Supporting intraoperative margin assessment using deep learning for automatic tumour segmentation in breast lumpectomy micro-PET-CT.
Authors
Affiliations (10)
Affiliations (10)
- Ghent University, Department of Electronics and Information Systems, MEDISIP, Ghent, Belgium. [email protected].
- XEOS Medical, R&D Department, Ghent, Belgium. [email protected].
- Ghent University Hospital, Department of Gynaecology, Ghent, Belgium.
- Ghent University Hospital, Department of Medical Imaging, Nuclear Medicine, Ghent, Belgium.
- Ghent University-Imec, Internet Technology and Data Science Lab (IDLab), Ghent, Belgium.
- Ghent University Hospital, Department of Pathology, Ghent, Belgium.
- Ghent University, Department of Diagnostic Sciences and CRIG, Ghent, Belgium.
- Ghent University, Department of Electronics and Information Systems, MEDISIP, Ghent, Belgium.
- Ghent University, CORE ARTH INFINITY, Ghent, Belgium.
- XEOS Medical, R&D Department, Ghent, Belgium.
Abstract
Complete tumour removal is vital in curative breast cancer (BCa) surgery to prevent recurrence. Recently, [<sup>18</sup>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.