Semi-Supervised Learning for Retinal Disease Detection: A BIOMISA Study
Bibliographic record
Abstract
Proper immediate identification of Age-related Macular Degeneration (AMD) together with Central Serous Retinopathy (CSR) and Macular Edema (ME) is crucial for protecting vision. OCT imaging achieves better condition detection through automated model-based detection processes. The majority of studies in this domain utilize supervised learning because these approaches need large labeled dataset resources. The method confronts two essential obstacles due to limited medical data labeling quality, expensive expert training costs, and with irregular medical condition distributions. The considered factors limit practical implementation of these methods and their meaningful expansions. The study evaluates how semi-supervised learning techniques analyze retinal diseases in images that originate from the BIOMISA Macula database while providing diagnostic details about AMD, CSR, and ME in addition to Normal retinal results. SSL functions uniquely from fully supervised methods through its unique capability to process labeled and unlabeled data, which lowers manual annotation needs while improving generalized output performance. SSL delivers better results than traditional supervised learning practices through its ability to manage class irregularities and process extensive medical image files. The establishment of SSL as an attractive third option in medical settings with limited labeled data proves through research findings. The study provides insights regarding SSL use in diagnosis of retinal diseases alongside demonstrating its medical potential in healthcare environments. Future investigation designs improved deep learning algorithms which would enable higher system scalability and cost-effective diagnostics for ophthalmic disease systems.
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 | 0.004 | 0.008 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| 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".