Joint Modeling of Image and Label Statistics for Enhancing Model Generalizability of Medical Image Segmentation

06/09/2022
by   Shangqi Gao, et al.
0

Although supervised deep-learning has achieved promising performance in medical image segmentation, many methods cannot generalize well on unseen data, limiting their real-world applicability. To address this problem, we propose a deep learning-based Bayesian framework, which jointly models image and label statistics, utilizing the domain-irrelevant contour of a medical image for segmentation. Specifically, we first decompose an image into components of contour and basis. Then, we model the expected label as a variable only related to the contour. Finally, we develop a variational Bayesian framework to infer the posterior distributions of these variables, including the contour, the basis, and the label. The framework is implemented with neural networks, thus is referred to as deep Bayesian segmentation. Results on the task of cross-sequence cardiac MRI segmentation show that our method set a new state of the art for model generalizability. Particularly, the BayeSeg model trained with LGE MRI generalized well on T2 images and outperformed other models with great margins, i.e., over 0.47 in terms of average Dice. Our code is available at https://zmiclab.github.io/projects.html.

READ FULL TEXT

page 7

page 8

research
03/03/2023

BayeSeg: Bayesian Modeling for Medical Image Segmentation with Interpretable Generalizability

Due to the cross-domain distribution shift aroused from diverse medical ...
research
08/10/2022

KiPA22 Report: U-Net with Contour Regularization for Renal Structures Segmentation

Three-dimensional (3D) integrated renal structures (IRS) segmentation is...
research
12/25/2019

Neural ODEs for Image Segmentation with Level Sets

We propose a novel approach for image segmentation that combines Neural ...
research
12/19/2022

Segmentation Ability Map: Interpret deep features for medical image segmentation

Deep convolutional neural networks (CNNs) have been widely used for medi...
research
04/12/2023

UniverSeg: Universal Medical Image Segmentation

While deep learning models have become the predominant method for medica...
research
04/06/2022

PlutoNet: An Efficient Polyp Segmentation Network

Polyps in the colon can turn into cancerous cells if not removed with ea...
research
10/25/2021

Interactive Segmentation via Deep Learning and B-Spline Explicit Active Surfaces

Automatic medical image segmentation via convolutional neural networks (...

Please sign up or login with your details

Forgot password? Click here to reset