A machine learning approach for automated assessment of retinal vasculature in the oxygen induced retinopathy model
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Preclinical studies of vascular retinal diseases rely on the assessment of developmental dystrophies in the oxygen induced retinopathy rodent model. The quantification of vessel tufts and avascular regions is typically computed manually from flat mounted retinas imaged using fluorescent probes that highlight the vascular network. Such manual measurements are time-consuming and hampered by user variability and bias, thus a rapid and objective method is needed. Here, we introduce a machine learning approach to segment and characterize vascular tufts, delineate the whole vasculature network, and identify and analyze avascular regions. Our quantitative retinal vascular assessment (QuRVA) technique uses a simple machine learning method and morphological analysis to provide reliable computations of vascular density and pathological vascular tuft regions, devoid of user intervention within seconds. We demonstrate the high degree of error and variability of manual segmentations, and designed, coded, and implemented a set of algorithms to perform this task in a fully automated manner. We benchmark and validate the results of our analysis pipeline using the consensus of several manually curated segmentations using commonly used computer tools. The source code of our implementation is released under version 3 of the GNU General Public License ( https://www.mathworks.com/matlabcentral/fileexchange/65699-javimazzaf-qurva ).
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.
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 | 0.005 | 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) | 0.000 | 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 it