Screening of Common Retinal Diseases Using Six-Category Models Based on EfficientNet

Application and visualization study of an intelligence-assisted classification model for common eye diseases using B-mode ultrasound images

Aim

Conventional approaches to diagnosing common eye diseases using B-mode ultrasonography are labor-intensive and time-consuming, must requiring expert intervention for accuracy. This study aims to address these challenges by proposing an intelligence-assisted analysis five-classification model for diagnosing common eye diseases using B-mode ultrasound images.

Methods

This research utilizes 2064 B-mode ultrasound images of the eye to train a novel model integrating artificial intelligence technology.

Results

The ConvNeXt-L model achieved outstanding performance with an accuracy rate of 84.3% and a Kappa value of 80.3%. Across five classifications (no obvious abnormality, vitreous opacity, posterior vitreous detachment, retinal detachment, and choroidal detachment), the model demonstrated sensitivity values of 93.2%, 67.6%, 86.1%, 89.4%, and 81.4%, respectively, and specificity values ranging from 94.6% to 98.1%. F1 scores ranged from 71% to 92%, while AUC values ranged from 89.7% to 97.8%.

Conclusion

Among various models compared, the ConvNeXt-L model exhibited superior performance. It effectively categorizes and visualizes pathological changes, providing essential assisted information for ophthalmologists and enhancing diagnostic accuracy and efficiency.

Prediction of spherical equivalent refraction and axial length in children based on machine learning

Purpose

Recently, the proportion of patients with high myopia has shown a continuous growing trend, more toward the younger age groups. This study aimed to predict the changes in spherical equivalent refraction (SER) and axial length (AL) in children using machine learning methods.

Methods

This study is a retrospective study. The cooperative ophthalmology hospital of this study collected data on 179 sets of childhood myopia examinations. The data collected included AL and SER from grades 1 to 6. This study used the six machine learning models to predict AL and SER based on the data. Six evaluation indicators were used to evaluate the prediction results of the models.

Results

For predicting SER in grade 6, grade 5, grade 4, grade 3, and grade 2, the best results were obtained through the multilayer perceptron (MLP) algorithm, MLP algorithm, orthogonal matching pursuit (OMP) algorithm, OMP algorithm, and OMP algorithm, respectively. The R² of the five models were 0.8997, 0.7839, 0.7177, 0.5118, and 0.1758, respectively. For predicting AL in grade 6, grade 5, grade 4, grade 3, and grade 2, the best results were obtained through the Extra Tree (ET) algorithm, MLP algorithm, kernel ridge (KR) algorithm, KR algorithm, and MLP algorithm, respectively. The R² of the five models were 0.7546, 0.5456, 0.8755, 0.9072, and 0.8534, respectively.

Conclusion

Therefore, in predicting SER, the OMP model performed better than the other models in most experiments. In predicting AL, the KR and MLP models were better than the other models in most experiments.

Automated detection of myopic maculopathy using five-category models based on vision outlooker for visual recognition

Purpose

To propose a five-category model for the automatic detection of myopic macular lesions to help grassroots medical institutions conduct preliminary screening of myopic macular lesions from limited number of color fundus images.

Methods

First, 1,750 fundus images of non-myopic retinal lesions and four categories of pathological myopic maculopathy were collected, graded, and labeled. Subsequently, three five-classification models based on Vision Outlooker for Visual Recognition (VOLO), EfficientNetV2, and ResNet50 for detecting myopic maculopathy were trained with data-augmented images, and the diagnostic results of the different trained models were compared and analyzed. The main evaluation metrics were sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV), area under the curve (AUC), kappa and accuracy, and receiver operating characteristic curve (ROC).

Results

The diagnostic accuracy of the VOLO-D2 model was 96.60% with a kappa value of 95.60%. All indicators used for the diagnosis of myopia-free macular degeneration were 100%. The sensitivity, NPV, specificity, and PPV for diagnosis of leopard fundus were 96.43, 98.33, 100, and 100%, respectively. The sensitivity, specificity, PPV, and NPV for the diagnosis of diffuse chorioretinal atrophy were 96.88, 98.59, 93.94, and 99.29%, respectively. The sensitivity, specificity, PPV, and NPV for the diagnosis of patchy chorioretinal atrophy were 92.31, 99.26, 97.30, and 97.81%, respectively. The sensitivity, specificity, PPV, and NPV for the diagnosis of macular atrophy were 100, 98.10, 84.21, and 100%, respectively.

Conclusion

The VOLO-D2 model accurately identified myopia-free macular lesions and four pathological myopia-related macular lesions with high sensitivity and specificity. It can be used in screening pathological myopic macular lesions and can help ophthalmologists and primary medical institution providers complete the initial screening diagnosis of patients.

