Artificial Intelligence (AI) is becoming more powerful but is barely used to counter the growth in health care burden. AI applications to increase efficiency in orthopedics are rare. We questioned if (1) we could train machine learning (ML) algorithms, based on answers from digitalized history taking questionnaires, to predict treatment of hip osteoartritis (either conservative or surgical); (2) such an algorithm could streamline clinical consultation. Multiple ML models were trained on 600 annotated (80% training, 20% test) digital history taking questionnaires, acquired before consultation. Best performing models, based on balanced accuracy and optimized automated hyperparameter tuning, were build into our daily clinical orthopedic practice. Fifty patients with hip complaints (>45 years) were prospectively predicted and planned (partly blinded, partly unblinded) for consultation with the physician assistant (conservative) or orthopedic surgeon (operative). Tailored patient information based on the prediction was automatically sent to a smartphone app. Level of evidence: IV. Random Forest and BernoulliNB were the most accurate ML models (0.75 balanced accuracy). Treatment prediction was correct in 45 out of 50 consultations (90%), p<0.0001 (sign and binomial test). Specialized consultations where conservatively predicted patients were seen by the physician assistant and surgical patients by the orthopedic surgeon were highly appreciated and effective. Treatment strategy of hip osteoartritis based on answers from digital history taking questionnaires was accurately predicted before patients entered the hospital. This can make outpatient consultation scheduling more efficient and tailor pre-consultation patient education

Introduction and Objective. Despite the low incidence of pilon fractures among lower limb injuries, their high-impact nature presents difficulties in surgical management and recovery. Current literature includes a wide range of different management strategies, however there is no universal treatment algorithm. We aim to determine clinical outcomes in patients with open and closed pilon fractures, managed using a treatment algorithm that was applied consistently over the span of this study. Materials and Methods. This retrospective study was conducted at a single institution, including 141 pilon fractures in 135 patients, from August 2014 to January 2021. AO/OTA classification was used to classify fractures. Among closed fractures, 12 had type 43A, 18 had type 43B, 61 had type 43C. Among open fractures, 11 had type 43A, 12 had type 43B, 27 had type 43C. Open fractures were further classified with Gustilo-Anderson (GA); type 1: n=8, type 2: n=10, type 3A: n=12, type 3B: n=20. Our treatment algorithm consisted of fine wire fixator (FWF) for severely comminuted closed fractures (AO/OTA type 43C3), or open fractures with severe soft tissue injury (GA type 3). Otherwise, open reduction internal fixation (ORIF) was performed. When required, minimally invasive osteosynthesis (MIO) was performed in combination with FWF to improve joint congruency. All open fractures, and closed fractures with severe soft tissue injury (skin contusion, fracture blister, severe oedema) were initially treated with temporary ankle-spanning external fixation. For all open fracture patients, surgical debridement, soft tissue cover with a free or pedicled flap were performed. For GA types 1 and 2, this was done with ORIF in the same operating session. Those with severe soft tissue injury (GA type 3) were treated with FWF four to six weeks after soft tissue management was completed. Primary outcome was AOFAS Ankle-Hindfoot score at 3, 6 and 12-months post-treatment. Secondary outcomes include time to partial weight-bear (PWB) and full weight-bear (FWB), bone union time. All complications were recorded. Results. Mean AOFAS score 3, 6, and 12 months post-treatment for open and closed fracture patients were 44.12 and 53.99 (p=0.007), 62.38 and 67.68 (p=0.203), 78.44 and 84.06 (p=0.256), respectively. 119 of the 141 fractures healed without further intervention (84.4%). Average time to bone union was 51.46 and 36.48 weeks for open and closed fractures, respectively (p=0.019). Union took longer in closed fracture patients treated with FWF than ORIF (p=0.025). On average, open and closed fracture patients took 12.29 and 10.76 weeks to PWB (p=0.361); 24.04 and 20.31 weeks to FWB (p=0.235), respectively. Common complications for open fractures were non-union (24%), post-traumatic arthritis (16%); for closed fractures they were post-traumatic arthritis (25%), superficial infection (22%). Open fracture was a risk factor for non-union (p=0.042; OR=2.558, 95% CI 1.016–6.441), bone defect (p=0.001; OR=5.973, 95% CI 1.986–17.967), and superficial infection (p<0.001; OR=4.167, 95% CI 1.978–8.781). Conclusions. The use of a two-staged approach involving temporary external fixation followed by definitive fixation, provides a stable milieu for soft tissue recovery. FWF combined with MIO, where required for severely comminuted closed fractures, and FWF for open fractures with severe soft tissue injury, are safe methods achieving low complication rates and good functional recovery

