Enhanced stroke risk prediction in hypertensive patients through deep learning integration of imaging and clinical
Hui Li1, Tianyu Zhang2, Guochao Han1
1Neuroelectrophysiology Department, The Second Affiliated Hospital of Qiqihar Medical College, No. 37, Zhonghua West Road, Jianhua District, Qiqihar, Heilongjiang Province, 161000, China.
Insights
This study developed a deep learning model integrating carotid ultrasound images and clinical data to predict stroke risk in hypertensive patients, achieving high accuracy for early detection and intervention.
Area of Science:
- Cardiovascular Imaging
- Artificial Intelligence in Medicine
- Stroke Prevention Research
Background:
- Stroke is a leading cause of death and disability globally, with hypertension as a major risk factor.
- Current stroke risk assessment methods often lack accuracy due to limited clinical parameters and exclusion of imaging features.
Purpose of the Study:
- To develop a deep learning multimodal model for precise stroke risk prediction in hypertensive patients.
- To integrate carotid ultrasound imaging with clinical data for enhanced risk stratification.
Main Methods:
- Utilized 2,176 carotid artery ultrasound images from 1,088 hypertensive patients.
- Employed ResNet50 for image segmentation and feature extraction, fused with clinical data via Vision Transformer (ViT).
- Risk stratification performed using a Radial Basis Probabilistic Neural Network (RBPNN), with performance evaluated by AUC, Dice, IoU, and Precision-Recall curves.
Main Results:
- The multimodal fusion model achieved an AUC of 0.97, Dice coefficient of 0.90, and IoU of 0.80.
- Ablation studies confirmed significant performance enhancement with ViT and RBPNN modules.
- The model demonstrated robust performance in high-risk subgroups, including diabetic and smoking patients.
Conclusions:
- The deep learning multimodal model accurately integrates imaging and clinical data for improved stroke risk prediction in hypertensive individuals.
- The model shows strong generalizability and potential for clinical application in early stroke screening and personalized prevention strategies.
Background:
Stroke is one of the leading causes of death and disability worldwide, with a significantly elevated incidence among individuals with hypertension. Conventional risk assessment methods primarily rely on a limited set of clinical parameters and often exclude imaging-derived structural features, resulting in suboptimal predictive accuracy.
Objective:
This study aimed to develop a deep learning-based multimodal stroke risk prediction model by integrating carotid ultrasound imaging with multidimensional clinical data to enable precise identification of high-risk individuals among hypertensive patients.
Methods:
A total of 2,176 carotid artery ultrasound images from 1,088 hypertensive patients were collected. ResNet50 was employed to automatically segment the carotid intima-media and extract key structural features. These imaging features, along with clinical variables such as age, blood pressure, and smoking history, were fused using a Vision Transformer (ViT) and fed into a Radial Basis Probabilistic Neural Network (RBPNN) for risk stratification. The model's performance was systematically evaluated using metrics including AUC, Dice coefficient, IoU, and Precision-Recall curves.
Results:
The proposed multimodal fusion model achieved outstanding performance on the test set, with an AUC of 0.97, a Dice coefficient of 0.90, and an IoU of 0.80. Ablation studies demonstrated that the inclusion of ViT and RBPNN modules significantly enhanced predictive accuracy. Subgroup analysis further confirmed the model's robust performance in high-risk populations, such as those with diabetes or smoking history.
Conclusion:
The deep learning-based multimodal fusion model effectively integrates carotid ultrasound imaging and clinical features, significantly improving the accuracy of stroke risk prediction in hypertensive patients. The model demonstrates strong generalizability and clinical application potential, offering a valuable tool for early screening and personalized intervention planning for stroke prevention.
Clinical Trial Number:
Not applicable.
Related Concept Videos
01:23Hypertension V: Nursing Management
01:27Regulation of Stroke Volume
Preload refers to the degree of stretch on the heart before it contracts. It's analogous to the stretching of a rubber band; the more it's stretched, the more forcefully it snaps back. This concept is encapsulated in the Frank-Starling law of the...
01:29Hypertension II: Pathophysiology
01:30Hypertension III: Clinical Manifestations and Diagnostic Studies
01:18Neural Regulation of Blood Pressure
Baroreceptor Reflex
Baroreceptors, located in the carotid sinuses and aortic arch, detect changes in blood pressure. When blood pressure rises, these stretch-sensitive receptors...
01:28Hypertension I: Introduction