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.

Intramedullary nailing is the standard fixation method for displaced diaphyseal fractures of the tibia in adults. Anecdotal clinical evidence indicates that current nail designs do not fit optimally for Asian patients. This study aimed to develop a method to quantitatively assess the fitting of two nail designs for Asian tibiae.

We used 3D models of two different tibial nail designs (ETN (Expert Tibia Nail) and ETN-Proximal-Bend, Synthes), and 20 CT-based 3D cortex models of Japanese cadaver tibiae. The nail models were positioned inside the medullary cavity of the intact bone models. The anatomical fitting between nail and bone was assessed by the extent of the nail protrusion from the medullary cavity into the cortical bone, which in a real bone would lead to axial malalignments of the main fragments. The fitting was quantified in terms of the total surface area, and the maximal distance of nail protrusion.

In all 20 bone models, the total area of the nail protruding from the medullary cavity was smaller for the ETN-Proximal-Bend (average 540 mm2) compared to the ETN (average 1044 mm2). Also, the maximal distance of the nail protruding from the medullary cavity was smaller for the ETN-Proximal-Bend (average 1.2 mm) compared to the ETN (average 2.7 mm). The differences were statistically significant (p < 0.05) for both the total surface area and the maximal distance measurements. For all bone models, the nail protrusion occurred on the posterior side in the middle third of the tibia. For 12 bones the protrusion was slightly lateral to the centre of the shaft, for seven bones it was centred, and for one bone it was medial to the shaft. The ETN-Proximal-Bend shows a statistical significantly better intramedullary fit with less cortical protrusion than the original ETN. The expected clinical implications of an improved anatomical nail fit are fewer complications with malreduction and malalignments, a lower likelihood for fracture extension and/or new fracture creation during the nail insertion as well as an easier handling for the nail insertion.

By utilising computer graphical methods we were able to conduct a quantitative fit assessment between implanted nail and bone geometry in 3D. In addition to the application in implant design, the developed method could potentially be suitable for pre-operative planning enabling the surgeon to choose the most appropriate nail design.

Successful treatment of bone fractures requires a balance between stability, to restore functional anatomy and allow early mobilisation (and thus avoid dystrophy). The healing occurs through complex interactions of inducing, enabling and inhibitory factors. The mechanical environment (e.g. stress and strain) in/around the fracture site regulates tissue changes throughout the healing process, including the formation of a fibro-cartilaginous callus and its progressive replacement by bone. The mechanical and biological environment is controlled substantially by the selection of the fracture stabilisation method achieving either absolute stability (mostly achieved with compression plating technique) or relative stability allowing a limited amount of dynamic fracture displacement across the fracture gap. A number of treatments may be used to accomplish these conditions, ranging from splinting with a plaster cast, external fixator or an intramedullary nail to rigid internal fixation using plates affixed to the bone fragments. Fixation methods are presently selected on the basis of general guidelines, but nevertheless the optimal stability/instability remains unclear and relies heavily on the surgeon’s experience. With the recently more and more widely used locking plates the question of the optimal fixation technique and applied stability to the fracture zone especially in simple fractures have raised again.

To fill this knowledge gap, an interdisciplinary approach with in vitro and in vivo experiments seems to be essential. Analysing clinical situations and the healing course with mathematical modelling and computational simulations can further aid to understand the healing conditions in respect to stability.

This presentation will give an overview on the role of the mechanical environment in fracture healing, and demonstrating clinical examples that highlight the relevance of this research.

Introduction: With the development and popularisation of minimally invasive surgical methods and implants, it is increasingly important that available implants are pre-contoured to the specific anatomical location for which they are designed. From a clinical point of view, an anatomically well fitting plate can greatly facilitate the process of closed reduction in terms of axial and rotational alignment of the main fragments. Furthermore, such a plate may additionally protrude less and therefore minimise soft tissue irritations. Clinical practice and recent research shows that some of the current pre-contoured fracture fixation plates do not fit well for certain patient groups ^1,2. This study reports on a method for optimising the shape of a plate for a given dataset.