Partial meniscectomy patients have a greater likelihood for the development of early osteoarthritis (OA). To prevent the onset of early OA, patient-specific treatment algorithms need to be created that predict patient risk to early OA after meniscectomy. The aim of this work was to identify patient-specific risk factors in partial meniscectomy patients that could potentially lead to early OA. Partial meniscectomy patients operated between 01/2017 and 12/2019 were evaluated in the study (n=317). Exclusion criteria were other pathologies or surgeries for the evaluated knee and meniscus (n = 114). Following informed consent, an online questionnaire containing demographics and the “Knee Injury and Osteoarthritis Outcome Score” (KOOS) questionnaire was sent to the patient. Based on the KOOS pain score, patients were classified into “low” (> 75) and “high” (< 75) risk patients, indicating risk to symptomatic OA. The “high risk” patients also underwent a follow-up including an MRI scan to understand whether they have developed early OA. From 203 participants, 96 patients responded to the questionnaire (116 did not respond) with 61 patients considered “low-risk” and 35 “high-risk” patients. Groups that showed a significant increased risk for OA were patients aged > 40 years, females, overweight (BMI >25 kg/m2 ≤ 30 kg/m2), and smokers (*p < 0.05). The “high-risk”-follow-up revealed a progression of early osteoarthritic cartilage changes in seven patients, with the remaining nineteen patients showing no changes in cartilage status or pain since time of operation. Additionally, eighteen patients in the high-risk group showed a varus or valgus axis deviation. Patient-specific factors for worse postoperative outcomes after partial meniscectomy and indicators for an “early OA” development were identified, providing the basis for a patient-specific treatment approach. Further analysis in a multicentre study and computational analysis of MRI scans is ongoing to develop a patient-specific treatment algorithm for meniscectomy patients

Summary. Management of metal on metal hip replacements can be accomplished with a simple algorithm including easily available metal ion levels and hip MRI with metal artifact reducing software. After revision serum metal ion levels can be expected to fall rapidly. Introduction. Metallic ion release may be related to bearing surface wear and thus serves as an indicator of the in-vivo performance of metal on metal articulations. The purpose of this prospective, controlled study was to compare new large head metal on metal hip components with established modular metal on metal and metal on polyethylene and to determine their effects on serum metal levels before and after revision. Patients & Methods. We performed a multi-surgeon, prospective, controlled trial to compare clinical, radiographic, and metal ion concentration in serum (cobalt and chrome) results across multiple devices including the Large Head ASR XL System (MoM-1), the Ultamet Advanced Modularity System (MoM-2), and as the control the Pinacle Acetabular Cup System with polyethylene liner (MoP). One hundred and fifty-one consecutive patients undergoing THA were enrolled in the study: MoM-1 n=97; MoM-2 n=22; MoP n=32. Clinical, radiographic, and venous blood assessments were performed pre-operatively, and post-operatively at 6 months, 1 year and 2 years, and after revision (1,3,6,12 months). All serum ion concentrations are reported in nmol/L. We are following metal ion levels after revision and have developed an algorithm to diagnose and manage patients with MoM THA. Results. MoM-1 patients had significantly increased average cobalt and chromium levels. Clinical scores improved after surgery in all groups and continued to improve in MoM-2 and MoP patients after 2 years but decreased slightly in the MoM-1 patients at 2 years. Average cup inclination angle did not differ significantly between the groups: MoM-1 50.2, MoM-2 47.8, and MoP 51.7. In the MoM-1 group 11 patients (11%) had significantly elevated ion levels (MoM-1 Outliers). Nine hips (9.3%) in 8 MoM-1 outlier patients required revision. Metal ion levels were not significantly different between MoM-2 and MoP groups. Metal ion levels after revision in the MoM-1 group decreased rapidly but at one year post-operatively have still not returned to an equivalent baseline comparable to the MoM-2 and MoP groups. All revisions were in the MoM-1 group. Chromium levels decreased more slowly than Cobalt levels. Discussion. To our knowledge this is the only data in the literature prospectively comparing ion levels among groups and reporting post revision ion levels. Average serum ion levels were elevated at all post-operative samples in the MoM-1 group but this was due to significantly elevated levels in a subset of outliers who required revision. Excluding the outliers there is not a significant difference in post-operative ion levels between the groups. There was no radiographic evidence of component malposition or aseptic loosening in any of the groups. Control groups (MoM-2, MoP) performed comparatively across all variables. We present an algorithm to diagnose and manage patients with metal on metal THA and offer evidence that metal ion levels do decrease after revision but still remain abnormally elevated at one-year post revision compared to the control group. A significant portion of MoM-1 performs comparatively to the controls in terms of ions

