Examining the effectiveness of a deep learning-based computer-aided breast cancer detection system for breast ultrasound

Tomoyuki Fujioka; Kazunori Kubota; Jen Feng Hsu; Ruey Feng Chang; Terumasa Sawada; Yoshimi Ide; Kanae Taruno; Meishi Hankyo; Tomoko Kurita; Seigo Nakamura; Ukihide Tateishi; Hiroyuki Takei

doi:10.1007/s10396-023-01332-9

Examining the effectiveness of a deep learning-based computer-aided breast cancer detection system for breast ultrasound

J Med Ultrason (2001). 2023 Oct;50(4):511-520. doi: 10.1007/s10396-023-01332-9. Epub 2023 Jul 4.

Authors

Tomoyuki Fujioka¹, Kazunori Kubota^{2

3}, Jen Feng Hsu⁴, Ruey Feng Chang⁴, Terumasa Sawada^{5

6}, Yoshimi Ide^{6

7}, Kanae Taruno⁶, Meishi Hankyo⁸, Tomoko Kurita⁸, Seigo Nakamura⁶, Ukihide Tateishi¹, Hiroyuki Takei⁸

Affiliations

¹ Department of Diagnostic Radiology, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo, 113-8510, Japan.
² Department of Diagnostic Radiology, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo, 113-8510, Japan. kubotard@dokkyomed.ac.jp.
³ Department of Radiology, Dokkyo Medical University Saitama Medical Center, 2-1-50 Minami-Koshigaya, Koshigaya, Saitama, 343-8555, Japan. kubotard@dokkyomed.ac.jp.
⁴ Department of Computer Science and Information Engineering, National Taiwan University, No. 1, Sec. 4, Roosevelt Rd, Taipei, 10617, Taiwan, ROC.
⁵ Department of Breast Surgery, NTT Medical Center Tokyo, 5-9-22 Higashi-Gotanda, Shinagawa-ku, Tokyo, 141-8625, Japan.
⁶ Department of Breast Surgical Oncology, Department of Surgery, Showa University School of Medicine, 1-5-8 Hatanodai, Shinagawa-ku, Tokyo, 142-8666, Japan.
⁷ Department of Breast Oncology, Kikuna Memorial Hospital, 4-4-27 Kikuna, Kohoku-ku, Yokohama, 222-0011, Japan.
⁸ Department of Breast Surgical Oncology, Nippon Medical School, 1-1-5 Sendagi, Bunkyo-ku, Tokyo, 113-8602, Japan.

Abstract

Purpose: This study aimed to evaluate the clinical usefulness of a deep learning-based computer-aided detection (CADe) system for breast ultrasound.

Methods: The set of 88 training images was expanded to 14,000 positive images and 50,000 negative images. The CADe system was trained to detect lesions in real- time using deep learning with an improved model of YOLOv3-tiny. Eighteen readers evaluated 52 test image sets with and without CADe. Jackknife alternative free-response receiver operating characteristic analysis was used to estimate the effectiveness of this system in improving lesion detection.

Result: The area under the curve (AUC) for image sets was 0.7726 with CADe and 0.6304 without CADe, with a 0.1422 difference, indicating that with CADe was significantly higher than that without CADe (p < 0.0001). The sensitivity per case was higher with CADe (95.4%) than without CADe (83.7%). The specificity of suspected breast cancer cases with CADe (86.6%) was higher than that without CADe (65.7%). The number of false positives per case (FPC) was lower with CADe (0.22) than without CADe (0.43).

Conclusion: The use of a deep learning-based CADe system for breast ultrasound by readers significantly improved their reading ability. This system is expected to contribute to highly accurate breast cancer screening and diagnosis.

Keywords: Artificial intelligence; Breast cancer; Computer-aided detection; Deep learning; Ultrasound.

MeSH terms

Breast Neoplasms* / diagnostic imaging
Computers
Deep Learning*
Female
Humans
ROC Curve