Necessity of Local Modification for Deep Learning Algorithms to Predict Diabetic Retinopathy

Abstract

Deep learning (DL) algorithms are used to diagnose diabetic retinopathy (DR). However, most of these algorithms have been trained using global data or data from patients of a single region. Using different model architectures (e.g., Inception-v3, ResNet101, and DenseNet121), we assessed the necessity of modifying the algorithms for universal society screening. We used the open-source dataset from the Kaggle Diabetic Retinopathy Detection competition to develop a model for the detection of DR severity. We used a local dataset from Taipei City Hospital to verify the necessity of model localization and validated the three aforementioned models with local datasets. The experimental results revealed that Inception-v3 outperformed ResNet101 and DenseNet121 with a foreign global dataset, whereas DenseNet121 outperformed Inception-v3 and ResNet101 with the local dataset. The quadratic weighted kappa score (κ) was used to evaluate the model performance. All models had 5–8% higher κ for the local dataset than for the foreign dataset. Confusion matrix analysis revealed that, compared with the local ophthalmologists’ diagnoses, the severity predicted by the three models was overestimated. Thus, DL algorithms using artificial intelligence based on global data must be locally modified to ensure the applicability of a well-trained model to make diagnoses in clinical environments.

Detection of Diabetic Retinopathy using Convolutional Neural Networks for Feature Extraction and Classification (DRFEC)

“…It has achieved comparatively an accuracy of 99.03% and 73.72% in InceptionV3 and 99.46% and 78.05% in Xception than Deepa et al [ 18 ]. Tsai et al [ 82 ] and AbdelMaksoud et al [ 1 ] have implemented a DenseNet121 based CNN using Kaggle’s EyePACs train-test dataset and TCH dataset, and using EyePACS Kaggle training dataset, IDRiD, MESSIDOR and APTOS 2019 datasets, and have achieved an accuracy of 84.05% and 84.67%, and 91.2% respectively. Whereas DRFEC has achieved an accuracy of 98.80% and 73.58% using DenseNet121 on 35,126 images.…”

“…It uses Decision Tree (DT), RF and Gradient Boosting (GB) for classification and compares the overall performance of ResNet152 with all the other models. Tsai et al [ 82 ] have proposed a DL model using Inception V3, ResNet101 and DenseNet121 on global Kaggle’s EyePACs dataset and local dataset from Taipei City Hospital (TCH), for DR detection and sets a higher overestimation rate on local dataset than global dataset due to differences in regional and ethnic factors. The proposed model uses a significant dataset for the DL models employed.…”

“…On this basis we reject our null hypothesis which claims on choosing and finding a best DL model for image classification, and accept our alternative hypothesis of finding and engineering a better DL model for DR image classification. Figures 61 and 62 depicts the plot of KSPA statistical test for 26 DL models, and comparison of them with contemporary and state-of-the-art works [1, 16,18,19,23,70,72,82], respectively.…”

Automated Binary Classification of Diabetic Retinopathy by SWIN Transformer

“…The evaluation metrics for model performance in DR detection included two-class, The performance is measured with various performance measures including accuracy, Specificity, Sensitivity, and F1-Score, [12], [13].…”

The Role of Medical Image Modalities and AI in the Early Detection, Diagnosis and Grading of Retinal Diseases: A Survey

“…an SVM was used for classification. For grading DR, Tsai et al [ 179 ] applied transfer learning using three models, i.e., Inception-v3, ResNet101, and DenseNet121.…”