Abstract. Objectives. Young patients receiving metallic bone implants after surgical resection of bone cancer require implants that last into adulthood, and ideally life-long. Porous implants with similar stiffness to bone can promote bone ingrowth and thus beneficial clinical outcomes. A mechanical remodelling stimulus, strain energy density (SED), is thought to be the primary control variable of the process of bone growth into porous implants. The sequential process of bone growth needs to be taken into account to develop an accurate and validated bone remodelling algorithm, which can be employed to improve porous implant design and achieve better clinical outcomes. Methods. A bone remodelling algorithm was developed, incorporating the concept of bone connectivity (sequential growth of bone from existing bone) to make the algorithm more physiologically relevant. The algorithm includes adaptive elastic modulus based on apparent bone density, using a node-based model to simulate local remodelling variations while alleviating numerical checkerboard problems. Strain energy density (SED) incorporating stress and strain effects in all directions was used as the primary stimulus for bone remodelling. The simulations were developed to run in MATLAB interfacing with the commercial FEA software ABAQUS and Python. The algorithm was applied to predict bone ingrowth into a porous implant for comparison against data from a sheep model. Results. The accuracy of the predicted bone remodelling was verified for standard loading cases (bending, torsion) against analytical calculations. Good convergence was achieved. The algorithm predicted good bone remodelling and growth into the investigated porous implant. Using the standard algorithm without connectivity, bone started to remodel at locations unconnected to any bone, which is physiologically implausible. The implementation of bone connectivity ensures the gradual process of bone growth into the implant pores from the sides. The bone connectivity algorithm predicted that the full remodelling required more time (approximately 50% longer), which should be considered when developing post-surgical rehabilitation strategies for patients. Both algorithms with and without bone connectivity implementation converged to same final stiffness (less than 0.01% difference). Almost all nodes reached the same density value, with only a limited number of nodes (less than 1%) in transition areas with a strong density gradient having noticeable differences. Conclusions. An improved bone remodelling algorithm based on strain energy density that modelled the sequential process of bone growth has been developed and tested. For a porous metallic bone implant the same final bone density distribution as for the original adaptive elasticity theory was predicted, with a slower and more fidelic process of growth from existing surrounding bone into the porous implant. 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

