Determination of Optimum Hyperparameters in Diagnosis of Strabismus Using Artificial Intelligence Model: Cross-Sectional Study
Bibliographic record
Abstract
Objective: This study aims to analyze the effectiveness of the internet-based, free Teachable Machine (TM) platform, which does not entail code knowledge, in detecting the presence and types of strabismus in the optimum hyperparameters. Material and Methods: The images obtained from the patients who presented to our clinic with the complaint of ocular deviation were analyzed, and 523 [176 esotropia (ET), 195 exotropia (XT), and 152 orthophoria (ORTHO)] images were included in this study. After the images were uploaded to the TM platform, 6 different batch sizes and 9 different learning rates were tested using the grid search method, with the number of epochs fixed at 4,000 to determine the optimum hyperparameter. Results: The highest overall test accuracy was 0.887, and the hyperparameters from which this accuracy was obtained were 200 for the number of epochs, 256 for the batch size, and 0.0005 for the learning rate. In the TM model trained with these parameters, accuracy values of ET: 0.96, ORTHO: 0.78 and XT: 0.9 were obtained in the subgroups, respectively. Conclusion: To achieve optimal accuracy at the stage of development of the artificial intelligence model, users should determine the appropriate hyperparameter values depending on the size of the available dataset and the complexity of the data. The results we obtained by determining the optimum hyperparameters have revealed that the presence of strabismus can be detected with high accuracy using TM, an internet-based, free deep learning platform that does not entail having code knowledge.
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 | 0.017 |
| 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 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".