Development of a robust eye exam diagnosis platform with a deep learning model

A difficulty-aware and task-augmentation method based on meta-learning model for few-shot diabetic retinopathy classification

Background

Accurate classification techniques are essential for the early diagnosis and treatment of patients with diabetic retinopathy (DR). However, the limited amount of annotated DR data poses a challenge for existing deep-learning models. This article proposes a difficulty-aware and task-augmentation method based on meta-learning (DaTa-ML) model for few-shot DR classification with fundus images.

Methods

The difficulty-aware (Da) method operates by dynamically modifying the cross-entropy loss function applied to learning tasks. This methodology has the ability to intelligently down-weight simpler tasks, while simultaneously prioritizing more challenging tasks. These adjustments occur automatically and aim to optimize the learning process. Additionally, the task-augmentation (Ta) method is used to enhance the meta-training process by augmenting the number of tasks through image rotation and improving the feature-extraction capability. To implement the expansion of the meta-training tasks, various task instances can be sampled during the meta-training stage. Ultimately, the proposed Ta method was introduced to optimize the initialization parameters and enhance the meta-generalization performance of the model. The DaTa-ML model showed promising results by effectively addressing the challenges associated with few-shot DR classification.

Results

The Asia Pacific Tele-Ophthalmology Society (APTOS) 2019 blindness detection data set was used to evaluate the DaTa-ML model. The results showed that with only 1% of the training data (5-way, 20-shot) and a single update step (training time reduced by 90%), the DaTa-ML model had an accuracy rate of 89.6% on the test data, which is a 1.7% improvement over the transfer-learning method [i.e., residual neural network (ResNet)50 pre-trained on ImageNet], and a 16.8% improvement over scratch-built models (i.e., ResNet50 without pre-trained weights), despite having fewer trainable parameters (the parameters used by the DaTa-ML model are only 0.47% of the ResNet50 parameters).

Conclusions

The DaTa-ML model provides a more efficient DR classification solution with little annotated data and has significant advantages over state-of-the-art methods. Thus, it could be used to guide and assist ophthalmologists to determine the severity of DR.

Evaluation of visual plasticity in patients with refractive amblyopia treated using a visual perceptual learning system

BACKGROUND

Amblyopia is a neurological deficit in binocular vision that affects 3% of the population and is the result of disruptions in early visual development.

OBJECTIVE

In this study, we used a visual perceptual learning system for the short-term treatment of children with ametropic amblyopia and evaluated the clinical efficacy of this system in terms of visual plasticity.

METHODS

We conducted a retrospective analysis of the clinical data of 114 children (228 eyes) with refractive amblyopia, who were aged 6.51 ± 1.51 years. Prior to the treatment, we evaluated all children with amblyopia using the visual information processing test. We determined the type of amblyopic defect according to the type of amblyopia, corrected visual acuity, and advanced visual function test results. Based on the type of defect, each child with amblyopia was given short-term visual perception training for 10 days. Finally, we compared the results of visual acuity and visual information processing tests before and after the treatment.

RESULTS

The best-corrected visual acuity of patients was better after 10 days of visual training than that before training (P< 0.05). The perceptual eye position after training improved with statistically significant differences in horizontal and vertical perceptual eye position (both P< 0.05) compared to that before training. The number of amblyopic children without suppression in both eyes was 81 cases (71.1%) after training which was higher than that (65 cases, or 57.0%) before training, with a statistically significant difference (P< 0.05). Binocular fine stereopsis and dynamic stereopsis improved after training with a statistically significant difference (both P< 0.05).

CONCLUSION

In this study, it was found that patients with amblyopia showed visual plasticity. Moreover, continuous visual perceptual learning improved the best-corrected visual acuity and recovered stereopsis in children with refractive amblyopia.

Multi-dimensional dense attention network for pixel-wise segmentation of optic disc in colour fundus images

BACKGROUND

Segmentation of retinal fragments like blood vessels, Optic Disc (OD), and Optic Cup (OC) enables the early detection of different retinal pathologies like Diabetic Retinopathy (DR), Glaucoma, etc.

OBJECTIVE

Accurate segmentation of OD remains challenging due to blurred boundaries, vessel occlusion, and other distractions and limitations. These days, deep learning is rapidly progressing in the segmentation of image pixels, and a number of network models have been proposed for end-to-end image segmentation. However, there are still certain limitations, such as limited ability to represent context, inadequate feature processing, limited receptive field, etc., which lead to the loss of local details and blurred boundaries.

METHODS

A multi-dimensional dense attention network, or MDDA-Net, is proposed for pixel-wise segmentation of OD in retinal images in order to address the aforementioned issues and produce more thorough and accurate segmentation results. In order to acquire powerful contexts when faced with limited context representation capabilities, a dense attention block is recommended. A triple-attention (TA) block is introduced in order to better extract the relationship between pixels and obtain more comprehensive information, with the goal of addressing the insufficient feature processing. In the meantime, a multi-scale context fusion (MCF) is suggested for acquiring the multi-scale contexts through context improvement.

RESULTS

Specifically, we provide a thorough assessment of the suggested approach on three difficult datasets. In the MESSIDOR and ORIGA data sets, the suggested MDDA-NET approach obtains accuracy levels of 99.28% and 98.95%, respectively.

CONCLUSION

The experimental results show that the MDDA-Net can obtain better performance than state-of-the-art deep learning models under the same environmental conditions.