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

Journal: NPJ breast cancer
Published Date:

Abstract

Complete tumour removal is vital in curative breast cancer (BCa) surgery to prevent recurrence. Recently, [F]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.

Authors

  • Luna Maris
    Ghent University, Department of Electronics and Information Systems, MEDISIP, Ghent, Belgium. luna.maris@ugent.be.
  • Menekse Göker
    Ghent University Hospital, Department of Gynaecology, Ghent, Belgium.
  • Kathia De Man
    Department of Nuclear Medicine, Ghent University Hospital, Ghent, Belgium.
  • Bliede Van den Broeck
    Ghent University Hospital, Department of Medical Imaging, Nuclear Medicine, Ghent, Belgium.
  • Sofie Van Hoecke
  • Koen Van de Vijver
    Department of Pathology, UZ Gent, Gent, Belgium.
  • Christian Vanhove
    Medical Image and Signal Processing (MEDISIP), Department of Electronics and Information Systems, Faculty of Engineering and Architecture, Ghent University, Ghent, Belgium.
  • Vincent Keereman
    Ghent University, Department of Electronics and Information Systems, MEDISIP, Ghent, Belgium.

Keywords

No keywords available for this article.