MAESTRO: A Public, Generalizable Model for Stroke Lesion Segmentation from T1 MRI Across the Recovery Continuum
Preprint
Abstract
Accurate stroke lesion segmentation is essential for large-scale neuroimaging studies, yet manual delineation remains labor-intensive, and existing automated methods often struggle to generalize across imaging protocols and stages of recovery. We developed MAESTRO, a deep learning framework for automated lesion segmentation across the stroke recovery continuum using T1-weighted (T1) MRI alone. We hypothesized that combining a transformer-based architecture with an image augmentation strategy would improve segmentation accuracy and robustness under heterogeneous imaging conditions. T1 MRI scans and expert-traced lesion masks from 955 stroke participants across 33 international cohorts were used to train and evaluate MAESTRO within the open-source nnU-Net framework. Performance was evaluated on a held-out test set using spatial and volumetric agreement metrics. An exploratory human-in-the-loop (HITL) evaluation compared correction of MAESTRO-generated segmentations with manual tracing from scratch. MAESTRO achieved the strongest performance across several evaluated model configurations, providing the most accurate lesion localization and lesion volume estimates (median Dice = 0.686; Pearson r = 0.861; ICC = 0.792). Segmentation performance was sensitive to lesion size and stroke chronicity but remained robust across diverse imaging conditions. Additionally, using a HITL workflow to correct MAESTRO segmentations reduced annotation time by 47.4% compared to manual tracing while improving accuracy relative to both automated and manual workflows. MAESTRO is publicly available to enable robust, automated stroke lesion segmentation from T1 MRI. When combined with human review and correction, MAESTRO offers a practical approach for generating standardized, high-quality lesion annotations, helping reduce a major barrier to large-scale stroke imaging studies.Keywords
This publication has 28 references indexed in Scilit:
- Associations between contralesional neuroplasticity and motor impairment through deep learning-derived MRI regional brain age in chronic stroke (ENIGMA): a multicohort, retrospective, observational studyThe Lancet Digital Health, 2026
- Non-inferiority of deep learning ischemic stroke segmentation on non-contrast CT within 16-hours compared to expert neuroradiologistsScientific Reports, 2023
- Quality assessment, variability and reproducibility of anatomical measurements derived from T1-weighted brain imaging: The RIN–Neuroimaging Network case studyPhysica Medica, 2023
- Robust machine learning segmentation for large-scale analysis of heterogeneous clinical brain MRI datasetsProceedings of the National Academy of Sciences, 2023
- A large, curated, open-source stroke neuroimaging dataset to improve lesion segmentation algorithmsScientific Data, 2022
- Automated multimodal segmentation of acute ischemic stroke lesions on clinical MR imagesMagnetic Resonance Imaging, 2022
- Joint super-resolution and synthesis of 1 mm isotropic MP-RAGE volumes from clinical MRI exams with scans of different orientation, resolution and contrastNeuroImage, 2021
- nnU-Net: a self-configuring method for deep learning-based biomedical image segmentationNature Methods, 2020
- The ENIGMA Stroke Recovery Working Group: Big data neuroimaging to study brain–behavior relationships after strokeHuman Brain Mapping, 2020
- A comparison of automated lesion segmentation approaches for chronic stroke T1‐weighted MRI dataHuman Brain Mapping, 2019