Abstract. Introduction. Precision health aims to develop personalised and proactive strategies for predicting, preventing, and treating complex diseases such as osteoarthritis (OA), a degenerative joint disease affecting over 300 million people worldwide. Due to OA heterogeneity, which makes developing effective treatments challenging, identifying patients at risk for accelerated disease progression is essential for efficient clinical trial design and new treatment target discovery and development. Objectives. This study aims to create a trustworthy and interpretable precision health tool that predicts rapid knee OA progression based on baseline patient characteristics using an advanced automated machine learning (autoML) framework, “Autoprognosis 2.0”. Methods. All available 2-year follow-up periods of 600 patients from the FNIH OA Biomarker Consortium were analysed using “Autoprognosis 2.0” in two separate approaches, with distinct definitions of clinical outcomes: multi-class predictions (categorising patients into non-progressors, pain-only progressors, radiographic-only progressors, and both pain and radiographic progressors) and binary predictions (categorising patients into non-progressors and progressors). Models were developed using a training set of 1352 instances and all available variables (including clinical, X-ray, MRI, and biochemical features), and validated through both stratified 10-fold cross-validation and hold-out validation on a testing set of 339 instances. Model performance was assessed using multiple evaluation metrics, such as AUC-ROC, AUC-PRC, F1-score, precision, and recall. Additionally, interpretability analyses were carried out to identify important predictors of rapid disease progression. Results. Our final models yielded high accuracy scores for both multi-class predictions (AUC-ROC: 0.858, 95% CI: 0.856–0.860; AUC-PRC: 0.675, 95% CI: 0.671–0.679; F1-score: 0.560, 95% CI: 0.554–0.566) and binary predictions (AUC-ROC: 0.717, 95% CI: 0.712–0.722; AUC-PRC: 0.620, 95% CI: 0.616–0.624; F1-score: 0.676, 95% CI: 0.673–0679). Important predictors of rapid disease progression included the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) scores and MRI features. Our models were further successfully validated using a hold-out dataset, which was previously omitted from model development and training (AUC-ROC: 0.877 for multi-class predictions; AUC-ROC: 0.746 for binary predictions). Additionally, accurate ML models were developed for predicting OA progression in a subgroup of patients aged 65 or younger (AUC-ROC: 0.862, 95% CI: 0.861–0.863 for multi-class predictions; AUC-ROC: 0.736, 95% CI: 0.734–0.738 for binary predictions). Conclusions. This study presents a reliable and interpretable precision health tool for predicting rapid knee OA progression using “Autoprognosis 2.0”. Our models provide accurate predictions and offer insights into important predictors of rapid disease progression. Furthermore, the transparency and interpretability of our methods may facilitate their acceptance by clinicians and patients, enabling effective utilisation in clinical practice. Future work should focus on refining these models by increasing the sample size, integrating additional features, and using independent datasets for external validation. Declaration of Interest. (b) declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported:I declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research project

Precision health aims to develop personalised and proactive strategies for predicting, preventing, and treating complex diseases such as osteoarthritis (OA). Due to OA heterogeneity, which makes developing effective treatments challenging, identifying patients at risk for accelerated disease progression is essential for efficient clinical trial design and new treatment target discovery and development. To create a reliable and interpretable precision health tool that predicts rapid knee OA progression over a 2-year period from baseline patient characteristics using an advanced automated machine learning (autoML) framework, “Autoprognosis 2.0”. All available 2-year follow-up periods of 600 patients from the FNIH OA Biomarker Consortium were analysed using “Autoprognosis 2.0” in two separate approaches, with distinct definitions of clinical outcomes: multi-class predictions (categorising disease progression into pain and/or radiographic progression) and binary predictions. Models were developed using a training set of 1352 instances and all available variables (including clinical, X-ray, MRI, and biochemical features), and validated through both stratified 10-fold cross-validation and hold-out validation on a testing set of 339 instances. Model performance was assessed using multiple evaluation metrics. Interpretability analyses were carried out to identify important predictors of progression. Our final models yielded higher accuracy scores for multi-class predictions (AUC-ROC: 0.858, 95% CI: 0.856-0.860) compared to binary predictions (AUC-ROC: 0.717, 95% CI: 0.712-0.722). Important predictors of rapid disease progression included WOMAC scores and MRI features. Additionally, accurate ML models were developed for predicting OA progression in a subgroup of patients aged 65 or younger. This study presents a reliable and interpretable precision health tool for predicting rapid knee OA progression. Our models provide accurate predictions and, importantly, allow specific predictors of rapid disease progression to be identified. Furthermore, the transparency and explainability of our methods may facilitate their acceptance by clinicians and patients, enabling effective translation to clinical practice

