Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 Jun 15;13(6):947.
doi: 10.3390/mi13060947.

Multi-Label Fundus Image Classification Using Attention Mechanisms and Feature Fusion

Zhenwei Li  1 , Mengying Xu  1 , Xiaoli Yang  1 , Yanqi Han  1
Affiliations

Multi-Label Fundus Image Classification Using Attention Mechanisms and Feature Fusion

Zhenwei Li et al. Micromachines (Basel). .

Abstract

Fundus diseases can cause irreversible vision loss in both eyes if not diagnosed and treated immediately. Due to the complexity of fundus diseases, the probability of fundus images containing two or more diseases is extremely high, while existing deep learning-based fundus image classification algorithms have low diagnostic accuracy in multi-labeled fundus images. In this paper, a multi-label classification of fundus disease with binocular fundus images is presented, using a neural network algorithm model based on attention mechanisms and feature fusion. The algorithm highlights detailed features in binocular fundus images, and then feeds them into a ResNet50 network with attention mechanisms to extract fundus image lesion features. The model obtains global features of binocular images through feature fusion and uses Softmax to classify multi-label fundus images. The ODIR binocular fundus image dataset was used to evaluate the network classification performance and conduct ablation experiments. The model's backend is the Tensorflow framework. Through experiments on the test images, this method achieved accuracy, precision, recall, and F1 values of 94.23%, 99.09%, 99.23%, and 99.16%, respectively.

Keywords: attention mechanisms; deep learning; feature fusion; fundus images; image classification.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure 1
Figure 1
Images of the ODIR dataset. (a) No disease. (b) Diabetic retinopathy and myopia.
Figure 2
Figure 2
Classification of attentional mechanisms (ф indicates no relevant classification).
Figure 3
Figure 3
BFPC-Net model. IAM: image augmentation model; RAM: residual attention module; FFM: feature fusion module.
Figure 4
Figure 4
Distribution of ODIR dataset. Normal (N), diabetes (D), glaucoma (G), cataract (C), age-related macular degeneration (A), hypertension (H), pathological myopia (M), other diseases/abnormalities (O).
Figure 5
Figure 5
Image preprocessing flow.
Figure 6
Figure 6
Image augmentation results.
Figure 7
Figure 7
Residual attention module.
Figure 8
Figure 8
BFPC-Net training set and validation set losses.
Figure 9
Figure 9
BFPC-Net training set and validation set accuracy.
Figure 10
Figure 10
BFPC-Net test classification indicators for each disease.
Figure 11
Figure 11
Different image size evaluation metrics for BFPC-Net.
See this image and copyright information in PMC

References

    1. Chen T.-L., Wang J., Yuan F. A preliminary study of a deep learning assisted diagnostic system with an artificial intelligence for detection of retina disease. Int. Eye Sci. 2020;20:1452–1455.
    1. Aggarwal A., Chakradar M., Bhatia M.S., Kumar M., Stephan T., Gupta S.K., Alsamhi S.H., Al-Dois H. COVID-19 Risk Prediction for Diabetic Patients Using Fuzzy Inference System and Machine Learning Approaches. J. Healthc. Eng. 2022;2022:1–10. doi: 10.1155/2022/4096950. - DOI - PMC - PubMed
    1. Chakradar M., Aggarwal A., Cheng X., Rani A., Kumar M., Shankar A. A Non-invasive Approach to Identify Insulin Resistance with Triglycerides and HDL-c Ratio Using Machine learning. Neural Process. Lett. 2021:1–21. doi: 10.1007/s11063-021-10461-6. - DOI
    1. Li C., Ye J., He J., Wang S., Qiao Y., Gu L. Dense Correlation Network for Automated Multi-Label Ocular Disease Detection with Paired Color Fundus Photographs; Proceedings of the 2020 IEEE 17th International Symposium on Biomedical Imaging (ISBI); Iowa City, IA, USA. 3–7 April 2020; pp. 1–4.
    1. Song W., Cao Y., Qiao Z., Wang Q., Yang J.-J. An Improved Semi-Supervised Learning Method on Cataract Fundus Image Classification; Proceedings of the IEEE 43rd Annual Computer Software and Applications Conference (COMPSAC); Milwaukee, WI, USA. 15–19 July 2019; pp. 362–367.

LinkOut - more resources