Developing an artificial intelligence diagnostic tool for paediatric distal radius fractures, a proof of concept study.

INTRODUCTION: In the UK, 1 in 50 children sustain a fractured bone yearly, yet studies have shown that 34% of children sustaining an injury do not have a visible fracture on initial radiographs. Wrist fractures are particularly difficult to identify because the growth plate poses diagnostic challenges when interpreting radiographs. METHODS: We developed Convolutional Neural Network (CNN) image recognition software to detect fractures in radiographs from children. A consecutive data set of 5,000 radiographs of the distal radius in children aged less than 19 years from 2014 to 2019 was used to train the CNN. In addition, transfer learning from a VGG16 CNN pretrained on non-radiological images was applied to improve generalisation of the network and the classification of radiographs. Hyperparameter tuning techniques were used to compare the model with the radiology reports that accompanied the original images to determine diagnostic test accuracy. RESULTS: The training set consisted of 2,881 radiographs with a fracture and 1,571 without; 548 radiographs were outliers. With additional augmentation, the final data set consisted of 15,498 images. The data set was randomly split into three subsets: training (70%), validation (10%) and test (20%). After training for 20 epochs, the diagnostic test accuracy was 85%. DISCUSSION: A CNN model is feasible in diagnosing paediatric wrist fractures. We demonstrated that this application could be utilised as a tool for improving diagnostic accuracy. Future work would involve developing automated treatment pathways for diagnosis, reducing unnecessary hospital visits and allowing staff redeployment to other areas.

MRI ankle and subtalar characteristics in haemochromatosis arthropathy: a case-control study.

AIM: To examine the magnetic resonance imaging (MRI) features of the ankle and subtalar joints that might distinguish genetic haemochromatosis (GH). MATERIALS AND METHODS: The present study was a retrospective case-control study comparing 30 MRI studies of GH patients with ankle or subtalar arthropathy with 30 matched controls with ankle pain. Anonymised images were scored using a semi-quantative tool adapted from the MRI osteoarthritis knee score. Scores were generated for bone marrow lesions size, number, and distinguishing the proportion of each lesion consisting of subchondral cyst versus oedema. Articular cartilage loss and osteophytes were documented. The primary comparator was bone marrow lesion size. Paired Student's t-test and the chi-squared test were utilised to compare outcomes. RESULTS: Bone marrow lesion/cyst size and number, presence and extent of full-thickness cartilage loss, and osteophyte scores were significantly higher in ankle joints of GH cases (p<0.01). In the middle subtalar articulation, there were significantly higher scores for full-thickness cartilage loss and extent and osteophytes in GH cases (p<0.05). There were no significant differences in the posterior subtalar articulation. CONCLUSION: The finding of both numerous and large cysts on ankle MRI should raise suspicion of GH. Other MRI features of potential diagnostic value include large osteophytes and the presence of extensive full-thickness cartilage loss in the ankle joint and middle subtalar articulation.

SPECT-CT in total hip arthroplasty.

Single photon emission computed tomography combined with computed tomography (SPECT-CT), which combines functional and anatomical imaging, provides superior spatial localization to plain radiography and is more tolerant to metallic artefacts than conventional imaging such as magnetic resonance imaging (MRI). It is increasingly used in musculoskeletal imaging to enable accurate anatomical localization of increased tracer uptake, and is particularly useful in assessing metal prosthesis and the surrounding bone following total hip arthroplasty (THA). In addition to detecting complications of THA, SPECT-CT enables multiplanar reconstruction and manipulation of imaging data, which may aid surgical planning. SPECT-CT is an important adjunct to conventional imaging in the management of post-THA complications. It is vital that radiologists are able to identify the specific features of different complications and use this novel imaging technique to guide management. In this article, the use of SPECT-CT to follow post-THA complications, including aseptic loosening, periprosthetic infection, histiocytic reactions, periprosthetic fractures, polyethylene wear, and pseudotumour formation, will be reviewed.