Related Experiment Video
Updated: Jul 15, 2025
Using Retinal Imaging to Study Dementia
Published on: November 6, 2017
Non-invasive and accurate risk evaluation of cerebrovascular disease using retinal fundus photo based on deep
Lin An1, Jia Qin1, Weili Jiang2
1Guangdong Weiren Meditech Co., Ltd, Foshan, Guangdong, China.
Insights
This study introduces an Efficient Attention model using retinal fundus photos for noninvasive cerebrovascular disease (CeVD) risk prediction. The model accurately assesses CeVD risk, offering a convenient, low-cost screening solution.
Area of Science:
- Ophthalmology
- Neurology
- Medical Imaging
Background:
- Cerebrovascular disease (CeVD) is a leading cause of death and disability globally.
- Retinal microvascular changes are linked to CeVD.
- Manual analysis of retinal images for CeVD risk is inefficient.
Purpose of the Study:
- To develop a noninvasive risk prediction model for cerebrovascular disease (CeVD).
- To utilize retinal fundus photography for accurate CeVD risk assessment.
- To improve upon existing methods for detecting cerebrovascular risks.
Main Methods:
- Proposed a novel Efficient Attention model integrating convolutional neural networks and attention mechanisms.
- Applied the model to analyze retinal fundus images for cerebrovascular risk factors.
- Compared the model's performance against conventional architectures like ResNet and Efficient-Net.
Main Results:
- The Efficient Attention model achieved an accuracy (ACC) of 0.834 ± 0.03, outperforming Efficient-Net by 3.6%.
- The model demonstrated an improved area under the receiver operating characteristic curve (AUC) of 0.904 ± 0.02, surpassing other methods by 2.2%.
- The model effectively reinforced salient features in fundus photos for enhanced risk assessment.
Conclusions:
- Efficient-Attention methods provide an accurate and effective tool for cerebrovascular risk assessment.
- Retinal fundus photography shows significant potential for predicting CeVD.
- This approach offers a noninvasive, convenient, and low-cost method for large-scale CeVD screening.
Background:
Cerebrovascular disease (CeVD) is a prominent contributor to global mortality and profound disability. Extensive research has unveiled a connection between CeVD and retinal microvascular abnormalities. Nonetheless, manual analysis of fundus images remains a laborious and time-consuming task. Consequently, our objective is to develop a risk prediction model that utilizes retinal fundus photo to noninvasively and accurately assess cerebrovascular risks.
Materials And Methods:
To leverage retinal fundus photo for CeVD risk evaluation, we proposed a novel model called Efficient Attention which combines the convolutional neural network with attention mechanism. This combination aims to reinforce the salient features present in fundus photos, consequently improving the accuracy and effectiveness of cerebrovascular risk assessment.
Result:
Our proposed model demonstrates notable advancements compared to the conventional ResNet and Efficient-Net architectures. The accuracy (ACC) of our model is 0.834 ± 0.03, surpassing Efficient-Net by a margin of 3.6%. Additionally, our model exhibits an improved area under the receiver operating characteristic curve (AUC) of 0.904 ± 0.02, surpassing other methods by a margin of 2.2%.
Conclusion:
This paper provides compelling evidence that Efficient-Attention methods can serve as effective and accurate tool for cerebrovascular risk. The results of the study strongly support the notion that retinal fundus photo holds great potential as a reliable predictor of CeVD, which offers a noninvasive, convenient and low-cost solution for large scale screening of CeVD.
