Meeting Banner
Abstract #0111

Machine Learning-Based Automating Breast Cancer Detection and Classification using DWI

Mami Iima1,2, Ryosuke Mizuno3, Masako Kataoka2, Akihiko Minami2,4, Maya Honda2,5, Keiho Imanishi6, Yunhao Zhang7, Hiroko Satake7, Rintaro Ito8, Shinji Naganawa7, and Yuji Nakamoto2
1Department of Fundamental Development for Advanced Low Invasive Diagnostic Imaging, Nagoya University Graduate School of Medicine, Nagoya, Japan, 2Diagnostic Imaging and Nuclear Medicine, Kyoto University Graduate School of Medicine, Kyoto, Japan, 3A.I. System Research Co. Ltd, Kyoto, Japan, 4Kyoto City Hospital, Kyoto, Japan, 5Kansai Electricity Hospital, Osaka, Japan, 6e-Growth Co., Ltd., Kyoto, Japan, 7Department of Radiology, Nagoya University Graduate School of Medicine, Nagoya, Japan, 8Department of Innovative BioMedical Visualization, Nagoya University Graduate School of Medicine, Nagoya, Japan

Synopsis

Keywords: Breast, Breast

Motivation: Breast MRI's subotimal specificity and labor-intensive DWI interpretation call for automated diagnostic solutions.

Goal(s): Investigate effectiveness of AI integration for breast tumor detection and characterization in DWI.

Approach: Retrospective analysis of 601 patients. Fine-tuned YOLO v5 for tumor identification and 2D CNN for malignancy assessment, utilizing whole-slice and bounding box techniques.

Results: Strong diagnostic performance observed. Whole-slice method: AUC 0.90, sensitivity 84.4%, specificity 84.0%. Bounding box method: AUC 0.87, sensitivity 87.5%, specificity 80.0%.

Impact: Potential to boost screening efficiency, minimize false positives, and improve patient care via more precise, swift diagnoses.

How to access this content:

For one year after publication, abstracts and videos are only open to registrants of this annual meeting. Registrants should use their existing login information. Non-registrant access can be purchased via the ISMRM E-Library.

After one year, current ISMRM & ISMRT members get free access to both the abstracts and videos. Non-members and non-registrants must purchase access via the ISMRM E-Library.

After two years, the meeting proceedings (abstracts) are opened to the public and require no login information. Videos remain behind password for access by members, registrants and E-Library customers.

Click here for more information on becoming a member.

Keywords