MicroRNA (miR) delivery to regulate chronic inflammation hold extraordinary promise, with new therapeutic possibilities emanating from their ability to fine-tune multiple target gene regulation pathways which is an important factor in controlling aberrant inflammatory reactions in complex multifactorial disease. However, several hurdles have prevented advancements in miR-based therapies. These include off-target effects of miRs, limited trafficking, and inefficient delivery. We propose a magnetically guided nanocarrier to transport therapeutically relevant miRs to assist self- resolving inflammation processes at injury sites and reduce the impact of chronic inflammation- related diseases such as tendinopathies. The high prevalence, significant socio-economic burden and increasing recognition of dysregulated immune mediated pathways in tendon disease provide a compelling rationale for exploring inflammation-targeting strategies as novel treatments in this condition. By combining cationic polymers, miR species (e.g., miR 29a, miR155 antagonist), and magnetic nanoparticles in the form of magnetoplexes with highly efficient magnetofection procedures, we developed inexpensive, easy-to-fabricate, and biocompatible systems with competent miR-binding and fast cellular uptake into different types of human cells, namely macrophages and tendon-derived cells. The system was shown to be cell-compatible and to successfully modulate the expression and production of inflammatory markers in tendon cells, with evidence of functional pro-healing changes in immune cell phenotypes. Hence, magnetoplexes represent a simple, safe, and non-viral nanoplatform that enables contactless miR delivery and high- precision control to reprogram cell profiles toward improved pro-regenerative environments. Acknowledgements: ERC CoG MagTendon No.772817; FCT Doctoral Grant SFRD/BD/144816/2019, and TERM. RES Hub (Norte-01-0145-FEDER-022190)

DVC is a novel full-field and contactless measurement technique for calculating displacements and strains inside bones (Grassi and Isaksson 2015) through the comparison of 3D reconstructions (CT, micro-CT, MRI, etc.) from unloaded and loaded samples. Recent in zero-strain tests to estimate the measurement precision by applying a known state of strain (Palanca, Tozzi et al. 2015) suggested that DVC is suitable to identify regions where bone tissue is yielded (i.e. subjected to high strains). Conversely to reliably measure strain in the physiological range a severe compromise with spatial resolution is necessary (Dall'Ara, Barber et al. 2014, Palanca, Tozzi et al. 2015). In order to use DVC to explore the relationship between the local physiological strain and bone microarchitecture, an error lower than 200 microstrain (an order of magnitude lower than the mean strain) and a spatial resolution of the strain measurement lower than 100 μm is required. The aim of this work is to define if, and to what extend, high-quality images obtained by synchrotron radiation micro computed tomography (SR-μCT) improve the precision of a global DVC approach. Cylindrical specimens of cortical and trabecular bone were extracted from a fresh bovine femur and embedded in acrylic resin. Both samples were scanned twice without any repositioning (‘repeated scantest’) at beamline l13–2 of Diamond Light Source (Oxford, UK). 4000 projections of 53 ms exposure were collected via fly-scanning with a CdWO. 4. scintillator-coupled pco.edge 5.5 detector with 4× magnification and an effective pixel size of 1.6μm. Strains were evaluated using a global DVC approach (ShIRT-FE) in two cubic volumes of interest (VOI) of 1,000 voxels in side length, for each specimen, exploring a DVC spatial resolution from 16 to 498 μm. The precision of measurements was evaluated extracting a similar indicator to (Liu and Morgan 2007). Precision improved with decreasing spatial resolution, confirming a trend similar to that obtained with ‘laboratory source’ μCT on similar specimens (Palanca, Tozzi et al. 2015). To obtain a precision of better than 200 microstrains the cortical and trabecular samples required spatial resolutions of 41 and 80 μm respectively. Comparing these results to those of previous studies, where similar specimens were scanned with ‘laboratory source’ μCT (effective voxel size of the order of ten μm) the errors were vastly reduced (approximately one order of magnitude). In fact, in order to obtain a precision of better than 200 microstrain, spatial resolutions of 550 (cortical) and 480 (trabecular) μm were needed (Dall'Ara, Barber et al. 2014). This work showed that using high-quality tomograms obtained by synchrotron radiation μCT decreases the measurement uncertainties of a global DVC approach with respect to those obtained with laboratory source μCT. DVC could therefore be used with μCT data to evaluate displacement and strain in the physiological range with remarkable spatial resolution