Abstract. Objectives. Ultrasound speckle tracking is a safe and non-invasive diagnostic tool to measure soft tissue deformation and strain. In orthopaedics, it could have broad application to measure how injury or surgery affects muscle, tendon or ligament biomechanics. However, its application requires custom tuning of the speckle-tracking algorithm then validation against gold-standard reference data. Implementing an experiment to acquire these data takes months and is expensive, and therefore prohibits use for new applications. Here, we present an alternative optimisation approach that automatically finds suitable machine and algorithmic settings without requiring gold-standard reference data. Methods. The optimisation routine consisted of two steps. First, convergence of the displacement field was tested to exclude the settings that would not track the underlying tissue motion (e.g. frame rates that were too low). Second, repeatability was maximised through a surrogate optimisation scheme. All settings that could influence the strain calculation were included, ranging from acquisition settings to post-processing smoothing and filtering settings, totalling >1,000,000 combinations of settings. The optimisation criterion minimised the normalised standard deviation between strain maps of repeat measures. The optimisation approach was validated for the medial collateral ligament (MCL) with quasi-static testing on porcine joints (n=3), and dynamic testing on a cadaveric human knee (n=1, female, aged 49). Porcine joints were fully dissected except for the MCL and loaded in a material-testing machine (0 to 3% strain at 0.2 Hz), which was captured using both ultrasound (>14 repeats per specimen) and optical digital image correlation (DIC). For the human cadaveric knee (undissected), 3 repeat ultrasound acquisitions were taken at 18 different anterior/posterior positions over the MCL while the knee was extended/flexed between 0° and 90° in a knee extension rig. Simultaneous optical tracking recorded the position of the ultrasound transducer, knee kinematics and the MCL attachments (which were digitised under direct visualisation post testing). Half of the data collected was used for optimisation of the speckle tracking algorithms for the porcine and human MCLs separately, with the remaining unseen data used as a validation test set. Results. For the porcine MCLs, ultrasound strains closely matched DIC strains (R. 2. > 0.98, RMSE < 0.59%) (Figure 1A). For the human MCL (Figure 1B), ultrasound strains matched the strains estimated from the optically tracked displacements of the MCL attachments. Furthermore, strains developed during flexion were highly correlated with AP position (R = 0.94) with strains decreasing the further posterior the transducer was on the ligament. This is in line with previously reported length change values for the posterior, intermediate and anterior bundles of the MCL. Conclusions. Ultrasound speckle tracking algorithms can be adapted for new applications without ground-truth data by using an optimisation approach that verifies displacement field convergence then minimises variance between repeat measurements. This optimisation routine was insensitive to anatomical variation and loading conditions, working for both porcine and human MCLs, and for quasi-static and dynamic loading. This will facilitate research into changes in musculoskeletal tissue motion due to abnormalities or pathologies. Declaration of Interest. (a) fully declare any financial or other potential conflict of interest

The optimal treatment strategy for post-traumatic long bone non-unions is subject of an ongoing discussion. At the Maastricht University Medical Center (MUMC+) the induced membrane technique is used to treat post-traumatic long bone non-unions. This technique uses a multimodal treatment algorithm involving bone marrow aspirate concentrate (BMAC), the reamer-irrigator-aspirator (RIA) and P-15 bioactive peptide (iFactor, Cerapedics). Bioactive glass (S53P4 BAG, Bonalive) is added when infection is suspected. This study aims to objectify the effect of this treatment algorithm on the health-related quality of life (HRQoL) of patients with post-traumatic long bone non-unions. We hypothesized that HRQoL would improve after treatment. From January 2020 to March 2023, consecutive patients who were referred to a multidisciplinary (trauma, orthopaedic and plastic surgery) non-union clinic at the MUMC+, The Netherlands, were evaluated using the Non-Union Scoring System (NUSS). The EQ-5D-5L questionnaire and the Lower Extremity Functional Scale (LEFS) were employed to obtain HRQoL outcomes both prior to and subsequent to surgery, with a follow-up at 6, 18 and 35 weeks. Seventy-six patients were assessed at baseline (T0), with a mean NUSS of 40 (± 13 SD). Thirty-eight patients had their first follow-up, six weeks after surgery (T1). Thirty-one patients had a second follow-up at 18 weeks (T2), and twenty patients had the third follow-up at 35 weeks (T3). The EQ-5D index mean at baseline was 0.480, followed by an index of 0.618 at T1, 0.636 at T2, and 0.702 at T3. A significant difference was found in the HRQoL score between T0 and T1, as well as T2 and T3 (p<0.001; p=0.011). The mean LEFS significantly increased from 26 before intervention to 34, 39, and 43 after treatment (p<0.001; p=0.033; p=0.016). This study demonstrated a significant improvement in the health-related quality of life of patients with post-traumatic long bone non-unions after the standardized treatment algorithm following the induced membrane technique

