HyMNet: A Multimodal Deep Learning System for Hypertension Prediction Using Fundus Images and Cardiometabolic Risk Factors
Study Objectives: This study aimed to develop a multimodal deep learning (MMDL) system called HyMNet, integrating fundus images and cardiometabolic factors (age and sex) to enhance hypertension (HTN) detection. Methods: HyMNet employed RETFound, a model pretrained on 1.6 million retinal images, for...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-10-01
|
| Series: | Bioengineering |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2306-5354/11/11/1080 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850217811207520256 |
|---|---|
| author | Mohammed Baharoon Hessa Almatar Reema Alduhayan Tariq Aldebasi Badr Alahmadi Yahya Bokhari Mohammed Alawad Ahmed Almazroa Abdulrhman Aljouie |
| author_facet | Mohammed Baharoon Hessa Almatar Reema Alduhayan Tariq Aldebasi Badr Alahmadi Yahya Bokhari Mohammed Alawad Ahmed Almazroa Abdulrhman Aljouie |
| author_sort | Mohammed Baharoon |
| collection | DOAJ |
| description | Study Objectives: This study aimed to develop a multimodal deep learning (MMDL) system called HyMNet, integrating fundus images and cardiometabolic factors (age and sex) to enhance hypertension (HTN) detection. Methods: HyMNet employed RETFound, a model pretrained on 1.6 million retinal images, for the fundus data, in conjunction with a fully connected neural network for age and sex. The two pathways were jointly trained by joining their feature vectors into a fusion network. The system was trained on 5016 retinal images from 1243 individuals provided by the Saudi Ministry of National Guard Health Affairs. The influence of diabetes on HTN detection was also assessed. Results: HyMNet surpassed the unimodal system, achieving an F1 score of 0.771 compared to 0.745 for the unimodal model. For diabetic patients, the F1 score was 0.796, while it was 0.466 for non-diabetic patients. Conclusions: HyMNet exhibited superior performance relative to unimodal approaches, with an F1 score of 0.771 for HyMNet compared to 0.752 for models trained on demographic data alone, underscoring the advantages of MMDL systems in HTN detection. The findings indicate that diabetes significantly impacts HTN prediction, enhancing detection accuracy among diabetic patients. Utilizing MMDL with diverse data sources could improve clinical applicability and generalization. |
| format | Article |
| id | doaj-art-3c916f8a20de458393d1855fce346ea2 |
| institution | OA Journals |
| issn | 2306-5354 |
| language | English |
| publishDate | 2024-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Bioengineering |
| spelling | doaj-art-3c916f8a20de458393d1855fce346ea22025-08-20T02:07:58ZengMDPI AGBioengineering2306-53542024-10-011111108010.3390/bioengineering11111080HyMNet: A Multimodal Deep Learning System for Hypertension Prediction Using Fundus Images and Cardiometabolic Risk FactorsMohammed Baharoon0Hessa Almatar1Reema Alduhayan2Tariq Aldebasi3Badr Alahmadi4Yahya Bokhari5Mohammed Alawad6Ahmed Almazroa7Abdulrhman Aljouie8AI and Bioinformatics Department, King Abdullah International Medical Research Center, King Saud bin Abdulaziz University for Health Sciences (KSAU-HS), Riyadh 11481, Saudi ArabiaAI and Bioinformatics Department, King Abdullah International Medical Research Center, King Saud bin Abdulaziz University for Health Sciences (KSAU-HS), Riyadh 11481, Saudi ArabiaAI and Bioinformatics Department, King Abdullah International Medical Research Center, King Saud bin Abdulaziz University for Health Sciences (KSAU-HS), Riyadh 11481, Saudi ArabiaOphthalmology Department, King Abdulaziz Medical City, Ministry of National Guard Health Affairs, Riyadh 14611, Saudi ArabiaOphthalmology Department, Prince Mohammad bin Abdulaziz Hospital, Ministry of National Guard Health Affairs, Al Madinah 42324, Saudi ArabiaAI and Bioinformatics Department, King Abdullah International Medical Research Center, King Saud bin Abdulaziz University for Health Sciences (KSAU-HS), Riyadh 11481, Saudi ArabiaNational Center for Artificial Intelligence (NCAI), Saudi Data and Artificial Intelligence Authority (SDAIA), Riyadh 12382, Saudi ArabiaAI and Bioinformatics Department, King Abdullah International Medical Research Center, King Saud bin Abdulaziz University for Health