Related Experiment Video
Updated: May 24, 2025
04:48
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
2.6K
Automated Tumor Segmentation in Breast-Conserving Surgery Using Deep Learning on Breast Tomosynthesis
Wen-Pei Wu1, Yu-Wen Chen2, Hwa-Koon Wu1
1Department of Medical Imaging, Changhua Christian Hospital, Changhua, Taiwan.
Journal of Imaging Informatics in Medicine
|March 3, 2025
Summary
This study introduces an AI model for precise breast cancer tumor segmentation during surgery. The deep learning system enhances intraoperative margin assessment, potentially improving surgical outcomes for breast cancer patients.
Area of Science:
- Oncology
- Medical Imaging
- Artificial Intelligence
Background:
- Breast cancer is a leading cause of death in women globally.
- Breast-conserving surgery (BCS) with radiation therapy is standard for early-stage disease.
- Accurate tumor margin delineation is crucial for effective BCS and reducing recurrence.
Purpose of the Study:
- To improve intraoperative tumor segmentation during BCS using digital breast tomosynthesis (DBT).
- To develop and evaluate a deep learning model for precise delineation of tumor margins.
Main Methods:
- Utilized an improved U-Net deep learning architecture with a convolutional block attention module (CBAM).
- Applied the model to segment tumor margins on DBT images from 51 patient cases.
- Compared automated segmentation results with manual contours and pathological assessments.
Main Results:
- The AI model demonstrated high accuracy in tumor segmentation.
- Achieved an Intersection over Union (IoU) of 0.866 and a Dice coefficient of 0.928.
- The system shows potential for enhanced intraoperative margin assessment.
Conclusions:
- The developed deep learning model effectively enhances intraoperative tumor segmentation during BCS.
- This technology has the potential to improve surgical precision and patient outcomes in breast cancer treatment.
- Accurate margin assessment via AI can aid surgeons in achieving complete tumor removal.