Introduction:. Most of the published papers on AI based diagnosis have focused on the algorithm's diagnostic performance in a ‘binary’ setting (i.e. disease vs no disease). However, no study evaluated the actual value for the clinicians of an AI based approach in diagnostic. Detection of Traumatic thoracolumbar (TL) fractures is challenging on planar radiographs, resulting in significant rates of missed diagnoses (30-60%), thus constituting a field in which a performance improvement is needed. Aim of this study is therefore to evaluate the value provided by AI generated saliency maps (SM), i.e. the maps that highlight the AI identified region of interests. Methods:. An AI model aimed at identifying TL fractures on plain radiographs was trained and tested on 567 single vertebrae images. Three expert spine surgeons established the Ground Truth (GT) using CT and MRI to confirm the presence of the fracture. From the test set, 12 cases (6 with a GT of fracture and 6 with a GT of no fracture, associated with varying levels of algorithm confidence) were selected and the corresponding SMs were generated and shown to 7 independent evaluators with different grade of experience; the evaluators were requested to: (1) identify the presence or absence of a fracture before and after the saliency map was shown; (2) grade, with a score from 1 (low) to 6 (high) the pertinency (correlation between the map and the human diagnosis), and the utility (the perceived utility in confirming or not the initial diagnosis) of the SM. Furthermore, the usefulness of the SM was evaluated through the rate of correct change in diagnosis after the maps had been shown. Finally, the obtained scores were correlated with the algorithm confidence for the specific case. Results:. Of the selected maps, 8 had an agreement between the AI diagnosis and the GT, while in 4 the diagnosis was discordant (67% accuracy). The pertinency of the map was found higher when the AI diagnosis was the same as the GT and the human diagnosis (respectively p-value = .021 and <.000). A positive and significant correlation between the AI confidence score and the perceived utility (Spearman: 27%, p-value=.0-27) was found. Furthermore, evaluator with experience < 5 year found the maps more useful than the experts (z-score=2.004; p-value=.0455). Among the 84 evaluation we found 12 diagnostic errors in respect to the GT, 6 (50%) of which were reverted after the saliency map evaluation (z statistic = 1.25 and p-value = .21). Discussion:. The perceived utility of AI generated SM correlate with the model confidence in the diagnosis. This highlights the fact that to be considered helpful, the AI must provide not only the diagnosis but also the case specific confidence. Furthermore, the perceived utility was higher among less experienced users, but overall, the SM were useful in improving the human diagnostic accuracy. Therefore, in this setting, the AI enhanced approach provides value in improving the human performance

Currently implemented accuracy metrics in open-source libraries for segmentation by supervised machine learning are typically one-dimensional scores [1]. While extremely relevant to evaluate applicability in clinics, anatomical location of segmentation errors is often neglected. This study aims to include the three-dimensional (3D) spatial information in the development of a novel framework for segmentation accuracy evaluation and comparison between different methods. Predicted and ground truth (manually segmented) segmentation masks are meshed into 3D surfaces. A template mesh of the same anatomical structure is then registered to all ground truth 3D surfaces. This ensures all surface points on the ground truth meshes to be in the same anatomically homologous order. Next, point-wise surface deviations between the registered ground truth mesh and the meshed segmentation prediction are calculated and allow for color plotting of point-wise descriptive statistics. Statistical parametric mapping includes point-wise false discovery rate (FDR) adjusted p-values (also referred to as q-values). The framework reads volumetric image data containing the segmentation masks of both ground truth and segmentation prediction. 3D color plots containing descriptive statistics (mean absolute value, maximal value,…) on point-wise segmentation errors are rendered. As an example, we compared segmentation results of nnUNet [2], UNet++ [3] and UNETR [4] by visualizing the mean absolute error (surface deviation from ground truth) as a color plot on the 3D model of bone and cartilage of the mean distal femur. A novel framework to evaluate segmentation accuracy is presented. Output includes anatomical information on the segmentation errors, as well as point-wise comparative statistics on different segmentation algorithms. Clearly, this allows for a better informed decision-making process when selecting the best algorithm for a specific clinical application

