Diabetic Retinopathy Recognition and Classification Using Transfer Learning Deep Neural Networks
Bibliographic record
Abstract
Diabetic retinopathy is one of the common causes of blindness with diabetes.Early diagnosis is important to prevent irreversible vision loss.Conventional methods for diagnosing diabetic retinopathy are often based on manual examination of retinal images, which can be time-consuming and subject to human error.The integration of machine-based automated diagnostic systems offers a promising solution to this challenge.Machine-based automated diagnosis of diabetic retinopathy can prevent vision loss with early detection and treatment.In this study, we investigated the performance of different transfer learning models-DenseNet, EfficientNet, VggNet, and ResNet-on a large dataset called Diabetic retinopathy from Kaggle, consisting of 35,108 retinal images in 5 classes.Out of which 28086 samples were used for training purpose and 7,022 samples for validation testing.While previous research has explored machine learning for retinopathy diagnosis, our research uniquely combines modern transfer learning models and evaluates the effectiveness of specific processing methods with Ben Graham's processing methods.This combination distinguishes us from existing methods by contributing to a significant increase in accuracy.In particular the accuracy of the proposed approach is 97.7%, our tests show that the diagnostic accuracy increases by about 4-5% when using Ben Graham preprocessing.The results of our research may help develop more accurate and efficient automated systems for diagnosing diabetic retinopathy, thereby improving patient outcomes.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 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 source (direct Gemma or distilled Codex), 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".