Magnetic resonance imaging (MRI) is a useful diagnostic tool in evaluating meniscus pathology in the knee. Data from available literature suggests sensitivity and specificity rates around 90% when compared to the gold standard findings at knee arthroscopy. We sought to evaluate the sensitivity, specificity and precision rate (positive predictive value) of MRI at diagnosing meniscus tears within our unit. A retrospective audit of a total of 79 MRI reports and arthroscopic findings spanning a one year period was carried out. There were 66 positive MRI reports and 13 negative reports. There were 6 false positives 4 false negatives when compared to arthroscopic findings. The sensitivity of MRI for detecting meniscus tears was 93.7% with 60 out of 64 tears detected. All 4 false negatives also had at least grade III osteoarthritic changes at arthroscopy. Specificity was rather low at 60% with MRI reporting 6 tears (false positives) out of 15 patients who had no tears found at arthroscopy. The positive predictive value (precision rate) of MRI detecting tears was 90.9%. This data shows that MRI in our unit has a comparable high sensitivity to that in various literature making it a useful tool at ruling out disease with a negative result in the clinical setting. A more useful parameter in the clinical setting is its high precision rate when faced with a positive result. However, its specificity is much lower than that in most published data. A total of 6 tears on MRI turned out not to be on arthroscopy meaning patients could have been subjected to an avoidable invasive procedure in the absence of any other indication. This highlights the importance of obtaining reports from experienced musculoskeletal radiologists and the need for surgeons to review MRI images and match them to clinical information prior to subjecting patients to surgery

Our aim was to determine the precision of the measurements of bone mineral density (BMD) by dual-energy x-ray absorptiometry in the proximal femur before and after implantation of an uncemented implant, with particular regard to the significance of retro- and prospective studies. We examined 60 patients to determine the difference in preoperative BMD between osteoarthritic and healthy hips. The results showed a preoperative BMD of the affected hip which was lower by a mean of 4% and by a maximum of 9% compared with the opposite side. In addition, measurements were made in the operated hip before and at ten days after operation to determine the effect of the implantation of an uncemented custom-made femoral stem. The mean increase in the BMD was 8% and the maximum was 24%. Previous retrospective studies have reported a marked loss of BMD on the operated side. The precision of double measurements using a special foot jig showed a modified coefficient of variation of 0.6% for the non-operated side in 15 patients and of 0.6% for the operated femur in 20 patients. The effect of rotation on the precision of the measurements after implantation of an uncemented femoral stem was determined in ten explanted femora and for the operated side in ten patients at 10° rotation and in 20 patients at 30° rotation. Rotation within 30° influenced the precision in studies in vivo and in vitro by a mean of 3% and in single cases in up to 60%. Precise prediction of the degree of loss of BMD is thus only possible in prospective cross-sectional measurements, since the effect of the difference in preoperative BMD, as well as the apparent increase in BMD after implantation of an uncemented stem, is not known from retrospective studies. The DEXA method is a reliable procedure for determining periprosthetic BMD when positioning and rotation are strictly controlled

Background. Trust in the validity of a measurement tool is critical to its function in both clinical and educational settings. Acetabular cup malposition within total hip arthroplasty (THA) can lead to increased dislocation rates, impingement and increased wear as a result of edge loading. We have developed a THA simulator incorporating a foam/Sawbone pelvis model with a modified Microsoft HoloLens® augmented reality (AR) headset. We aimed to measure the trueness, precision, reliability and reproducibility of this platform for translating spatial measurements of acetabular cup orientation to angular values before developing it as a training tool. Methods. A MicronTracker® stereoscopic camera was integrated onto a HoloLens® AR system. Trueness and precision values were obtained through comparison of the AR system measurements to a gold-standard motion capture system”s (OptiTrack®) measurements for acetabular cup orientation on a benchtop trainer, in six clinically relevant pairs of anteversion and inclination angles. Four surgeons performed these six orientations, and repeated each orientation twice. Pearson”s coefficients and Bland-Altman plots were computed to assess correlation and agreement between the AR and Motion Capture systems. Intraclass correlation coefficients (ICC) were calculated to evaluate the degree of repeatability and reproducibility of the AR system by comparing repeated tasks and between surgeons, respectively. Results. The trueness of the AR system was 0.24° (95% CI limit 0.92°) for inclination and 0.90° (95% CI limit 1.8°) for anteversion. Precision was 0.46° for inclination and 0.91° for anteversion. There was significant correlation between the two methods for both inclination (r = 0.996, p<0.001) and anteversion (r = 0.974, p<0.001). Repeatability for the AR system was 0.995 for inclination and 0.989 for anteversion. Reproducibility for the AR system was 0.999 for inclination and 0.995 for anteversion. Conclusion. Measurements obtained from the enhanced HoloLens® AR system were accurate and precise in regards to determining angular measurements of acetabular cup orientation. They exceeded those of currently used methods of cup angle determination such as CT and computer-assisted navigation. Measurements obtained were also highly repeatable and reproducible, therefore this platform is accurately validated for use in a THA training simulator

