Explainable DCNN Decision Framework for Breast Lesion Classification from Ultrasound Images Based on Cancer Characteristics

• Published in: Bioengineering (Basel) Vol. 11; no. 5; p. 453
• Main Authors: AlZoubi, Alaa, Eskandari, Ali, Yu, Harry, Du, Hongbo
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 01-05-2024
• Subjects: Ablation; Analysis; Artificial neural networks; Breast; Breast cancer; Calcification; Cancer; cancer recognition; Care and treatment; Classification; Datasets; Decision making; Decisions; deep convolutional neural networks; Deep learning; Diagnosis; Image analysis; Image classification; Image processing; Lesions; Machine learning; Magnetic resonance imaging; Medical imaging; Methods; Neural networks; Salience; saliency maps; Transfer learning; Ultrasonic imaging; ultrasonography; Ultrasound; Ultrasound imaging; visual explanations; Visualization; United Kingdom; ultrasonography; breast cancer; cancer characteristics; cancer recognition; deep convolutional neural networks; visual explanations; saliency maps
• ISSN: 2306-5354; 2306-5354
• DOI: 10.3390/bioengineering11050453

In recent years, deep convolutional neural networks (DCNNs) have shown promising performance in medical image analysis, including breast lesion classification in 2D ultrasound (US) images. Despite the outstanding performance of DCNN solutions, explaining their decisions remains an open investigation. Yet, the explainability of DCNN models has become essential for healthcare systems to accept and trust the models. This paper presents a novel framework for explaining DCNN classification decisions of lesions in ultrasound images using the saliency maps linking the DCNN decisions to known cancer characteristics in the medical domain. The proposed framework consists of three main phases. First, DCNN models for classification in ultrasound images are built. Next, selected methods for visualization are applied to obtain saliency maps on the input images of the DCNN models. In the final phase, the visualization outputs and domain-known cancer characteristics are mapped. The paper then demonstrates the use of the framework for breast lesion classification from ultrasound images. We first follow the transfer learning approach and build two DCNN models. We then analyze the visualization outputs of the trained DCNN models using the EGrad-CAM and Ablation-CAM methods. We map the DCNN model decisions of benign and malignant lesions through the visualization outputs to the characteristics such as echogenicity, calcification, shape, and margin. A retrospective dataset of 1298 US images collected from different hospitals is used to evaluate the effectiveness of the framework. The test results show that these characteristics contribute differently to the benign and malignant lesions' decisions. Our study provides the foundation for other researchers to explain the DCNN classification decisions of other cancer types.