With advances in artificial intelligence, the use of computer-aided detection and diagnosis in clinical imaging is gaining traction. Typically, very large datasets are required to train machine-learning models, potentially limiting use of this technology when only small datasets are available. This study investigated whether pretraining of fracture detection models on large, existing datasets could improve the performance of the model when locating and classifying wrist fractures in a small X-ray image dataset. This concept is termed “transfer learning”. Firstly, three detection models, namely, the faster region-based convolutional neural network (faster R-CNN), you only look once version eight (YOLOv8), and RetinaNet, were pretrained using the large, freely available dataset, common objects in context (COCO) (330000 images). Secondly, these models were pretrained using an open-source wrist X-ray dataset called “Graz Paediatric Wrist Digital X-rays” (GRAZPEDWRI-DX) on a (1) fracture detection dataset (20327 images) and (2) fracture location and classification dataset (14390 images). An orthopaedic surgeon classified the small available dataset of 776 distal radius X-rays (Introduction
Method
Inaccurate identification of implants on X-rays may lead to prolonged surgical duration as well as increased complexity and costs during implant removal. Deep learning models may help to address this problem, although they typically require large datasets to effectively train models in detecting and classifying objects, e.g. implants. This can limit applicability for instances when only smaller datasets are available. Transfer learning can be used to overcome this limitation by leveraging large, publicly available datasets to pre-train detection and classification models. The aim of this study was to assess the effectiveness of deep learning models in implant localisation and classification on a lower limb X-ray dataset. Firstly, detection models were evaluated on their ability to localise four categories of implants, e.g. plates, screws, pins, and intramedullary nails. Detection models (Faster R-CNN, YOLOv5, EfficientDet) were pre-trained on the large, freely available COCO dataset (330000 images). Secondly, classification models (DenseNet121, Inception V3, ResNet18, ResNet101) were evaluated on their ability to classify five types of intramedullary nails. Localisation and classification accuracy were evaluated on a smaller image dataset (204 images).Introduction
Method
Introduction
Materials and Methods
To determine the outcomes of cases converted from an external fixator to an internal fixation device in the management of limb reconstructions and deformity corrections. A retrospective review of 18 patients, that underwent a conversion procedure to internal fixation following long term external fixation use, was done. This comprised 24 limbs. Inclusion criteria: All cases of long term external fixator use converted to internal fixation over a 5 year period. Average external fixation time, pin site care, conversion timing, surgical device used as well as outcome were documented.Purpose of the study:
Method:
Is circular external fixation a safe and effective method of managing closed distal third tibia fractures. These fractures are conventionally treated with plaster casts, intramedullary nails or plate fixation. These treatment modalities have complication rates in the literature of up to 16% malunion, 12% non-union, and 17% deep infections. Retrospective review of 18 patients with closed distal third tibia fractures, with or without extension into the ankle joint, treated with circular fixator systems and minimal percutaneous internal fixation of the intra-articular fragment if required. Patients were followed up for time to union, malunion incidence as well as incidence of pin tract and deep infection. Distal third fractures which were extra articular or with simple intra articular extension were included. (AO 43 A, B1, C1, C2 + AO 42 in distal third) Patients with pilon fractures (AO 43 B2, B3 and C3) were excluded.Purpose of the study:
Description and Methods:
This study attempts to establish whether biomechanical manipulation through distraction can result in fracture union. A retrospective clinical audit of 15 patients with delayed or hypertrophic non-unions treated successfully with closed distraction in circular external fixation. Average time to union, complications and complication rates were also reviewed. Inclusion criteria: all patients with delayed or hypertrophic non-union, treated by closed distraction between 2004 and 2011.Purpose:
Method:
To determine whether HIV seropositivity is a significant risk factor for infection following open fractures of long bones, we undertook a prospective cohort-type analytical study. We obtained Ethics Committee approval and invited suitable patients (adults with open fractures of long bones) presenting to our institution to participate in the study. Written informed consent was obtained and the patient registered. Baseline blood tests were done (Hb, HIV, Albumin, CD4 count), after which treatment proceeded in the standard manner. At set intervals, patient records were accessed and information captured in a database. Patients were followed up until bony union had occurred. To date, 50 patients have been recruited to the study. Of these, 19 were followed up until bony union. All but one of these patients was male and the mean age was 34.5 years. Fractures included three femora, two humeri, two radii and 15 tibiae. There were five Gustilo-Anderson grade-I, seven grade-II, six grade-IIIa and three grade-IIIb open fractures. Three of the 19 patients tested positive for HIV (15.8%). Infection occurred in five patients (26.3%), none of whom was HIV-positive. The strongest predictor for infection was the time delay to wound inspection, with mean delays of 56.9 hours and 100.2 hours respectively in patients who did and did not develop infection. Although our study is small, it suggests that asymptomatic HIV seropositivity is not a significant risk factor for infection following open fractures of long bones. Delayed wound inspection puts patients at increased risk.
