Advanced search
1 file | 2.95 MB Add to list

Supporting intraoperative margin assessment using deep learning for automatic tumour segmentation in breast lumpectomy micro-PET-CT

Luna Maris (UGent) , Menekse GOKER (UGent) , Kathia De Man (UGent) , Bliede Van den Broeck (UGent) , Sofie Van Hoecke (UGent) , Koen Van de Vijver (UGent) , Christian Vanhove (UGent) and Vincent Keereman (UGent)
Author
Organization
Abstract
Complete tumour removal is vital in curative breast cancer (BCa) surgery to prevent recurrence. Recently, [18F]FDG micro-PET-CT of lumpectomy specimens has shown promise for intraoperative margin assessment (IMA). To aid interpretation, we trained a 2D Residual U-Net to delineate invasive carcinoma of no special type in micro-PET-CT lumpectomy images. We collected 53 BCa lamella images from 19 patients with true histopathology-defined tumour segmentations. Group five-fold cross-validation yielded a dice similarity coefficient of 0.71 +/- 0.20 for segmentation. Afterwards, an ensemble model was generated to segment tumours and predict margin status. Comparing predicted and true histopathological margin status in a separate set of 31 micro-PET-CT lumpectomy images of 31 patients achieved an F1 score of 84%, closely matching the mean performance of seven physicians who manually interpreted the same images. This model represents an important step towards a decision-support system that enhances micro-PET-CT-based IMA in BCa, facilitating its clinical adoption.
Keywords
CONSERVING SURGERY, CARCINOMA, SPECIMENS, GUIDELINE, PROTOCOL

Downloads

  • Maris - 2025 - Supporting intraoperative margin assessment using .pdf
    • full text (Published version)
    • |
    • open access
    • |
    • PDF
    • |
    • 2.95 MB

Citation

Please use this url to cite or link to this publication:

MLA
Maris, Luna, et al. “Supporting Intraoperative Margin Assessment Using Deep Learning for Automatic Tumour Segmentation in Breast Lumpectomy Micro-PET-CT.” NPJ BREAST CANCER, vol. 11, no. 1, 2025, doi:10.1038/s41523-025-00797-w.
APA
Maris, L., GOKER, M., De Man, K., Van den Broeck, B., Van Hoecke, S., Van de Vijver, K., … Keereman, V. (2025). Supporting intraoperative margin assessment using deep learning for automatic tumour segmentation in breast lumpectomy micro-PET-CT. NPJ BREAST CANCER, 11(1). https://doi.org/10.1038/s41523-025-00797-w
Chicago author-date
Maris, Luna, Menekse GOKER, Kathia De Man, Bliede Van den Broeck, Sofie Van Hoecke, Koen Van de Vijver, Christian Vanhove, and Vincent Keereman. 2025. “Supporting Intraoperative Margin Assessment Using Deep Learning for Automatic Tumour Segmentation in Breast Lumpectomy Micro-PET-CT.” NPJ BREAST CANCER 11 (1). https://doi.org/10.1038/s41523-025-00797-w.
Chicago author-date (all authors)
Maris, Luna, Menekse GOKER, Kathia De Man, Bliede Van den Broeck, Sofie Van Hoecke, Koen Van de Vijver, Christian Vanhove, and Vincent Keereman. 2025. “Supporting Intraoperative Margin Assessment Using Deep Learning for Automatic Tumour Segmentation in Breast Lumpectomy Micro-PET-CT.” NPJ BREAST CANCER 11 (1). doi:10.1038/s41523-025-00797-w.
Vancouver
1.
Maris L, GOKER M, De Man K, Van den Broeck B, Van Hoecke S, Van de Vijver K, et al. Supporting intraoperative margin assessment using deep learning for automatic tumour segmentation in breast lumpectomy micro-PET-CT. NPJ BREAST CANCER. 2025;11(1).
IEEE
[1]
L. Maris et al., “Supporting intraoperative margin assessment using deep learning for automatic tumour segmentation in breast lumpectomy micro-PET-CT,” NPJ BREAST CANCER, vol. 11, no. 1, 2025.
@article{01K2Y47TEGWE6RD6ZW474ZQ7VJ,
  abstract     = {{Complete tumour removal is vital in curative breast cancer (BCa) surgery to prevent recurrence. Recently, [18F]FDG micro-PET-CT of lumpectomy specimens has shown promise for intraoperative margin assessment (IMA). To aid interpretation, we trained a 2D Residual U-Net to delineate invasive carcinoma of no special type in micro-PET-CT lumpectomy images. We collected 53 BCa lamella images from 19 patients with true histopathology-defined tumour segmentations. Group five-fold cross-validation yielded a dice similarity coefficient of 0.71 +/- 0.20 for segmentation. Afterwards, an ensemble model was generated to segment tumours and predict margin status. Comparing predicted and true histopathological margin status in a separate set of 31 micro-PET-CT lumpectomy images of 31 patients achieved an F1 score of 84%, closely matching the mean performance of seven physicians who manually interpreted the same images. This model represents an important step towards a decision-support system that enhances micro-PET-CT-based IMA in BCa, facilitating its clinical adoption.}},
  articleno    = {{88}},
  author       = {{Maris, Luna and GOKER, Menekse and De Man, Kathia and Van den Broeck, Bliede and Van Hoecke, Sofie and Van de Vijver, Koen and Vanhove, Christian and Keereman, Vincent}},
  issn         = {{2374-4677}},
  journal      = {{NPJ BREAST CANCER}},
  keywords     = {{CONSERVING SURGERY,CARCINOMA,SPECIMENS,GUIDELINE,PROTOCOL}},
  language     = {{eng}},
  number       = {{1}},
  pages        = {{12}},
  title        = {{Supporting intraoperative margin assessment using deep learning for automatic tumour segmentation in breast lumpectomy micro-PET-CT}},
  url          = {{http://doi.org/10.1038/s41523-025-00797-w}},
  volume       = {{11}},
  year         = {{2025}},
}

Altmetric
View in Altmetric
Web of Science
Times cited: