Stroke Lesion Segmentation in Clinical Workflows: A Modular, Lightweight, and Deployment-Ready Tool

TL;DR

This paper presents StrokeSeg, a lightweight, modular tool that converts research stroke lesion segmentation models into easy-to-deploy clinical applications, maintaining high accuracy while reducing dependencies and size.

Contribution

The paper introduces StrokeSeg, a modular framework that simplifies deployment of stroke lesion segmentation models with reduced dependencies and size, enabling clinical use.

Findings

Segmentation performance equivalent to original models (Dice difference <10^{-3})

Model size reduced by approximately 50% using Float16 quantisation

Framework supports both graphical and command-line interfaces

Abstract

Deep learning frameworks such as nnU-Net achieve state-of-the-art performance in brain lesion segmentation but remain difficult to deploy clinically due to heavy dependencies and monolithic design. We introduce \textit{StrokeSeg}, a modular and lightweight framework that translates research-grade stroke lesion segmentation models into deployable applications. Preprocessing, inference, and postprocessing are decoupled: preprocessing relies on the Anima toolbox with BIDS-compliant outputs, and inference uses ONNX Runtime with \texttt{Float16} quantisation, reducing model size by about 50\%. \textit{StrokeSeg} provides both graphical and command-line interfaces and is distributed as Python scripts and as a standalone Windows executable. On a held-out set of 300 sub-acute and chronic stroke subjects, segmentation performance was equivalent to the original PyTorch pipeline (Dice difference $<10^{-3}$), demonstrating that high-performing research pipelines can be transformed into portable, clinically usable tools.