Enhancing the Quality of Compressed Breast Ultrasound Imagery through Application of Wavelet Convolutional Neural Networks
Bibliographic record
Abstract
Breast cancer, a pervasive and life-threatening malignancy, predominantly affects women worldwide.Despite the widespread adoption of imaging technologies such as mammography for early-stage breast cancer detection, access to such specialized imaging equipment remains limited in low-income countries.Conversely, ultrasound imaging has demonstrated its efficacy as a cost-effective tool for tumor identification.The advent of portable ultrasound devices facilitates rapid and precise lesion diagnosis in the breast, circumventing the need for hospital visits.Nevertheless, the images procured by portable ultrasound devices are typically necessitated to be transmitted in a compressed format for remote evaluation by physicians.This compression process often introduces artifacts in medical images, complicating the delineation of tumorous regions.To address this challenge, we introduce a deep-learning solution in this paper.A novel wavelet convolutional neural network (CNN) architecture is conceived to learn and subsequently diminish the artifacts present in compressed ultrasound images.To achieve this, a diverse dataset comprising various types of breast ultrasound imagesmalignant, benign, and normalis utilized.Experimental outcomes indicate that the proposed method surpasses the denoising CNN in mitigating artifacts in compressed ultrasound images.This improved performance is particularly evident in the most compressed images, which are of significant interest.This research underscores the potential of deploying deep-learning techniques to enhance the quality of compressed medical images, thereby facilitating more accurate and efficient remote diagnoses.
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 | 0.000 | 0.001 |
| 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 | 0.000 | 0.001 |
| 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 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".