Automated Wrist Ultrasound Image Bone Enhancement and Segmentation Using Deep Learning
Bibliographic record
Abstract
Wrist fractures are among the most common upper limb injuries in children and are currently examined by X-rays. Ultrasound imaging offers a radiation-free alternative that can be used for a fast and effective evaluation by assessing the severity of suspected fractures before further referral to X-rays. This project proposes a segmentation framework based on the nnU-Net model to segment bony structures such as the epiphysis and metaphysis commonly seen in wrist ultrasound images. As a preprocessing step, we use an image enhancement technique such as Contrast-Limited Adaptive Histogram Equalization (CLAHE) and report the accuracy of segmentation with and without preprocessing.Experiments were conducted on 16,865 training and 3,822 testing ultrasound images from 74 and 18 subjects, respectively. The results show that training and testing on CLAHE-enhanced images improves segmentation performance, achieving a DICE score of 0.874 compared to 0.872 without pre-processing.Clinical relevance-This study shows the feasibility of using automatic segmentation of wrist ultrasound images acquired by lightly trained users in a pediatric emergency setting. It combines CLAHE-based image enhancement with semantic segmentation using the nnUNet-inspired model to segment the epiphysis, metaphysis, and carpal bone from wrist ultrasound images. In an emergency care setting, this approach could be integrated into an effective triage tool to assess the severity of pediatric wrist injuries and reduce the need for x-ray examination in cases with no fractures.
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 | 0.001 |
| 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 | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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".