Precision medicine tailoring the patient pathway based on the risk, prognosis, and treatment response may bring benefits to the patients. To identify risk factors contributing to the early failure of treatment (development of events of interest) and when possible to change the prognosis via modifying these factors may improve the outcome and/or lower the risk of complications. There is an emerging goal to identify such parameters in total knee arthroplasty (TKA) thus lower the risk of revision surgery. The goal of this study was to identify factors explaining the risk for early revision of TKA using an artificial intelligence method appropriate for this task. We applied a patient similarity network (PSN) for the identification of risk factors associated with early reoperations (n=109, 5.8%) in patients with TKA (n=1885). Next, an algorithm based on formal concept analysis was developed to support the patient decision on how to change modifying personal characteristics with respect to the estimated probability of reoperations. The early reoperations were less frequent in women (4.4%, median time to reoperation 4.5 mo) than in men (8.2%, 10 mo), reaching the highest incidence in younger men (10.9%)

Mobility plays an important role, in particular for patients with osteoporosis and after trauma surgery, both as an outcome and as treatment. Mobility is closely linked to the patient”s quality of life and exercise is a powerful additional treatment option. In order to be able to generate an evidence base to evaluate various surgical and non-surgical treatment options, objective measurements of patient mobility and exercise over a certain time period are needed. Wearables are a promising candidate, with obvious advantages compared to questionnaires and/or PROs. However, when extracting parameters with wearables, one often faces the problem of algorithms not performing well enough for special cases like slow gait speeds or impaired gait, as they typically appear in this patient group. We plan to further extend the applicability of the actibelt system (3D accelerometer, 100Hz), in particular to improve the measurement precision of real-world walking speed in slow and impaired walking. We are using a special measurement wheel including a rotating 3D accelerometer that allows to capture high quality real-world walking speed and distance measurements, and a mobile high resolution camera system. In a first block 20 patients with osteoporosis were included in the study at the Ludwigs-Maximilians-University”s Department of General, Trauma and Reconstructive Surgery in Munich, Germany and equipped with an actibelt. Patients were asked to walk as “normal” as possible, while wearing their usual apparel, in the building and outside the building. They climbed stairs and had to deal with all unexpected “stop and go” events that appear in real-world walking. Various gait parameters will be extracted from the recorded data and compared to the gold standard. We will then tune the existing algorithms as well as new algorithms (e.g. step detection based on continuous wavelet transformation) to explore potential improvements of both step detection and speed estimation algorithms. Further refinement and validation using real world data is warranted

Chronic low back pain (cLBP) is a complex, multifaceted disorder where biological, psychological, and social factors affect its onset and trajectory. Consequently, cLBP encompasses many different disease variants, with multiple patient-specific mechanisms. The goal of NIH Back Pain Consortium (BACPAC) Research Program is to develop understanding of cLBP mechanisms and to develop algorithms that optimally match specific treatments to individual patients. To accomplish this, one research activity of BACPAC is to develop theoretical models for chronic low back pain based on the current state of knowledge in the scientific community, and to interrogate the relationships implied by the theoretical models using data generated by or available to BACPAC. The models consider biopsychosocial perspectives, and encompass both peripheral (i.e. low back) and central (i.e. spinal and supra-spinal) factors as well as proposed mechanisms of action of cLBP treatments. However, absent explanations, models/algorithms may fall short of regulatory requirements and clinician expectations, and ultimately may not be embraced by physicians and patients. To address this, BACPAC is developing a clinical utility roadmap (CUR) to clarify how models will be used in practice for selecting optimal treatments, monitoring response to treatment, and reducing health care utilization. This presentation will review the goals of BACPAC and how theoretical models and CUR are being used to support computational knowledge networks to integrate data from deeply phenotyped cLBP patients.

