BD-StableNet: a deep stable learning model with an automatic lesion area detection function for predicting malignancy in BI-RADS category 3-4A lesions

Hui Qu; Guanglei Chen; Tong Li; Mingchen Zou; Jiaxi Liu; Canwei Dong; Ye Tian; Caigang Liu; Xiaoyu Cui

doi:10.1088/1361-6560/ad953e

BD-StableNet: a deep stable learning model with an automatic lesion area detection function for predicting malignancy in BI-RADS category 3-4A lesions

Phys Med Biol. 2024 Nov 20. doi: 10.1088/1361-6560/ad953e. Online ahead of print.

Authors

Hui Qu¹, Guanglei Chen², Tong Li³, Mingchen Zou³, Jiaxi Liu³, Canwei Dong³, Ye Tian³, Caigang Liu⁴, Xiaoyu Cui⁵

Affiliations

¹ Northeastern University, College of Medicine and Biological Information Engineering, Shenyang, 110169, CHINA.
² Department of Oncology, Shengjing Hospital of China Medical University, No.36 Sanhao Street, Heping District, Shengjing Hospital of China Medical University, Shenyang, Liaoning, China, 110004, Shenyang, Liaoning, 110004, CHINA.
³ College of Medicine and Biological Information Engineering, Northeastern University, NO. 3-11, Wenhua Road, Heping District, Northeastern University, Shenyang, Liaoning, China, 110819, Shenyang, Liaoning, 110819, CHINA.
⁴ Department of Oncology, Innovative Cancer Drug Research and Engineering Center of Liaoning Province, Cancer Stem Cell and Translation Medicine Lab, Shengjing Hospital of China Medical University, Shenyang, Liaoning, China., Shengjing Hospital of China Medical University, No.36 Sanhao Street, Heping District, Shengjing Hospital of China Medical University, Shenyang, Liaoning, China, 110004, Shenyang, Liaoning, 110004, CHINA.
⁵ the College of Medicine and Biological Information Engineering, Northeastern University, NO. 3-11, Wenhua Road, Heping District, Northeastern University, Shenyang, Liaoning, China, 110819, Shenyang, 110819, CHINA.

PMID: 39569908
DOI: 10.1088/1361-6560/ad953e

Abstract

The latest developments combining deep learning technology and medical image data have attracted wide attention and provide efficient noninvasive methods for the early diagnosis of breast cancer. The success of this task often depends on a large amount of data annotated by medical experts, which is time-consuming and may not always be feasible in the biomedical field. The lack of interpretability has greatly hindered the application of deep learning in the medical field. Currently, deep stable learning, including causal inference, make deep learning models more predictive and interpretable. In this study, to distinguish malignant tumors in Breast Imaging-Reporting and Data System (BI-RADS) category 3-4A breast lesions, we propose BD-StableNet, a deep stable learning model for the automatic detection of lesion areas. In this retrospective study, we collected 3103 breast ultrasound images (1418 benign and 1685 malignant lesions) from 493 patients (361 benign and 132 malignant lesion patients) for model training and testing. Compared with other mainstream deep learning models, BD-StableNet has better prediction performance (accuracy = 0.952, area under the curve (AUC) = 0.982, precision = 0.970, recall = 0.941, F1-score = 0.955 and specificity = 0.965). The lesion area prediction and class activation map (CAM) results both verify that our proposed model is highly interpretable. The results indicate that BD-StableNet significantly enhances diagnostic accuracy and interpretability, offering a promising noninvasive approach for the diagnosis of BI-RADS category 3-4A breast lesions. Clinically, the use of BD-StableNet could reduce unnecessary biopsies, improve diagnostic efficiency, and ultimately enhance patient outcomes by providing more precise and reliable assessments of breast lesions.

Keywords: BI-RADS; breast cancer; deep stable learning; diagnosis; ultrasound.