Artificial Intelligence-Enabled Retinal Vasculometry at Scale Utilizing the UK Biobank, CLSA, and NEL DESP Datasets
Bibliographic record
Abstract
Retinal imaging offers a non-invasive means to assess the circulatory system, with morphological features of retinal vessels serving as biomarkers for systemic disease. QUARTZ (QUantitative Analysis of Retinal vessel Topology and siZe) is a fully automated artificial intelligence-enabled retinal vasculometry system designed to process large-scale retinal image datasets to obtain quantitative measures of vessel morphology for use in epidemiological studies. Previously reliant on traditional image processing and machine learning, QUARTZ has now transitioned to a deep learning pipeline. Currently individually trained versions are tailored to specific datasets. Evaluation using the UK Biobank retinal dataset shows improvements in performance metrics: the F1 score for vessel segmentation increased from 0.7753 to 0.8472, accuracy for the A/V segment-level decision increased from 0.8524 to 0.9022, the detection rate for optic disc localization increased from 0.9760 to 0.9933, and the F1 score for image quality classification increased from 0.8872 to 0.9750. QUARTZ distinguishes itself from other deep learning based retinal vasculometry systems through its efficient use of data, extracting valuable information despite issues such as low levels of illumination. The high performance of QUARTZ is consistent across two other extensive retinal datasets, namely the Canadian Longitudinal Study on Aging (CLSA) and the North East London Diabetic Eye Screening Programme (NEL DESP). Evaluation on subsets was preceded by the automatic processing of entire retinal datasets by QUARTZ, processing over 1.4 million images. These retinal vasculometry outputs will serve as a valuable resource for epidemiological studies.
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) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| 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) | 0.001 | 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".