Our aim was to determine whether tantalum markers improved the accuracy and/or precision of methods for the measurement of migration in total hip replacement based on conventional measurements without mathematical correction of the data, and with Ein Bild Roentgen Analyse – Femoral Component Analysis (EBRA-FCA) which allows a computerised correction. Three observers independently analysed 13 series of roentgen-stereophotogrammetric-analysis (RSA)-compatible radiographs (88). Data were obtained from conventional measurements, EBRA-FCA and the RSA method and all the results were compared with the RSA data. Radiological evaluation was also used to quantify in how many radiographs the intraosseous position of the bone markers had been simulated. The results showed that tantalum markers improve reliability whereas they do not affect accuracy for conventional measurements and for EBRA-FCA. Because of the danger of third-body wear their implantation should be avoided unless they are an integral part of the method

Introduction. Precision error (PE) in Dual Energy X-Ray Absorptiometry (DXA) is important for accurate monitoring of changes in Bone-Mineral-Density (BMD). It has been demonstrated that BMD PE increases with increasing BMI. In vivo PE for the Trabecular-Bone-Score (TBS) has not been reported. This study aimed to evaluate the short-term PE (STPE)) of BMD and TBS and to investigate the effect of obesity on DXA PE. Method. DXA lumbar spine scans (L1–L4) were performed using GE Lunar Prodigy. STPE was measured in 91 women (Group A) at a single visit by duplicating scans with repositioning in-between. PE was calculated as the percentage coefficient of variation (%CV). Group A was sub-divided into four groups based on BMI (A.1. <25kg/m2, A.2. 25–29.9kg/m2, A.3. 30–35kg/m2 and A.4. >35kg/m2) to assess the effect of obesity on STPE. Abnormally different vertebrae were excluded from the analysis in accordance with The International Society for Clinical Densitometry (ISCD) recommendations. Results. The Group A STPE was 1.26 % for BMD and 2.04% for TBS. Short-term PE for BMD and TBS respectively in the BMI subgroups was: A.1. 1.07% and 1.82%, A.2. 1.34% and 2.26%, A.3. 1.25% and 2.35%, A.4. 1.68% and 1.82%. Conclusion. The results show that STPE is higher for TBS than for BMD. Short-term PE for both BMD and TBS are adversely affected by increasing BMI but this effect is mitigated in the highest BMI category where use of the ‘thick’ scanning mode improves signal to noise ratio

Weight-bearing is a known stimulus for bone remodelling and a reduction in weight-bearing is associated with reduced bone mineral density (BMD) in affected limbs post lower limb fracture. This study investigated short and long-term precision of a method for measuring relative left/right weight-bearing using two sets of identical calibrated scales. The effect of imbalance on BMD at the hip and on lower limb lean tissue mass (LLTM) was also assessed. 46 postmenopausal women, with no history of leg or ankle fracture, were measured three times whilst standing astride two scales (Seca, Germany). 34 of the participants were re-measured after 6 months by the same method. Bilateral hip and total body dual x-ray absorptiometry measurements were performed using a GE Lunar Prodigy (Bedford, MA). Precision errors in weight-bearing measures were calculated using the root mean square coefficient of variation (RMSCV%). The correlations at the first visit between left/right differences in weight-bearing and differences in BMD and LLTM were calculated. The short-term RMSCV% for left and right weights were 4.20% and 4.25% respectively and the long-term RMSCV% were 6.91% and 6.90%. Differences in left/right weight-bearing ranged from 0 to 24% (SD 8.63%) at visit 1 and 0 to 30% (SD 10.71%) at visit 2. Using data from visit 1, the relationship between hip BMD differences and left/right weight-bearing differences were investigated, with no significant correlations found. However, a weak, but statistically significant correlation of r=0.35 (p=0.02) was found for differences in LLTM and left/right weight-bearing differences. In conclusion, left/right weight-bearing measured using two scales is a precise method for evaluating differences in weight-bearing in the short and long-term. Differences in left/right weight-bearing in this population varied by up to 30%. Participants showed a high degree of consistency in their long-term balance in a natural standing posture. Inequalities in left/right weight-bearing did not correlate significantly with BMD at the hip, but demonstrated a weak but statistically significant correlation with lean tissue mass

