Detection of Breast Cancer Lumpectomy Margin with SAM-incorporated Forward-Forward Contrastive Learning
This addresses the critical need to reduce re-excision rates in breast cancer surgery, though it appears to be an incremental improvement over existing deep learning methods.
The paper tackled the problem of inaccurate intraoperative margin assessment in breast cancer lumpectomy by proposing a deep learning framework combining Segment Anything Model with Forward-Forward Contrastive Learning, achieving an AUC of 0.8455 for margin classification and a 27.4% improvement in Dice similarity for segmentation while reducing inference time to 47 milliseconds per image.
Complete removal of cancer tumors with a negative specimen margin during lumpectomy is essential in reducing breast cancer recurrence. However, 2D specimen radiography (SR), the current method used to assess intraoperative specimen margin status, has limited accuracy, resulting in nearly a quarter of patients requiring additional surgery. To address this, we propose a novel deep learning framework combining the Segment Anything Model (SAM) with Forward-Forward Contrastive Learning (FFCL), a pre-training strategy leveraging both local and global contrastive learning for patch-level classification of SR images. After annotating SR images with regions of known maligancy, non-malignant tissue, and pathology-confirmed margins, we pre-train a ResNet-18 backbone with FFCL to classify margin status, then reconstruct coarse binary masks to prompt SAM for refined tumor margin segmentation. Our approach achieved an AUC of 0.8455 for margin classification and segmented margins with a 27.4% improvement in Dice similarity over baseline models, while reducing inference time to 47 milliseconds per image. These results demonstrate that FFCL-SAM significantly enhances both the speed and accuracy of intraoperative margin assessment, with strong potential to reduce re-excision rates and improve surgical outcomes in breast cancer treatment. Our code is available at https://github.com/tbwa233/FFCL-SAM/.