Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

LightMedSeg: Lightweight 3D Medical Image Segmentation with Learned Spatial Anchors

Kavyansh Tyagi, Vishwas Rathi, Puneet Goyal

TL;DR

LightMedSeg is a modular UNet-style segmentation architecture that integrates anatomical priors with adaptive context modeling that achieves segmentation accuracy within a few Dice points of heavy transformer baselines and is a deployable and data-efficient solution for 3D medical image segmentation.

Abstract

Accurate and efficient 3D medical image segmentation is essential for clinical AI, where models must remain reliable under stringent memory, latency, and data availability constraints. Transformer-based methods achieve strong accuracy but suffer from excessive parameters, high FLOPs, and limited generalization. We propose LightMedSeg, a modular UNet-style segmentation architecture that integrates anatomical priors with adaptive context modeling. Anchor-conditioned FiLM modulation enables anatomy-aware feature calibration, while a local structural prior module and texture-aware routing dynamically allocate representational capacity to boundary-rich regions. Computational redundancy is minimized through ghost and depthwise convolutions, and multi-scale features are adaptively fused via a learned skip router with anchor-relative spatial position bias. Despite requiring only 0.48M parameters and 14.64~GFLOPs, LightMedSeg achieves segmentation accuracy within a few Dice points of heavy transformer baselines. Therefore, LightMedSeg is a deployable and data-efficient solution for 3D medical image segmentation. Code will be released publicly upon acceptance.

LightMedSeg: Lightweight 3D Medical Image Segmentation with Learned Spatial Anchors

TL;DR

LightMedSeg is a modular UNet-style segmentation architecture that integrates anatomical priors with adaptive context modeling that achieves segmentation accuracy within a few Dice points of heavy transformer baselines and is a deployable and data-efficient solution for 3D medical image segmentation.

Abstract

Accurate and efficient 3D medical image segmentation is essential for clinical AI, where models must remain reliable under stringent memory, latency, and data availability constraints. Transformer-based methods achieve strong accuracy but suffer from excessive parameters, high FLOPs, and limited generalization. We propose LightMedSeg, a modular UNet-style segmentation architecture that integrates anatomical priors with adaptive context modeling. Anchor-conditioned FiLM modulation enables anatomy-aware feature calibration, while a local structural prior module and texture-aware routing dynamically allocate representational capacity to boundary-rich regions. Computational redundancy is minimized through ghost and depthwise convolutions, and multi-scale features are adaptively fused via a learned skip router with anchor-relative spatial position bias. Despite requiring only 0.48M parameters and 14.64~GFLOPs, LightMedSeg achieves segmentation accuracy within a few Dice points of heavy transformer baselines. Therefore, LightMedSeg is a deployable and data-efficient solution for 3D medical image segmentation. Code will be released publicly upon acceptance.
Paper Structure (12 sections, 17 equations, 3 figures, 4 tables)

This paper contains 12 sections, 17 equations, 3 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Proposed Method
  4. Patch Embedding Stem
  5. Global Anchor Detector
  6. Local Structural Prior Module (LSPM)
  7. Encoder Hierarchy
  8. Multi-Scale Skip Fusion
  9. Adaptive Decoder
  10. Training Objective
  11. Experimental Results
  12. Conclusion

Figures (3)

  • Figure 1: Architecture of LightMedSeg. The input volume is first embedded by a stride-2 GhostConv3D stem. The Global Anchor Detector and the Local Structural Prior Module (LSPM) extract spatial anchors $\mathbf{S}$ and a texture routing map $\mathbf{T}$ from the stem features. These priors guide (i) adaptive feature mixing before the encoder, (ii) anchor-conditioned, texture-routed processing at every encoder stage, (iii) multi-scale learned skip fusion, and (iv) an adaptive decoder with content-tied spatial position bias. A final ConvTranspose3D restores voxel-wise class logits at the original resolution.
  • Figure 2: Visual comparison of original MRI, ground truth, and predicted segmentation for four representative BraTS cases.
  • Figure 3: Visual comparison of original MRI, ground truth, and predicted segmentation for four ACDC cases.