Enhanced stroke risk prediction in hypertensive patients through deep learning integration of imaging and clinical data
Abstract 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 struct...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-07-01
|
| Series: | BMC Medical Informatics and Decision Making |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12911-025-03120-6 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849764386182266880 |
|---|---|
| author | Hui Li Tianyu Zhang Guochao Han Zonghui Huang Huiyu Xiao Yunzhe Ni Bo Liu Wennan Lin Yuan Lin |
| author_facet | Hui Li Tianyu Zhang Guochao Han Zonghui Huang Huiyu Xiao Yunzhe Ni Bo Liu Wennan Lin Yuan Lin |
| author_sort | Hui Li |
| collection | DOAJ |
| description | Abstract 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. Graphical Abstract |
| format | Article |
| id | doaj-art-d3c92e5a0ea345f68d5baed04dd03f6a |
| institution | DOAJ |
| issn | 1472-6947 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Medical Informatics and Decision Making |
| spelling | doaj-art-d3c92e5a0ea345f68d5baed04dd03f6a2025-08-20T03:05:09ZengBMCBMC Medical Informatics and Decision Making1472-69472025-07-0125111710.1186/s12911-025-03120-6Enhanced stroke risk prediction in hypertensive patients through deep learning integration of imaging and clinical dataHui Li0Tianyu Zhang1Guochao Han2Zonghui Huang3Huiyu Xiao4Yunzhe Ni5Bo Liu6Wennan Lin7Yuan Lin8Neuroelectrophysiology Department, The Second Affiliated Hospital of Qiqihar Medical CollegeImaging Department, The Second Affiliated Hospital of Qiqihar Medical CollegeNeuroelectrophysiology Department, The Second Affiliated Hospital of Qiqihar Medical CollegeImaging Department, The Second Affiliated Hospital of Qiqihar Medical CollegeImaging Department, The Second Affiliated Hospital of Qiqihar Medical CollegeImaging Department, The Second Affiliated Hospital of Qiqihar Medical CollegeDepartment of Neurology, The Second Affiliated Hospital of Qiqihar Medical CollegeDepartment of General Medicine, The Second Affiliated Hospital of Qiqihar Medical CollegeNeuroelectrophysiology Department, The Second Affiliated Hospital of Qiqihar Medical CollegeAbstract 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. Graphical Abstracthttps://doi.org/10.1186/s12911-025-03120-6Stroke predictionDeep learningCarotid artery ultrasoundMultimodal data fusionResNet50Vision transformer |
| spellingShingle | Hui Li Tianyu Zhang Guochao Han Zonghui Huang Huiyu Xiao Yunzhe Ni Bo Liu Wennan Lin Yuan Lin Enhanced stroke risk prediction in hypertensive patients through deep learning integration of imaging and clinical data BMC Medical Informatics and Decision Making Stroke prediction Deep learning Carotid artery ultrasound Multimodal data fusion ResNet50 Vision transformer |
| title | Enhanced stroke risk prediction in hypertensive patients through deep learning integration of imaging and clinical data |
| title_full | Enhanced stroke risk prediction in hypertensive patients through deep learning integration of imaging and clinical data |
| title_fullStr | Enhanced stroke risk prediction in hypertensive patients through deep learning integration of imaging and clinical data |
| title_full_unstemmed | Enhanced stroke risk prediction in hypertensive patients through deep learning integration of imaging and clinical data |
| title_short | Enhanced stroke risk prediction in hypertensive patients through deep learning integration of imaging and clinical data |
| title_sort | enhanced stroke risk prediction in hypertensive patients through deep learning integration of imaging and clinical data |
| topic | Stroke prediction Deep learning Carotid artery ultrasound Multimodal data fusion ResNet50 Vision transformer |
| url | https://doi.org/10.1186/s12911-025-03120-6 |
| work_keys_str_mv | AT huili enhancedstrokeriskpredictioninhypertensivepatientsthroughdeeplearningintegrationofimagingandclinicaldata AT tianyuzhang enhancedstrokeriskpredictioninhypertensivepatientsthroughdeeplearningintegrationofimagingandclinicaldata AT guochaohan enhancedstrokeriskpredictioninhypertensivepatientsthroughdeeplearningintegrationofimagingandclinicaldata AT zonghuihuang enhancedstrokeriskpredictioninhypertensivepatientsthroughdeeplearningintegrationofimagingandclinicaldata AT huiyuxiao enhancedstrokeriskpredictioninhypertensivepatientsthroughdeeplearningintegrationofimagingandclinicaldata AT yunzheni enhancedstrokeriskpredictioninhypertensivepatientsthroughdeeplearningintegrationofimagingandclinicaldata AT boliu enhancedstrokeriskpredictioninhypertensivepatientsthroughdeeplearningintegrationofimagingandclinicaldata AT wennanlin enhancedstrokeriskpredictioninhypertensivepatientsthroughdeeplearningintegrationofimagingandclinicaldata AT yuanlin enhancedstrokeriskpredictioninhypertensivepatientsthroughdeeplearningintegrationofimagingandclinicaldata |