Abstract. Objectives. Current literature on pilon fracture includes a range of different management strategies, however there is no universal treatment algorithm. We aim to determine clinical outcomes in patients with open and closed pilon fractures, managed using a treatment algorithm applied consistently over the span of this study. Methods. 135 patients over a 6-year period were included. Primary outcome was AOFAS score at 3, 6, 12-months post-injury. Secondary outcomes include time to partial weight-bear (PWB), full weight-bear (FWB), bone union time, follow-up time. AO/OTA classification was used (43A: n=23, 43B: n=30, 43C: n=82). Treatment algorithm consisted of fine wire fixator (FWF) for severely comminuted closed fractures (AO/OTA type 43C3), or open fractures with severe soft tissue injury (GA type 3). Otherwise, open reduction internal fixation (ORIF) was performed. When required, minimally invasive osteosynthesis was performed in combination with FWF to improve joint congruency. Results. Mean AOFAS score 3, 6, and 12 months post-treatment for open and closed fracture patients were 44.12 and 53.99 (p=0.007), 62.38 and 67.68 (p=0.203), 78.44 and 84.06 (p=0.256), respectively. 119 of 141 fractures healed without further intervention (84.4%). Average time to union was 51.46 and 36.48 weeks for open and closed fractures, respectively (p=0.019). On average, open, and closed fracture patients took 12.29 and 10.76 weeks to PWB (p=0.361); 24.04 and 20.31 weeks to FWB (p=0.235), respectively. Common complications for open fractures were non-union (24%), post-traumatic arthritis (16%); for closed fractures they were post-traumatic arthritis (25%), superficial infection (22%). Open fracture was a risk factor for non-union (p=0.042;OR=2.558,95% CI 1.016–6.441), bone defect (p=0.001;OR=5.973,95% CI 1.986–17.967), and superficial infection (p<0.001;OR=4.167,95% CI 1.978–8.781). Conclusions. FWF with minimally invasive osteosynthesis, where required for severely comminuted closed fractures, and FWF for open fractures with severe soft tissue injury, are safe methods achieving low complication rates and good functional recovery

Patients with advanced cancer can develop bone metastases in the femur which are often painful and increase the risk of pathological fracture. Accurate segmentation of bone metastases is, amongst others, important to improve patient-specific computer models which calculate fracture risk, and for radiotherapy planning to determine exact radiation fields. Deep learning algorithms have shown to be promising to improve segmentation accuracy for metastatic lesions, but require reliable segmentations as training input. The aim of this study was to investigate the inter- and intra-operator reliability of manual segmentation of femoral metastatic lesions and to define a set of lesions which can serve as a training dataset for deep learning algorithms. F. CT-scans of 60 advanced cancer patients with a femur affected with bone metastases (20 osteolytic, 20 osteoblastic and 20 mixed) were used in this study. Two operators were trained by an experienced radiologist and then segmented the metastatic lesions in all femurs twice with a four-week time interval. 3D and 2D Dice coefficients (DCs) were calculated to quantify the inter- and intra-operator reliability of the segmentations. We defined a DC>0.7 as good reliability, in line with a statistical image segmentation study. Mean first and second inter-operator 3D-DCs were 0.54 (±0.28) and 0.50 (±0.32), respectively. Mean intra-operator I and II 3D-DCs were 0.56 (±0.28) and 0.71 (±0.23), respectively. Larger lesions (>60 cm. 3. ) scored higher DCs in comparison with smaller lesions. This study reveals that manual segmentation of metastatic lesions is challenging and that the current manual segmentation approach resulted in dissatisfying outcomes, particularly for lesions with small volumes. However, segmentation of larger lesions resulted in a good inter- and intra-operator reliability. In addition, we were able to select 521 slices with good segmentation reliability that can be used to create a training dataset for deep learning algorithms. By using deep learning algorithms, we aim for more accurate automated lesion segmentations which might be used in computer modelling and radiotherapy planning

Syndesmotic ankle lesions involve disruption of the osseous tibiofibular mortise configuration as well as ligamentous structures stabilizing the ankle joint. Incomplete diagnosis and maltreatment of these injuries is frequent, resulting in chronic pain and progressive instability thus promoting development of ankle osteoarthritis in the long term. Although the pathogenesis is not fully understood, abnormal mechanics has been implicated as a principal determinant of ankle joint degeneration after syndesmotic ankle lesions. Therefore, the focus of this presentation will be on our recent development of a computationally efficient algorithm to calculate the contact pressure distribution in patients with a syndesmotic ankle lesion, enabling us to stratify the risk of OA development in the long term and thereby guiding patient treatment

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

We performed this systematic overview on the overlapping meta-analyses that analyzed autologous platelet-rich plasma (PRP) as an adjuvant in the repair of rotator cuff tears and identify the studies which provide the current best evidence on this subject and generate recommendations for the same. We conducted independent and duplicate electronic database searches in PubMed, Web of Science, Scopus, Embase, Cochrane Database of Systematic Reviews, and the Database of Abstracts of Reviews of Effects on September 8, 2021, to identify meta-analyses that analyzed the efficacy of PRP as an adjuvant in the repair of rotator cuff tears. Methodological quality assessment was made using Oxford Levels of Evidence, AMSTAR scoring, and AMSTAR 2 grades and used the Jadad decision algorithm to generate recommendations. 20 meta-analyses fulfilling the eligibility criteria were included. The AMSTAR scores of the included studies varied from 6–10 (mean:7.9). All the included studies had critically low reliability in their summary of results due to their methodological flaws according to AMSTAR 2 grades. The initial size of the tear and type of repair performed do not seem to affect the benefit of PRPs. Among the different preparations used, leucocyte poor (LP)-PRP possibly offers the greatest benefit as a biological augment in these situations. Based on this systematic overview, we give a Level II recommendation that intra-operative use of PRPs at the bone-tendon interface can augment the healing rate, reduce re-tears, enhance the functional outcomes and mitigate pain in patients undergoing arthroscopic rotator cuff repair

