Video-SwinUNet: Spatio-temporal Deep Learning Framework for VFSS Instance Segmentation

TL;DR

This paper introduces Video-SwinUNet, a deep learning framework that effectively captures spatio-temporal features for medical video segmentation, significantly improving performance on VFSS datasets by integrating CNNs, transformers, and temporal feature blending.

Contribution

The paper proposes a novel spatio-temporal deep learning framework combining CNNs, transformers, and temporal feature blending for improved medical video segmentation.

Findings

Achieved a dice coefficient of 0.8986 and 0.8186 on VFSS2022 datasets.

Outperforms existing methods significantly in segmentation benchmarks.

Demonstrates effective cross-dataset transferability of the model.

Abstract

This paper presents a deep learning framework for medical video segmentation. Convolution neural network (CNN) and transformer-based methods have achieved great milestones in medical image segmentation tasks due to their incredible semantic feature encoding and global information comprehension abilities. However, most existing approaches ignore a salient aspect of medical video data - the temporal dimension. Our proposed framework explicitly extracts features from neighbouring frames across the temporal dimension and incorporates them with a temporal feature blender, which then tokenises the high-level spatio-temporal feature to form a strong global feature encoded via a Swin Transformer. The final segmentation results are produced via a UNet-like encoder-decoder architecture. Our model outperforms other approaches by a significant margin and improves the segmentation benchmarks on the VFSS2022 dataset, achieving a dice coefficient of 0.8986 and 0.8186 for the two datasets tested. Our studies also show the efficacy of the temporal feature blending scheme and cross-dataset transferability of learned capabilities. Code and models are fully available at https://github.com/SimonZeng7108/Video-SwinUNet.