Shang et al. examine long-term changes in memory, daily activities, depression, and frailty in stroke survivors and matched controls in China, the UK, Europe, and the US. Stroke patients have faster declines in memory and more rapid increases in depression and difficulties with daily tasks, before and after diagnosis, compared to controls.

Scientists at the Keck School of Medicine of USC have developed an automated MRI system designed to improve the evaluation of experimental stroke treatments. This open-source tool, tested on over 2,400 rodents across six research centers, reduces variability in preclinical studies by standardizing measurements of brain injury, swelling, and tissue loss. The system employs deep learning to separate brain tissue from surrounding structures while using transparent, rule-based methods to ensure reproducibility. Key advantages include matching human expert assessments, minimizing scanner-related inconsistencies, and supporting multicenter collaboration. By enabling large-scale, standardized analysis, the tool aims to accelerate the transition of promising therapies from lab research to clinical trials. The study highlights challenges in early-stage stroke treatment research, where differences in methodologies often hinder reproducibility. This innovation represents a significant step toward more reliable preclinical evaluation, potentially improving outcomes for ischemic stroke patients.