Aims

The purpose of this study was to develop a convolutional neural network (CNN) for fracture detection, classification, and identification of greater tuberosity displacement ≥ 1 cm, neck-shaft angle (NSA) ≤ 100°, shaft translation, and articular fracture involvement, on plain radiographs.

Distal radius fractures (DRFs) are one of the most common types of fracture and one which is often treated surgically. Standard X-rays are obtained for DRFs, and in most cases that have an intra-articular component, a routine CT is also performed. However, it is estimated that CT is only required in 20% of cases and therefore routine CT's results in the overutilisation of resources burdening radiology and emergency departments. In this study, we explore the feasibility of using deep learning to differentiate intra- and extra-articular DRFs automatically and help streamline which fractures require a CT. Retrospectively x-ray images were retrieved from 615 DRF patients who were treated with an ORIF at the Royal Brisbane and Women's Hospital. The images were classified into AO Type A, B or C fractures by three training registrars supervised by a consultant. Deep learning was utilised in a two-stage process: 1) localise and focus the region of interest around the wrist using the YOLOv5 object detection network and 2) classify the fracture using a EfficientNet-B3 network to differentiate intra- and extra-articular fractures. The distal radius region of interest (ROI) detection stage using the ensemble model of YOLO networks detected all ROIs on the test set with no false positives. The average intersection over union between the YOLO detections and the ROI ground truth was Error! Digit expected.. The DRF classification stage using the EfficientNet-B3 ensemble achieved an area under the receiver operating characteristic curve of 0.82 for differentiating intra-articular fractures. The proposed DRF classification framework using ensemble models of YOLO and EfficientNet achieved satisfactory performance in intra- and extra-articular fracture classification. This work demonstrates the potential in automatic fracture characterization using deep learning and can serve to streamline decision making for axial imaging helping to reduce unnecessary CT scans

Background: Complex fractures of the tibial plateau can be difficult to characterize on plain radiographs and two-dimensional computed tomography scans. We tested the hypothesis that three-dimensional computed tomography reconstructions improve the reliability of tibial plateau fracture characterization and classification. Methods: Forty-five consecutive intra-articular fractures of the tibial plateau were evaluated by six independent observers for the presence of six fracture characteristics that are not specifically included in currently used classification schemes:. posteromedial shear fracture;. coronal plane fracture;. lateral condylar impaction;. medial condylar impaction;. tibial spine involvement;. separation of tibial tubercle necessitating anteroposterior lag screw fixation. In addition, fractures were classified according to the AO/OTA Comprehensive Classification of Fractures, the Schatzker classification system and the Hohl and Moore system. Two rounds of evaluation were performed and then compared. First, a combination of plain radiographs and two-dimensional computed tomography scans (2D) were evaluated, and then, four weeks later, a combination of radiographs, two-dimensional computed tomography scans, and three-dimensional reconstructions of computed tomography scans (3D) were assessed. Results: Interobserver agreement improved for all classification systems after the addition of three-dimensional reconstructions (AO/OTA κ2D = 0.536 versus κ3D = 0.545; Schatzker κ2D = 0.545 versus κ3D = 0.596; Hohl and Moore κ2D = 0.668 versus κ3D = 0.746). Three-dimensional computed tomography reconstructions also improved the average intraobserver reliability for all fracture characteristics, from κ2D = 0.624 (substantial agreement) to κ3D = 0.687 (substantial agreement). The addition of three-dimensional images had limited infiuence on the average interobserver reliability for the recognition of specific fracture characteristics (κ2D = 0.488 versus κ3D = 0.485, both moderate agreement). Three-dimensional computed tomography images improved interobserver reliability for the recognition of coronal plane fractures from fair (κ2D = 0.398) to moderate (κ3D = 0.418) but this difference was not statistically significant. Conclusions: Three-dimensional computed tomography is helpful for;. individual orthopaedic surgeons for preoperative planning (improves intraobserver reliability for the recognition of fracture characteristics), and for. comparison of clinical outcomes in the orthopaedic literature (improves interobserver reliability of classification systems)

Aims

The prevalence of ipsilateral total hip arthroplasty (THA) and total knee arthroplasty (TKA) is rising in concert with life expectancy, putting more patients at risk for interprosthetic femur fractures (IPFFs). Our study aimed to assess treatment methodologies, implant survivorship, and IPFF clinical outcomes.

Combined techniques of fracture mechanics and confocal Raman microprobe spectroscopy were applied to characterize, after increasing periods of environmental exposure, bulk and surface toughness values in an advanced alumina/zirconia composite. This material is used in joint prostheses (BIOLOX. ®. delta femoral heads, manufactured by CeramTec AG). Besides conventional fracture mechanics characterizations, including different types of fracture toughness test, Raman and fluorescence microprobe spectroscopy provided a microscopic insight into the effect of environmentally assisted processes of zirconia phase transformation at the surface on the fracture toughness of the material. We have found that the tetragonal-to-monoclinic polymorphic transformation occurs in the studied composite material as a consequence of an environmentally assisted process, although severe exposures are needed for to obtain a substantial increase of the monoclinic content. Such severe exposures in vitro correspond to exposures in human body of several lifetimes. The effect of an exposure of 10 h in autoclave (in vitro accelerated test) was carefully examined, because this span of time corresponds:. to the period of time recommended for testing in vitro by ISO standard; and,. to approximately the lifetime expected for a prosthesis in vivo. The main experimental outcomes of confocal Raman spectroscopy and fracture mechanics assessments can be summarized as follows:. the crack-tip toughness level measured in the as-received material was comprehensive of a tangible contribution by transformation toughening, thus showing that phase transformation in the zirconia dispersoids plays a positive role in the toughening behavior of the material;. after the material was environmentally aged in vitro for periods of the order of hundreds of hours, its surface toughness was reduced by about one-third; but, even in the case of such a severe exposure, the surface toughness of the composite was at least the same as that of monolithic alumina;. the observed decrease of fracture toughness by about one-third was limited to the very surface of the material (i.e., to a layer of the order of the tens of microns) and did not affect the bulk fracture behavior of the composite. It appears that concerns arising from the brittleness of alumina-based materials and, thus, from their vulnerability to fracture due to unexpected load situation, can be successfully counteracted by properly adding a dispersion of zirconia particles to the alumina matrix. Such an addition enables the obtainment of a composite material, whose fracture resistance is greatly enhanced by a crack-shielding effect due to phase-transformation processes occurring in the zirconia dispersoids