PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.
Esophageal adenocarcinoma (EAC), which can arise from Barrett’s esophagus (BE), has a 5-year survival rate of < 20%. Unfortunately, the severe sampling limitations associated with conventional histology may limit the sensitivity for detecting EAC and dysplasia (a precursor lesion to EAC) through regular endoscopic screening of BE patients. We have developed a non-destructive 3D pathology workflow to provide comprehensive evaluation of whole biopsies and a deep learning-based computational triage method that automatically segments potentially neoplastic regions (dysplasia or EAC) to guide pathologist review. A preliminary clinical validation study shows that our AI-assisted 3D workflow enables neoplasia to be identified with higher sensitivity on a per-biopsy level than conventional slide-based 2D histology.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.
The alert did not successfully save. Please try again later.
Lindsey A. Barner, Gan Gao, Deepti M. Reddi, Lydia Lan, Wynn Burke, William M. Grady, Jonathan T. C. Liu, "Enhanced detection of neoplasia in esophageal biopsies via non-destructive 3D pathology with deep-learning triage," Proc. SPIE PC12363, Multiscale Imaging and Spectroscopy IV, PC1236302 (17 March 2023); https://doi.org/10.1117/12.2657485