An Implementable Deep Learning Approach for Automated Risk Assessment of Stroke in Patients with Carotid Atherosclerotic Plaque - Takeaways - MDSpire

An Implementable Deep Learning Approach for Automated Risk Assessment of Stroke in Patients with Carotid Atherosclerotic Plaque

  • By

  • Yafei Gao

  • Hao Wang

  • Dingwen Zhou

  • Peipei Mai

  • Xiaona Li

  • Panpan Li

  • Yongxin Li

  • Hua Wang

  • April 21, 2026

  • 0 min

Share

  • 1

    The study developed deep learning models to enhance the accuracy of stroke risk assessment in patients with carotid atherosclerotic plaques.

  • 2

    ResNet-50 outperformed traditional machine learning models, achieving an AUC of 0.982 in predicting stroke risk.

  • 3

    The research analyzed 666 carotid plaque ultrasound images from 299 stroke patients and 367 non-stroke controls.

  • 4

    Deep learning models eliminate the need for manual feature engineering, improving objectivity and predictive performance.

  • 5

    This study highlights the potential of deep learning as a preferred clinical tool for stroke risk stratification.

Original Source(s)

An Implementable Deep Learning Approach for Automated Risk Assessment of Stroke in Patients with Carotid Atherosclerotic Plaque

  • by Yafei Gao, Hao Wang, Dingwen Zhou, Peipei Mai, Xiaona Li, Panpan Li, Yongxin Li, Hua Wang

  • April 21, 2026

Related Content

Comparative Study on the Efficacy and Safety of Different Left Atrial Appendage Occluders in One-Stop Atrial Fibrillation Procedures

Retinal Age as Disease Biomarker

Fundus-based AI screening can indicate “retinal age” of patients, as well as flag systemic disease risks, says new study

  • May 27, 2026

  • 3 min
Poststroke anxiety and depression: epidemiology, mechanisms, and management strategies