From retina to brain: how deep learning closes the gap in silent stroke screening - Takeaways - MDSpire
Coming Soon: Introducing MDSpire News. Learn more
Conexiant’s news site is now MDSpire News. Learn more

Bridging the Gap in Silent Stroke Detection: The Role of Deep Learning in Analyzing Retinal Images

  • By

  • Minyan Ge

  • Yuchun Wang

  • Shumao Xu

  • November 13, 2025

Share

  • 1

    Silent brain infarctions (SBIs) affect 20% of adults and significantly increase future stroke risk, yet they often evade detection without MRI.

  • 2

    DeepRETStroke is an AI system that analyzes retinal images to simultaneously detect SBIs and predict stroke risk, achieving 85.2% sensitivity.

  • 3

    The model was trained on approximately 900,000 retinal images, utilizing a three-stage pipeline for enhanced cerebrovascular health insights.

  • 4

    DeepRETStroke demonstrated superior performance in SBI detection and stroke prediction compared to traditional clinical metadata models.

  • 5

    Challenges remain in generalizability and potential biases due to reliance on Chinese cohorts, necessitating diverse data partnerships for broader applicability.

Original Source(s)

From retina to brain: how deep learning closes the gap in silent stroke screening

  • by Minyan Ge, Yuchun Wang, Shumao Xu

  • November 13, 2025

Related Content

Lifetime Prevalence of Technology-Facilitated Sexual Violence and Other Coercive Sexual Acts Against Women

Factors Associated With Pediatric Health Care Professionals’ Intention to Adopt Socially Assistive Robots: Mixed Methods Study

RadCoT may improve LLM detection of report errors

The radiology-specific framework produced its largest gains in detecting interpretation and omission or insertion errors across 7 language models.