Methods: Forty-five 3D models of the outer bone contour of Japanese tibiae were generated from CT image data. The fit between bone models and the undersurface (bone facing) of a distal medial tibia plate was quantified using the reverse engineering software RapidForm2006 (Inus Technology Inc., Korea). An anatomical fit of the plate was defined with four criteria². The current plate shape was optimised in the virtual environment using computer graphical tools of RapidForm2006.

Results: The current plate shape achieved an anatomical fit on 13% of tibias from the dataset, whereas the modified plate achieved an anatomical fit for 67% of the bone models.

Conclusion: Computer graphical methods enable optimisation of the shape of an existing plate resulting in a significantly improved fitting for the bones of the available dataset.

Volume and density of fracture callus are important outcome parameters in fracture healing studies. These values provide an indication for the recovery of the mechanical function of the bone. Traditionally, fracture callus’ have been evaluated from radiographs, which represent 2D projections of the three-dimensional structures, therefore such an analysis can be affected by many artefacts. With the availability of Computer Tomography (CT) scanners for the evaluation of healing bones, it is now possible to perform precise, three-dimensional reconstructions of the fracture callus and therefore to evaluate true three-dimensional callus volumes and bone mineral densities. We wanted to make use of this technology in the evaluation of a study looking at the healing of a multifragmentary fracture in sheep after 4 and 8 weeks of healing time (Wullschleger et al, ANZORS, 2006). Our goal was to develop a protocol that would allow for the standardised calculation of cortical bone and callus tissue volumes with minimal user influence. Here, we report on the development of this evaluation protocol and some early results.

A clinical CT scanner was used to scan the experimental limbs, immediately after the sheep had been euthanized. Further analysis of the CT dataset was accomplished with the commercial computer software Amira. The region of interest was cropped to a 9 cm section of the bone shaft, guaranteed to comprise the entire fracture callus. Next, the cortical bone and the callus tissue were segmented by choosing appropriate threshold values for the measured grey levels. The volume of the segmented regions was then calculated by the software.

The application of this protocol to six CT scans from our experimental study resulted in average callus volumes of 12.21 ± 1.96 (standard deviation) cm2 after 4 weeks healing time and 14.28 ± 1.58 cm2 after 8 weeks healing time.

In conclusion, we demonstrated the feasibility of using CT data for a quantitative 3D analysis of callus volumes. While this technique is undoubtedly superior to the estimation of callus volumes from two-dimensional radiographs, the absolute accuracy of the results will need to be determined by comparison with histological data.

Bilateral mandibular lengthening is widely accepted during mandibular distraction osteogenesis. However, distraction osteogenesis are sometimes associated with clinical complications such as open bite deformity, lateral displacement of temporo-mandibular joint, premature consolidation and pin loosening. Although distraction osteogenesis aims to develop pure tensile strain on the regenerate tissue however, in real life situation due to differences in device orientation, materials and misalignment it is often subjected to complex stress and strain regimes.

The objective of this study was to characterise the mechanical environment (stress and strain) in the Finite Element Models (FEM) of regenerate tissue during two different device orientations:

(a) device placed parallel to the mandibular body

Furthermore, the influence of misalignment from above two idealised orientations was also investigated.

The distraction protocol in this study was similar to previous study of Loboa et al (2005). FE models were developed at four stages: end of latency, distraction day two, distraction day five and distraction day eight. At each time period a distraction of 1mm was applied to the model as it is most widely used distraction rate. In these models two primary distraction vectors were simulated; first when the device is parallel to the body of the mandible and second when the device is parallel to the axis of distraction.

