Diabetic Retinopathy Detection Through Retinal Vessel Segmentation Using Swin U-Net Model
Bibliographic record
Abstract
Diabetic Retinopathy (DR) is a common ocular illness that affects people in their working years and can cause vision loss. Blood vessel segmentation is crucial for the early detection and management of DR. It helps identify early warning signs including hemorrhages and microaneurysms, allowing for prompt intervention to prevent vision loss. In addition, it provides quantitative measures such as tortuosity and vascular density, which aid in making treatment decisions and charting the progression of the disease over time. But artificial intelligence (AI), and in particular deep learning (DL), is about to change medicine gradually and affect almost every area of human life. Significant improvements in diagnostic technologies have made it easier to gain insight into retinal conditions. Auto-mated screening methods considerably expedite DR detection, especially in resource-constrained contexts. Automatic blood vessel segmentation significantly improves patient outcomes in diabetic retinopathy screening programs by facilitating timely intervention and precise treatment recommendations. In this work, we apply a segmentation method to the color fundus images. First, we started by preprocessing using the CLAHE method, data augmentation, and resizing. Swin-UNet was then applied to segment retinal vessels on the RITE, Drive, STARE, and ChaseDB1 datasets. The used Swin U-Net model obtained a specificity of 92.83%, a sensitivity of 94.67%, and a Dice score of 91.24% on the ChaseDB1, STARE, and Drive datasets. Our model obtained a specificity of 86%, a sensitivity of 85%, and a Dice score of 87% on the RITE dataset.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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".