3-D breast nodule detection on automated breast ultrasound using faster region-based convolutional neural networks and U-Net

• Published in: Scientific reports Vol. 13; no. 1; pp. 22625 - 14
• Main Authors: Oh, Kangrok, Lee, Si Eun, Kim, Eun-Kyung
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 18.12.2023; Nature Publishing Group; Nature Portfolio
• Subjects: 631/67/1347; 692/4028/67; Automation; Breast - diagnostic imaging; Breast cancer; Breast Density; Breast Neoplasms - diagnostic imaging; Cancer screening; Early Detection of Cancer - methods; Female; Humanities and Social Sciences; Humans; Mammography; Mammography - methods; Medical screening; multidisciplinary; Neural networks; Neural Networks, Computer; Nodules; Science; Science (multidisciplinary); Ultrasonic imaging; Ultrasonography, Mammary - methods; Ultrasound
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-023-49794-8

Mammography is currently the most commonly used modality for breast cancer screening. However, its sensitivity is relatively low in women with dense breasts. Dense breast tissues show a relatively high rate of interval cancers and are at high risk for developing breast cancer. As a supplemental screening tool, ultrasonography is a widely adopted imaging modality to standard mammography, especially for dense breasts. Lately, automated breast ultrasound imaging has gained attention due to its advantages over hand-held ultrasound imaging. However, automated breast ultrasound imaging requires considerable time and effort for reading because of the lengthy data. Hence, developing a computer-aided nodule detection system for automated breast ultrasound is invaluable and impactful practically. This study proposes a three-dimensional breast nodule detection system based on a simple two-dimensional deep-learning model exploiting automated breast ultrasound. Additionally, we provide several postprocessing steps to reduce false positives. In our experiments using the in-house automated breast ultrasound datasets, a sensitivity of 93.65 % with 8.6 false positives is achieved on unseen test data at best.