Saved in:
Bibliographic Details
Main Authors: Ward, Tyler, Wang, Xiaoqin, McFarland, Braxton, Ahamed, Md Atik, Nozad, Sahar, Arshad, Talal, Nebbache, Hafsa, Chen, Jin, Imran, Abdullah
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2506.21006
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911023305326592
author Ward, Tyler
Wang, Xiaoqin
McFarland, Braxton
Ahamed, Md Atik
Nozad, Sahar
Arshad, Talal
Nebbache, Hafsa
Chen, Jin
Imran, Abdullah
author_facet Ward, Tyler
Wang, Xiaoqin
McFarland, Braxton
Ahamed, Md Atik
Nozad, Sahar
Arshad, Talal
Nebbache, Hafsa
Chen, Jin
Imran, Abdullah
contents 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/.
format Preprint
id arxiv_https___arxiv_org_abs_2506_21006
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Detection of Breast Cancer Lumpectomy Margin with SAM-incorporated Forward-Forward Contrastive Learning
Ward, Tyler
Wang, Xiaoqin
McFarland, Braxton
Ahamed, Md Atik
Nozad, Sahar
Arshad, Talal
Nebbache, Hafsa
Chen, Jin
Imran, Abdullah
Computer Vision and Pattern Recognition
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/.
title Detection of Breast Cancer Lumpectomy Margin with SAM-incorporated Forward-Forward Contrastive Learning
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.21006