Results indicate that when the device is placed parallel to the mandible the effect of distraction vector variation due to misalignment in transverse plane (±150 at an interval of 50 ; + indicate lateral and indicates medial) is symmetric and variation in the stress and strain regimes on regenerative tissue are less than 3%. However, when the device is placed parallel to axis of distraction the corresponding change is asymmetric and almost double in magnitude. The greatest differences were seen when misalignment is towards lateral side (+150). Similarly in the sagittal plane variations up to 17% were developed due to 0- 400 change in the distraction vector orientation. Thus the orientation of device which determines the distraction vector plays an important role in determining the mechanical environment around regenerative tissue. The results suggest that implications of misalignment of the device and its sensitivity from the ideal situation should be well understood during clinical planning.

With the development and popularisation of minimally invasive surgical methods and implants for fracture fixation, it is increasingly important that available implants are pre-contoured to the specific anatomical location for which they are designed. Due to differences in the bone morphology it is impossible to design single implants that are universally applicable for the entire human population. A recent study on the fit of a distal periarticular medial tibia plate to Japanese bones supported the need for shape optimisation [1]. The present study aimed to quantify and optimise the fit of the same plate for an extended dataset of Japanese tibiae.

Forty-five 3D models of the outer bone contour of Japanese tibiae were used. The average age of the specimens was 67 years with an average height of 156 cm. All bone models were considered to be within a normal range without any bony pathology. An anatomical fit of the plate was defined with four criteria [1]. The current plate shape was optimised based on the quantitative results of the plate fitting. Two different optimised plate shapes were generated.

The current plate shape achieved an anatomical fit on 13% of tibias from the dataset. Plate 1 achieved an anatomical fit for 42% and Plate 2 a fit for 67% of the bone models. If either Plate 1 or Plate 2 is used, then the anatomical fit can be increased from 13% to 82% for the same dataset. For 12 (27%) of the 45 bones both modified plate shapes were fitting.

The results for the fit of the current plate shape are comparable to findings of a similar study on the anatomical fitting of proximal tibia plates [2]. The obtained results indicate that for the available dataset no further modification is warranted for the shaft region of the modified plates. Further optimization of the distal regions of Plate 1 and Plate 2 will be possible. This study shows that in order to achieve an anatomical fit of the plate to the medial Malleolus at least two different plate shapes will be required.

Volume and density of fracture callus are important outcome parameters in fracture healing studies. These values provide an indication for the recovery of the mechanical function of the bone. Traditionally, fracture callus’ have been evaluated from radiographs, which represent 2D projections of the three-dimensional structures; therefore such an analysis can be affected by many artefacts. With the availability of Computer Tomography (CT) scanners for the evaluation of healing bones, it is now possible to perform precise, three-dimensional reconstructions of the fracture callus and therefore to evaluate true three-dimensional callus volumes and bone mineral densities. We wanted to make use of this technology in the evaluation of a study looking at the healing of a multifragmentary fracture in sheep after 4 and 8 weeks of healing time (Wullschleger et al, ANZORS, 2006). Our goal was to develop a protocol that would allow for the standardised calculation of cortical bone and callus tissue volumes with minimal user influence. Here, we report on the development of this evaluation protocol and some early results.

A clinical CT scanner was used to scan the experimental limbs, immediately after the sheep had been euthanized. Further analysis of the CT dataset was accomplished with the commercial computer software Amira. The region of interest was cropped to a 9 cm section of the bone shaft, guaranteed to comprise the entire fracture callus. Next, the cortical bone and the callus tissue were segmented by choosing appropriate threshold values for the measured grey levels. The volume of the segmented regions was then calculated by the software.

The application of this protocol to six CT scans from our experimental study resulted in average callus volumes of 12.21 ± 1.96 (standard deviation) cm2 after 4 weeks healing time and 14.28 ± 1.58 cm2 after 8 weeks healing time.

In conclusion, we demonstrated the feasibility of using CT data for a quantitative 3D analysis of callus volumes. While this technique is undoubtedly superior to the estimation of callus volumes from two-dimensional radiographs, the absolute accuracy of the results will need to be determined by comparison with histological data.