manual ROI and 0.84 when using the DL-generated ROI. In the more challenging pure AD cases, DL-generated ROI yielded an AUC of 0.77, significantly lower than 0.86 for AD associated with other features.

Conclusion: DL could detect AD on DBT, and the diagnostic performance was comparable to manual ROI. The strategy worked for pure AD, but the performance was worse than that for AD with other features.

Keywords: Architectural distortion; Breast cancer; Deep learning; Digital breast tomosynthesis; Radiomics.