Thickened 2D Networks for Efficient 3D Medical Image Segmentation

TL;DR

This paper introduces a method that enhances 2D networks with thickened inputs and attention mechanisms to efficiently perform 3D medical image segmentation, achieving high accuracy with lower latency.

Contribution

The paper proposes a novel thickened 2D network approach with early multiplexing and slice sensitive attention to incorporate 3D context, improving segmentation performance.

Findings

Higher segmentation accuracy on abdominal organs

Lower inference latency compared to traditional 3D methods

Effective in capturing 3D context with 2D networks

Abstract

There has been a debate in 3D medical image segmentation on whether to use 2D or 3D networks, where both pipelines have advantages and disadvantages. 2D methods enjoy a low inference time and greater transfer-ability while 3D methods are superior in performance for hard targets requiring contextual information. This paper investigates efficient 3D segmentation from another perspective, which uses 2D networks to mimic 3D segmentation. To compensate the lack of contextual information in 2D manner, we propose to thicken the 2D network inputs by feeding multiple slices as multiple channels into 2D networks and thus 3D contextual information is incorporated. We also put forward to use early-stage multiplexing and slice sensitive attention to solve the confusion problem of information loss which occurs when 2D networks face thickened inputs. With this design, we achieve a higher performance while maintaining a lower inference latency on a few abdominal organs from CT scans, in particular when the organ has a peculiar 3D shape and thus strongly requires contextual information, demonstrating our method's effectiveness and ability in capturing 3D information. We also point out that "thickened" 2D inputs pave a new method of 3D segmentation, and look forward to more efforts in this direction. Experiments on segmenting a few abdominal targets in particular blood vessels which require strong 3D contexts demonstrate the advantages of our approach.