The heat produced by drills, saws and PMMA cement in the handling of bone can cause thermal necrosis. Thermal necrosis could be a factor in the formation of a fibrous tissue membrane and impaired bony ingrowth into porous prostheses. This has been proposed to lead to non-union of osteotomies and fractures, the failure of the bone-cement interface and the failure of resurfacing arthroplasty. We compared three proprietary blades (the De Soutter, the Stryker Dual Cut and the Stryker Precision) in an in-vitro setting with porcine tibiae, using thermocouples embedded in the bone below the cutting surface to measure the increases in bone temperature. There was a significant (p=0.001) difference in the change in temperature (δT) between the blade types. The mean increase in temperature was highest for the De Soutter, 2.84°C (SD: 1.83°C, range 0.48°C to 9.30°C); mean δT was 1.81°C (SD: 1.00°C, range 0.18°C to 4.85°C) for the Precision and 1.68°C (SD: 0.95°C, range 0.24°C to 5.67°C). Performing paired tests, there was no significant difference in δT between the Precision and Dual Cut blades (p=0.340), but both these blades had significantly (p=0.003 for Precision vs De Soutter, p<0.001 for Dual Cut vs De Soutter) lower values for δT than the Dual Cut

Chronic inflammatory events have been associated to almost every chronic disease, including cardiovascular-, neurodegenerative- and autoimmune- diseases, cancer, and host-implant rejection. Given the toll of chronic inflammation in healthcare and socioeconomical costs developing strategies to resolve and control chronic states of inflammation remain a priority for the significant benefit of patients. Macrophages (Mφ) hold a central role both in the initiation and resolution of inflammatory events, assuming different functional profiles. The outstanding features of Mφ counting with the easy access to tissues, and the extended networking make Mφ excellent candidates for precision therapy. Moreover, sophisticated macrophage-oriented systems could offer innovative immune-regulatory alternatives to effectively regulate chronic environments that traditional pharmacological agents cannot provide. We propose magnetically assisted systems for balancing Mφ functions at the injury site. This platform combines polymers, inflammatory miRNA antagonists and magnetically responsive nanoparticles to stimulate Mφ functions towards pro-regenerative phenotypes. Strategies with magnetically assisted systems include contactless presentation of immune-modulatory molecules, cell internalization of regulatory agents for functional programming via magnetofection, and multiple payload delivery and release. Overall, Mφ-oriented systems stimulated pro-regenerative functions of Mφ supporting magnetically assisted theranostic nanoplatforms for precision therapies, envisioning safer and more effective control over the distribution of sensitive nanotherapeutics for the treatments of chronical inflammatory conditions. Acknowledgements: ERC CoG MagTendon No.772817; FCT Doctoral Grant SFRD/BD/144816/2019, and TERM. RES Hub (Norte-01-0145-FEDER-022190)

Introduction. 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”. Method. 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 (Arbeidsgmeischaft für Osteosynthesefragen Foundation / Orthopaedic Trauma Association; AO/OTA), on which the models were tested. Result. Detection models without pre-training on the large datasets were the least precise when tested on the small distal radius dataset. The model with the best accuracy to detect and classify wrist fractures was the YOLOv8 model pretrained on the GRAZPEDWRI-DX fracture detection dataset (mean average precision at intersection over union of 50=59.7%). This model showed up to 33.6% improved detection precision compared to the same models with no pre-training. Conclusion. Optimisation of machine-learning models can be challenging when only relatively small datasets are available. The findings of this study support the potential of transfer learning from large datasets to improve model performance in smaller datasets. This is encouraging for wider application of machine-learning technology in medical imaging evaluation, including less common orthopaedic pathologies

