A preliminary numerical investigation of convolutional neural network (CNN) techniques for filtering high-intensity focused ultrasound (HIFU) noise in images
Bibliographic record
Abstract
High-intensity focused ultrasound (HIFU) is a minimally invasive medical procedure that uses ultrasonic waves to ablate or heat tissue with the aim of treating tumors and tremors. Diagnostic ultrasound imaging is the primary mode of imaging during HIFU treatments due to its real-time capabilities. However, HIFU noise, produced from therapeutic ultrasound components, interfere with the diagnostic ultrasound components and cause difficulty in monitoring changes to tissue during treatment. In a multitude of HIFU treatments, deep learning has been used as a tool to detect coagulation, monitor temperature, and segment tumors. Convolutional neural network (CNN) models are a series of deep learning algorithms that can assign importance to aspects of an inputted image and differentiate one from the other. Based on previous methods of filtering, CNNs too can be trained to filter raw RF signals received by an ultrasound probe for subsequent real-time treatment feedback with HIFU. Here, we were able to present a preliminary investigation of a CNN approach for HIFU noise reduction. To do this, we used acoustic wave simulations from k-Wave, a time-domain, full-wave model for ultrasound wave propagation, in combination with the Deep Learning Toolbox from MATLAB. Subsequent analyses studied the performance of noise reduction via the proposed regression model.
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 | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 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".