Optimized Apodizations with Training Simulations (OATS): Learned depth-dependent apodizations via differentiable beamforming for reduced operator tuning
Bibliographic record
Abstract
Ultrasound is a point-of-care imaging modality that allows for real-time operation. While real-time capabilities are advantageous, one potential concern is the operator dependency of the modality as settings selected by the operator can alter the image appearance. Improvement of the B-mode image to necessitate fewer adjustments can reduce the operator dependency. Here, we propose a supervised learning framework (Optimal Apodizations with Training Simulations - OATS) to devise new apodization weights for image quality improvement. Our framework relies on the use of a differentiable beamformer to iteratively optimize apodization weights by comparing differences between simulated ground truth images and the corresponding post-beamformed images (simulated training set of over 200 images). We experimentally verified that these apodization weights resulted in higher quality B-mode images on both simulated and real-world data for focused and unfocused imaging scenarios. In the focused imaging scenario, the OATS-apodized images demonstrated reduced sidelobe artifact, improved lateral resolution (11 %), and improved signal equalization across depth when compared to a conventional Hanning apodization. In the unfocused imaging scenario, we observed reduced sidelobe artifacts and improved tissue-to-lesion contrast by up to 13 dB when compared against fixed F-number beamforming. Additionally, the OATS apodization weights were physically interpretable and learned to emulate image formation parameters such as time-gain compensation, F-number limited aperture, and transmit focus through the supervised learning procedure. Overall, the proposed framework successfully learned generalizable receive apodizations to improve image quality.
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 | 0.001 |
| Science and technology studies | 0.001 | 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".