Enhancing Diabetic Retinopathy Diagnosis with Deep Neural Networks
Bibliographic record
Abstract
DR is a critical microvascular secondary complication of diabetes, and if not identified in time, it often leads to irreversible blindness. Traditional screening techniques rely on manual fundus examination, which is not only time-consuming but mostly unavailable in under-resourced settings. This work, therefore, calls for the use of computer-aided detection and classification using fundus images of the retina. In this analysis, pre-trained Convolutional Neural Networks like ResNet50, VGG16, DenseNet, MobileNet, and EfficientNet have been fine-tuned, with image enhancement using contrast and data augmentation to emphasize subtle retinal lesions like microaneurysms and exudates. Experimental results indicated that ResNet performed best among the systems with an accuracy of 74%, thus outperforming others in classifying DR into multiple severity stages. For ease of access, this system is designed for deployment through a lightweight web-based application with real-time prediction, visualization of results, and personalized suggestions for treatment. Furthermore, this paper discusses the integration of Vision Transformers for improved lesion detection at the fine-grain level. This work contributes toward SDG 3 - Good Health and Well-being, as it proposes a scalable, accurate, affordable DR screening that would otherwise reduce clinical burden and improve healthcare equity.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".