Classification of dry and wet macular degeneration based on the ConvNeXT model

Purpose

To assess the value of an automated classification model for dry and wet macular degeneration based on the ConvNeXT model.

Methods

A total of 672 fundus images of normal, dry, and wet macular degeneration were collected from the Affiliated Eye Hospital of Nanjing Medical University and the fundus images of dry macular degeneration were expanded. The ConvNeXT three-category model was trained on the original and expanded datasets, and compared to the results of the VGG16, ResNet18, ResNet50, EfficientNetB7, and RegNet three-category models. A total of 289 fundus images were used to test the models, and the classification results of the models on different datasets were compared. The main evaluation indicators were sensitivity, specificity, F1-score, area under the curve (AUC), accuracy, and kappa.

Results

Using 289 fundus images, three-category models trained on the original and expanded datasets were assessed. The ConvNeXT model trained on the expanded dataset was the most effective, with a diagnostic accuracy of 96.89%, kappa value of 94.99%, and high diagnostic consistency. The sensitivity, specificity, F1-score, and AUC values for normal fundus images were 100.00, 99.41, 99.59, and 99.80%, respectively. The sensitivity, specificity, F1-score, and AUC values for dry macular degeneration diagnosis were 87.50, 98.76, 90.32, and 97.10%, respectively. The sensitivity, specificity, F1-score, and AUC values for wet macular degeneration diagnosis were 97.52, 97.02, 96.72, and 99.10%, respectively.

Conclusion

The ConvNeXT-based category model for dry and wet macular degeneration automatically identified dry and wet macular degeneration, aiding rapid, and accurate clinical diagnosis.

GCLDNet: Gastric cancer lesion detection network combining level feature aggregation and attention feature fusion

Background

Analysis of histopathological slices of gastric cancer is the gold standard for diagnosing gastric cancer, while manual identification is time-consuming and highly relies on the experience of pathologists. Artificial intelligence methods, particularly deep learning, can assist pathologists in finding cancerous tissues and realizing automated detection. However, due to the variety of shapes and sizes of gastric cancer lesions, as well as many interfering factors, GCHIs have a high level of complexity and difficulty in accurately finding the lesion region. Traditional deep learning methods cannot effectively extract discriminative features because of their simple decoding method so they cannot detect lesions accurately, and there is less research dedicated to detecting gastric cancer lesions.

Methods

We propose a gastric cancer lesion detection network (GCLDNet). At first, GCLDNet designs a level feature aggregation structure in decoder, which can effectively fuse deep and shallow features of GCHIs. Second, an attention feature fusion module is introduced to accurately locate the lesion area, which merges attention features of different scales and obtains rich discriminative information focusing on lesion. Finally, focal Tversky loss (FTL) is employed as a loss function to depress false-negative predictions and mine difficult samples.

Results

Experimental results on two GCHI datasets of SEED and BOT show that DSCs of the GCLDNet are 0.8265 and 0.8991, ACCs are 0.8827 and 0.8949, JIs are 0.7092 and 0.8182, and PREs are 0.7820 and 0.8763, respectively.

Conclusions

Experimental results demonstrate the effectiveness of GCLDNet in the detection of gastric cancer lesions. Compared with other state-of-the-art (SOTA) detection methods, the GCLDNet obtains a more satisfactory performance. This research can provide good auxiliary support for pathologists in clinical diagnosis.

Automated measurement of the disc-fovea angle based on DeepLabv3

Purpose

To assess the value of automatic disc-fovea angle (DFA) measurement using the DeepLabv3+ segmentation model.

Methods

A total of 682 normal fundus image datasets were collected from the Eye Hospital of Nanjing Medical University. The following parts of the images were labeled and subsequently reviewed by ophthalmologists: optic disc center, macular center, optic disc area, and virtual macular area. A total of 477 normal fundus images were used to train DeepLabv3+, U-Net, and PSPNet model, which were used to obtain the optic disc area and virtual macular area. Then, the coordinates of the optic disc center and macular center were obstained by using the minimum outer circle technique. Finally the DFA was calculated.

Results

In this study, 205 normal fundus images were used to test the model. The experimental results showed that the errors in automatic DFA measurement using DeepLabv3+, U-Net, and PSPNet segmentation models were 0.76°, 1.4°, and 2.12°, respectively. The mean intersection over union (MIoU), mean pixel accuracy (MPA), average error in the center of the optic disc, and average error in the center of the virtual macula obstained by using DeepLabv3+ model was 94.77%, 97.32%, 10.94 pixels, and 13.44 pixels, respectively. The automatic DFA measurement using DeepLabv3+ got the less error than the errors that using the other segmentation models. Therefore, the DeepLabv3+ segmentation model was finally chosen to measure DFA automatically.

Conclusions

The DeepLabv3+ segmentation model -based automatic segmentation techniques can produce accurate and rapid DFA measurements.