Sciences (KSAU-HS), Riyadh 11481, Saudi ArabiaAI and Bioinformatics Department, King Abdullah International Medical Research Center, King Saud bin Abdulaziz University for Health Sciences (KSAU-HS), Riyadh 11481, Saudi ArabiaStudy Objectives: This study aimed to develop a multimodal deep learning (MMDL) system called HyMNet, integrating fundus images and cardiometabolic factors (age and sex) to enhance hypertension (HTN) detection. Methods: HyMNet employed RETFound, a model pretrained on 1.6 million retinal images, for the fundus data, in conjunction with a fully connected neural network for age and sex. The two pathways were jointly trained by joining their feature vectors into a fusion network. The system was trained on 5016 retinal images from 1243 individuals provided by the Saudi Ministry of National Guard Health Affairs. The influence of diabetes on HTN detection was also assessed. Results: HyMNet surpassed the unimodal system, achieving an F1 score of 0.771 compared to 0.745 for the unimodal model. For diabetic patients, the F1 score was 0.796, while it was 0.466 for non-diabetic patients. Conclusions: HyMNet exhibited superior performance relative to unimodal approaches, with an F1 score of 0.771 for HyMNet compared to 0.752 for models trained on demographic data alone, underscoring the advantages of MMDL systems in HTN detection. The findings indicate that diabetes significantly impacts HTN prediction, enhancing detection accuracy among diabetic patients. Utilizing MMDL with diverse data sources could improve clinical applicability and generalization.https://www.mdpi.com/2306-5354/11/11/1080artificial intelligencemachine learningcomputer visioncardiovascular diseaseshypertension detectionfundus images |
| spellingShingle | Mohammed Baharoon Hessa Almatar Reema Alduhayan Tariq Aldebasi Badr Alahmadi Yahya Bokhari Mohammed Alawad Ahmed Almazroa Abdulrhman Aljouie HyMNet: A Multimodal Deep Learning System for Hypertension Prediction Using Fundus Images and Cardiometabolic Risk Factors Bioengineering artificial intelligence machine learning computer vision cardiovascular diseases hypertension detection fundus images |
| title | HyMNet: A Multimodal Deep Learning System for Hypertension Prediction Using Fundus Images and Cardiometabolic Risk Factors |
| title_full | HyMNet: A Multimodal Deep Learning System for Hypertension Prediction Using Fundus Images and Cardiometabolic Risk Factors |
| title_fullStr | HyMNet: A Multimodal Deep Learning System for Hypertension Prediction Using Fundus Images and Cardiometabolic Risk Factors |
| title_full_unstemmed | HyMNet: A Multimodal Deep Learning System for Hypertension Prediction Using Fundus Images and Cardiometabolic Risk Factors |
| title_short | HyMNet: A Multimodal Deep Learning System for Hypertension Prediction Using Fundus Images and Cardiometabolic Risk Factors |
| title_sort | hymnet a multimodal deep learning system for hypertension prediction using fundus images and cardiometabolic risk factors |
| topic | artificial intelligence machine learning computer vision cardiovascular diseases hypertension detection fundus images |
| url | https://www.mdpi.com/2306-5354/11/11/1080 |
| work_keys_str_mv | AT mohammedbaharoon hymnetamultimodaldeeplearningsystemforhypertensionpredictionusingfundusimagesandcardiometabolicriskfactors AT hessaalmatar hymnetamultimodaldeeplearningsystemforhypertensionpredictionusingfundusimagesandcardiometabolicriskfactors AT reemaalduhayan hymnetamultimodaldeeplearningsystemforhypertensionpredictionusingfundusimagesandcardiometabolicriskfactors AT tariqaldebasi hymnetamultimodaldeeplearningsystemforhypertensionpredictionusingfundusimagesandcardiometabolicriskfactors AT badralahmadi hymnetamultimodaldeeplearningsystemforhypertensionpredictionusingfundusimagesandcardiometabolicriskfactors AT yahyabokhari hymnetamultimodaldeeplearningsystemforhypertensionpredictionusingfundusimagesandcardiometabolicriskfactors AT mohammedalawad hymnetamultimodaldeeplearningsystemforhypertensionpredictionusingfundusimagesandcardiometabolicriskfactors AT ahmedalmazroa hymnetamultimodaldeeplearningsystemforhypertensionpredictionusingfundusimagesandcardiometabolicriskfactors AT abdulrhmanaljouie hymnetamultimodaldeeplearningsystemforhypertensionpredictionusingfundusimagesandcardiometabolicriskfactors |