Deep learning measures pancreatic duct width from 3d scans
Uncertainty-aware Coarse-to-Fine Main Pancreatic Duct Segmentation for Width Measurement in 3D MRCP
Computer Vision and Pattern Recognition
Summary
Measuring the main pancreatic duct's size in 3D scans helps doctors spot problems like cancer or blockages. The authors developed a computer method that first finds the duct roughly, then zooms in to trace its shape more carefully. Their method can also estimate how confident it is about the measurements. Tested on real patient scans, it found duct width with good accuracy and detected dilated ducts reasonably well. This could help doctors get consistent measurements without needing to do everything by hand.
What this means in practice
- •For radiology departments: automate measurement of pancreatic duct width from MRCP scans to support diagnosis and reduce variability between readers
- •For medical imaging software developers: integrate uncertainty-aware segmentation and width estimation of pancreatic ducts into diagnostic imaging tools using 3D MRCP data
Authors
Zipei Zhang, Jaakko Sahlsten, Mikko Saukkoriipi, Johanna Ronkainen, Mona Lehto, Akseli Aittoniemi, Irina Rinta-Kiikka, Kimmo Kaski
Abstract
Assessment of the main pancreatic duct (MPD) from volumetric magnetic resonance cholangiopancreatography (MRCP) scans is an essential task for ductal dilatation diagnosis and pancreatic cancer analysis. However, manual assessment is affected by reader variability in which automated analysis may assist. Nevertheless, fully automated and uncertainty-aware MPD analysis from volumetric MRCP remains underexplored area of research. In this study, we propose an automatic and uncertainty-aware deep learning-based pipeline for MPD segmentation and maximum-width analysis from MRCP scans. A coarse localisation model first identifies the ductal region-of-interest, followed by a fine model that segments the MPD within the region. The dataset consists of 256 MPD-segmented MRCP scans from 69 patients. A randomly selected holdout test subset of 14 patients with 55 MRCP volumes was used for evaluation. We employed a five-model cross-validation ensemble for improved performance and uncertainty estimation. The best MPD segmentation performance was 0.634 DSC, 0.744 precision, and 0.593 recall. The mean absolute error for maximum MPD width prediction was 0.355~mm and dilated MPD ($\geq 5$~mm) was detected with 75.0\% sensitivity and 74.4\% specificity. These novel results suggest that automated MPD segmentation can be used to aid MRCP-based duct measurements.