Automated Detection of Shoulder Rotator Cuff Tendon Tears from Ultrasound Images by CNN-Autoencoder
Bibliographic record
Abstract
Ultrasound can effectively diagnose rotator cuff muscle tears in the shoulder, one of the most common musculoskeletal injuries. Expertise in performing and interpreting these scans is limited, and in some jurisdictions, it involves months-long wait times. This project aims to provide the core AI process for the assessment of shoulder injuries; specifically identifying full-thickness rotator cuff tears within minutes, rather than months. We propose a two-step approach starting with segmentation of the humeral cortex and subacromial bursa followed by the classification of tears based on these regions. Automatic segmentation using ultrasound is challenging due to speckle noise, low contrast, and artifacts. The proposed segmentation method leverages a CNN-autoencoder, that predicts the boundary contour points of the humeral cortex and subacromial bursa directly from raw images. Compared to an end-to-end black box classifier, our two-stage approach is more explainable and focuses on clinically relevant landmarks. The distinctive feature of this segmentation technique is, it predicts the segmentation contour points rather than the popular pixel-wise semantic segmentation. The study was performed on the dataset acquired from 206 patients. All the methods were trained using 10,080 images and evaluated using 2520 images. Our proposed segmentation method achieved an average Dice coefficient (DC) of 94.2% and a Hausdorff Distance (HD) of 2.8 mm, outperforming the U-Net model, which yielded 90.5% for the DC and 6.8 mm for the HD. After the segmentation, a classification network, VGG-16 achieved 81.0% accuracy (sensitivity 78.5%, specificity 76.2%) in classifying rotator cuff tendons as intact or torn from US images. AI-driven ultrasound for rotator cuff tear detection enhances early, accurate diagnosis and improves patient care, and this automated tool could be used by lightly trained users at initial point-of-care facilities like family physician clinics and emergency rooms.
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 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| 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".