We dive into the orthobiolocigal world of platelet rich plasma in the treatment of musculoskeletal conditions. We'll showcase some remarkable msk, and non-msk case studies, from diagnosis to treatment, all the way to healing. Some warnings for users in misguiding patients. We'll touch on the radiology setting for sonographer performed injections, and some ultrasound guided technical injection considerations for success.

Ultrasound-guided cortisone injections are very common procedures that many sonographers are involved with and widely used in musculoskeletal medicine, yet for many of us and our patients, important questions remain regarding optimal use and safety, and there is significant variation in clinical practice. This presentation reviews clinical rationale of ultrasound-guided cortisone injections and pathology, patient and technique variables which can influence clinical outcomes.

Shear wave elastography (SWE) is emerging as a valuable adjunct to conventional musculoskeletal ultrasound, offering quantitative insight into tissue stiffness and mechanical properties. This presentation explores the evolving role of SWE in modern MSK practice.

Introduction: Supraspinatus tendinopathy is a frequent cause of shoulder pain, affecting both quality of life and functional ability. Ultrasound offers benefits such as lower cost, greater accessibility, and real-time evaluation; however, its accuracy depends on the availability of a skilled operator. Using a deep learning (DL) model for automated supraspinatus tear classification and diagnosis may be beneficial in settings with limited resources. Therefore, the aim of this study was to develop and internally validate a DL-based AI model for automated detection and multi-class classification of supraspinatus tendon tears (no tear, partial-thickness tear, or full-thickness tear) on US images.
Methods: A retrospective dataset of 640 de-identified two-dimensional US images of the supraspinatus tendon was curated from a single collaborating orthopaedic centre. All cases had ground-truth labels established by radiologist reports and confirmation via arthroscopy performed by a senior orthopaedic surgeon. Images were annotated using open-source tools under expert guidance. A ResNet-based DL model was trained for simultaneous detection, classification, and (where indicated) localisation of tears. Split training validation: 80/20. Internal validation employed 4-fold cross-validation
Results: The model achieved strong performance on internal validation. Average accuracy across the four folds was 0.88. Per-class performance (overall, 4-fold) was highest for no-tear (precision 0.91, recall 0.97, F1-score 0.94) and full-thickness tears (precision 0.93, recall 0.82, F1-score 0.87), with partial-thickness tears showing more modest metrics (precision 0.81, recall 0.86, F1-score 0.83). The best single fold reached 0.91 accuracy.
Conclusions: This DL AI model demonstrates promising internal validity for automated supraspinatus tear classification on US, and it may have the potential to serve as a triage tool in under-resourced settings using portable wireless US probes operated by non-specialists. Further external validation is required before clinical deployment.