To detect early signs of infection infrared thermography has been suggested to provide quantitative information. Our vision is to invent a pin site infection thermographic surveillance tool for patients at home. A preliminary step to this goal is the aim of this study, to automate the process of locating the pin and detecting the pin sites in thermal images efficiently, exactly, and reliably for extracting pin site temperatures. A total of 1708 pin sites was investigated with Thermography and augmented by 9 different methods in to totally 10.409 images. The dataset was divided into a training set (n=8325), a validation set (n=1040), and a test set (n=1044) of images. The Pin Detection Model (PDM) was developed as follows: A You Only Look Once (YOLOv5) based object detection model with a Complete Detection Intersection over Union (CDIoU), it was pre-trained and finetuned by the through transfer learning. The basic performance of the YOLOv5 with CDIoU model was compared with other conventional models (FCOS and YOLOv4) for deep and transition learning to improve performance and precision. Maximum Temperature Extraction (MTE) Based on Region of Interest (ROI) for all pin sites was generated by the model. Inference of MTE using PDM with infected and un-infected datasets was investigated. An automatic tool that can identify and annotate pin sites on conventional images using bounding boxes was established. The bounding box was transferred to the infrared image. The PMD algorithm was built on YOLOv5 with CDIoU and has a precision of 0.976. The model offers the pin site detection in 1.8 milliseconds. The thermal data from ROI at the pin site was automatically extracted. These results enable automatic pin site annotation on thermography. The model tracks the correlation between temperature and infection from the detected pin sites and demonstrates it is a promising tool for automatic pin site detection and maximum temperature extraction for further infection studies. Our work for automatic pin site annotation on thermography paves the way for future research on infection assessment using thermography

Non-linear methods in statistical shape analysis have become increasingly important in orthopedic research as they allow for more accurate and robust analysis of complex shape data such as articulated joints, bony defects and cartilage loss. These methods involve the use of non-linear transformations to describe shapes, rather than the traditional linear approaches, and have been shown to improve the precision and sensitivity of shape analysis in a variety of applications. In orthopedic research, non-linear methods have been used to study a range of topics, including the analysis of bone shape and structure in relation to osteoarthritis, the assessment of joint deformities and their impact on joint function, and the prediction of patient outcomes following surgical interventions. Overall, the use of non-linear methods in statistical shape analysis has the potential to advance our understanding of the relationship between shape and function in the musculoskeletal system and improve the diagnosis and treatment of orthopedic conditions

Primary bone tumors are rare, complex and highly heterogeneous. Its diagnostic and treatment are a challenge for the multidisciplinary team. Developments on tumor biomarkers, immunohistochemistry, histology, molecular, bioinformatics, and genetics are fundamental for an early diagnosis and identification of prognostic factors. The personalized medicine allows an effective patient tailored treatment. The bone biopsy is essential for diagnosis. Treatment may include systemic therapy and local therapy. Frequently, a limb salvage surgery includes wide resection and reconstruction with endoprosthesis, biological or composites. The risk for local recurrence and distant metastases depends on the primary tumor and treatment response. Cancer patients are living longer and bone metastases are increasing. Bone is the third most frequently location for distant lesions. Bone metastases are associated to pain, pathological fractures, functional impairment, and neurological deficits. It impacts survival and patient quality of life. The treatment of metastatic disease is a challenge due to its complexity and heterogeneity, vascularization, reduced size and limited access. It requires a multidisciplinary treatment and depending on different factors it is palliative or curative-like treatment. For multiple bone metastases it is important to relief pain and increases function in order to provide the best quality of life and expect to prolong survival. Advances in nanotechnology, bioinformatics, and genomics, will increase biomarkers for early detection, prognosis, and targeted treatment effectiveness. We are taking the leap forward in precision medicine and personalized care

The computational modelling and 3D technology are finding more and more applications in the medical field. Orthopedic surgery is one of the specialties that can benefit the most from this solution. Three case reports drawn from the experience of the authors’ Orthopedic Clinic are illustraded to highlight the benefits of applying this technology. Drawing on the extensive experience gained within the authors’ Operating Unit, three cases regarding different body segments have been selected to prove the importance of 3D technology in preoperative planning and during the surgery. A sternal transplant by allograft from a cryopreserved cadaver, the realization of a custom made implant of the glenoid component in a two-stage revision of a reverse shoulder arthroplasty, and a case of revision on a hip prosthesis with acetabular bone loss (Paprosky 3B) treated with custom system. In all cases the surgery was planned using 3D processing software and models of the affected bone segments, printed by 3D printer, and based on CT scans of the patients. The surgical implant was managed with dedicated instruments. The use of 3D technology can improve the results of orthopedic surgery in many ways: by optimizing the outcomes of the operation as it allows a preliminary study of the bone loss and an evalutation of feasibility of the surgery, it improves the precision of the positioning of the implant, especially in the context of severe deformity and bone loss, and it reduces the operating time; by improving surgeon training; by increasing patient involvement in decision making and informed consent. 3D technology, by offering targeted and customized solutions, is a valid tool to obtain the tailored care that every patient needs and deserves, also providing the surgeon with an important help in cases of great complexity