Automatic Segmentation of Ischaemic Stroke Lesions Using Transformers and Convolutional Neural Networks Applied to Multimodal Neuroimaging.

Overview

1. Escuela Técnica Superior de Ingeniería Industrial, Campus Muralla del Mar, Universidad Politécnica de Cartagena, C/Doctor Fleming, s/n, 30202 Cartagena, Spain; (J.F.Z.P.); (J.M.-A.)

Abstract

Ischaemic stroke constitutes a leading cause of global disability. Rapid extraction of the infarct core from multimodal computed tomography perfusion (CTP) imaging guides reperfusion therapy and clinical decision-making. Deep learning algorithms automate this delineation, yet hospital translation is hindered by high-dimensional data, inter-scanner variability, and the low contrast of early ischaemia. Architectural comparisons in the literature frequently carry methodological biases originating from disparate preprocessing protocols and data partitions. This study reduces these variables by evaluating three segmentation strategies under a shared preprocessing pipeline and an identical data partition using the ISLES 2024 dataset. Three models were trained on the same 133-patient partition using a shared preprocessing pipeline based on morphological skull-stripping and modality-specific clinical intensity ranges. The data, preprocessing, and partitions are held constant across models, while framework-dependent factors (optimiser, patch size, physical field of view, spatial resampling, augmentation policy, and model capacity) remain coupled to each architecture and are therefore treated as part of the compared strategy rather than as fully isolated variables. The first of these is a single-stage 5-channel nnU-Net, followed by a two-stage cascaded nnU-Net (2 and 7 channels) and a lightweight Transformer (SegFormer3D). Evaluation on a fixed 15-patient held-out test set isolated the architectural performance. The cascade model achieved the highest Dice Similarity Coefficient (0.224). The single-stage nnU-Net provided the most precise volumetric estimation, recording an Absolute Volume Difference (AVD) of 23.70 mL and a lesion-wise F1-score of 7.60%. On the other hand, SegFormer3D returned the lowest overall metrics (DSC 0.163, AVD 27.28 mL, F1 2.30%). In the small held-out cohort, paired statistical testing did not reveal significant differences between models, so the reported orderings describe the present dataset and experimental configuration rather than a general architectural law. Within these limits, the local inductive bias of the convolutional models retained an empirical advantage over the single Transformer evaluated when processing this moderately sized neuroimaging dataset, and complex cascade topologies offered only marginal gains compared with a well-calibrated single-stage network. Although the predictive segmentation of infarcted tissue at acute stages still demands computational improvements, these results suggest that preprocessing quality is at least as decisive for clinical impact as increasing the complexity of neural architectures.

Reproduced under the paper's license (CC BY), from the paper cited above.

Code

The paper says that its authors' code is available on request: it was not published with the paper, so there is nothing to verify.

The paper's code and data availability statement is in the Data section.

Tracing map

A tracing map links a paper to the code its authors published: this paper has none (its code is available on request), so it has no map.

Data

No dataset and no data link were found in the paper.

Data Availability Statement

This study is based exclusively on the publicly available ISLES 2024 challenge dataset, which was released by the challenge organisers in anonymised form and can be obtained through the official ISLES 2024 challenge repository [7]. No new clinical data were collected for this work. The specific patient identifiers used for the training, validation, and held-out test partitions, together with the preprocessing and evaluation code required to reproduce the reported results, are described in Appendix A and are available from the corresponding author on reasonable request.

Reproduced under the paper's license (CC BY), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 1, 27 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 3 authors, 6 keywords, 11 MeSH terms, 1 funder, 17 references.

Cite

This paper

Cegarra, P. M., Pérez, J. F. Z., & Martínez-Alajarín, J. (2026). Automatic Segmentation of Ischaemic Stroke Lesions Using Transformers and Convolutional Neural Networks Applied to Multimodal Neuroimaging. Sensors (Basel, Switzerland), 26(16), 5021. https://doi.org/10.3390/s26165021

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.3390/jimaging12070288

GDNet: A Robust 2.5D Multimodal MRI Brain Tumor Segmentation Framework with EMA Stabilization and Tumor-Aware Sampling.

Journal: Journal of imaging

In common: methods / tools, 4 references

[2] doi:10.3390/s26154963

NeuroSPFNet: Vision-Language Semantic Prior-Guided Frequency-Enhanced Network for Brain Tumor Segmentation in Multimodal MRI.

Journal: Sensors (Basel, Switzerland)

In common: methods / tools, 3 references

[3] doi:10.3390/diagnostics16172806

Brain Tumor Segmentation and Grading on MRI Using Deep Learning: A Systematic Literature Review and Benchmark-Driven Comparative Analysis.

Journal: Diagnostics (Basel, Switzerland)

In common: methods / tools, 3 references

[4] doi:10.1038/s41598-026-53829-1 [code]

Attention U-Net with differential privacy in federated learning framework for brain stroke lesion segmentation.

Journal: Scientific reports

In common: stroke, methods / tools, 2 references

[5] doi:10.3389/frai.2026.1771088 [code]

Few-shot deployment of pretrained MRI transformers in brain imaging tasks.

Journal: Frontiers in artificial intelligence

In common: methods / tools, 2 references

[6] doi:10.1038/s41598-026-51942-9

Quantum SVM-driven framework for accurate brain stroke classification.

Journal: Scientific reports

In common: stroke, methods / tools, 1 reference

[7] doi:10.3389/fonc.2026.1834400

BrainFusionNet: an attention-augmented deep convolutional framework with hybrid loss optimisation and test-time augmentation for multi-class brain tumour detection in magnetic resonance images.

Journal: Frontiers in oncology

In common: methods / tools, 2 references

[8] doi:10.1371/journal.pone.0354511 [code]

TokenUNet: A new case for transformers integration in efficient and interpretable 3D UNets for brain imaging segmentation.

Journal: PloS one

In common: methods / tools, 2 references

[9] doi:10.3389/fonc.2026.1928589

LiteFreqMamba:lightweight frequency-enhanced Mamba for efficient 3D brain tumor segmentation.

Journal: Frontiers in oncology

In common: methods / tools, 2 references

[10] doi:10.21037/qims-2026-0792 [code]

An nnU-Net-based framework with adaptive feature representation for 3D brain tumor segmentation.

Journal: Quantitative imaging in medicine and surgery

In common: methods / tools, 2 references

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.