Varus ankle osteoarthritis (OA) is typically associated with peritalar instability, which may result in altered subtalar joint position. This study aimed to determine the extent to which total ankle replacement (TAR) in varus ankle OA can restore the subtalar position alignment using 3-dimensional semi-automated measurements on WBCT. Fourteen patients (15 ankles, mean age 61) who underwent TAR for varus ankle OA were retrospectively analyzed using semi- automated measurements of the hindfoot based on pre-and postoperative weightbearing WBCT (WBCT) imaging. Eight 3-dimensional angular measurements were obtained to quantify the ankle and subtalar joint alignment. Twenty healthy individuals were served as a control groups and were used for reliability assessments. All ankle and hindfoot angles improved between preoperative and a minimum of 1 year (mean 2.1 years) postoperative and were statistically significant in 6 out of 8 angles (P<0.05). Values The post-op angles were in a similar range to as those of healthy controls were achieved in all measurements and did not demonstrated statistical difference (P>0.05). Our findings indicate that talus repositioning after TAR within the ankle mortise improves restores the subtalar position joint alignment within normal values. These data inform foot and ankle surgeons on the amount of correction at the level of the subtalar joint that can be expected after TAR. This may contribute to improved biomechanics of the hindfoot complex. However, future studies are required to implement these findings in surgical algorithms for TAR in prescence of hindfoot deformity

The most important outcome predictor of Legg-Calvé-Perthes disease (LCPD) is the shape of the healed femoral head. However, the deformity of the femoral head is currently evaluated by non-reproducible, categorical, and qualitative classifications. In this regard, recent advances in computer vision might provide the opportunity to automatically detect and delineate the outlines of bone in radiographic images for calculating a continuous measure of femoral head deformity. This study aimed to construct a pipeline for accurately detecting and delineating the proximal femur in radiographs of LCPD patients employing existing algorithms. To detect the proximal femur, the pretrained stateof-the-art object detection model, YOLOv5, was trained on 1580 manually annotated radiographs, validated on 338 radiographs, and tested on 338 radiographs. Additionally, 200 radiographs of shoulders and chests were added to the dataset to make the model more robust to false positives and increase generalizability. The convolutional neural network architecture, U-Net, was then employed to segment the detected proximal femur. The network was trained on 80 manually annotated radiographs using real-time data augmentation to increase the number of training images and enhance the generalizability of the segmentation model. The network was validated on 60 radiographs and tested on 60 radiographs. The object detection model achieved a mean Average Precision (mAP) of 0.998 using an Intersection over Union (IoU) threshold of 0.5, and a mAP of 0.712 over IoU thresholds of 0.5 to 0.95 on the test set. The segmentation model achieved an accuracy score of 0.912, a Dice Coefficient of 0.937, and a binary IoU score of 0.854 on the test set. The proposed fully automatic proximal femur detection and segmentation system provides a promising method for accurately detecting and delineating the proximal femoral bone contour in radiographic images, which is necessary for further image analysis

Several emerging reports suggest an important involvement of the hindfoot alignment in the outcome of knee osteotomy. At present, studies lack a comprehensive overview. Therefore, we aimed to systematically review all biomechanical and clinical studies investigating the role of the hindfoot alignment in the setting of osteotomies around the knee. A systematic literature search was conducted on multiple databases combining “knee osteotomy” and “hindfoot/ankle alignment” search terms. Articles were screened and included according to the PRISMA guidelines. A quality assessment was conducted using the Quality Appraisal for Cadaveric Studies (QUACS) - and modified methodologic index for non-randomized studies (MINORS) scales. Three cadaveric, fourteen retrospective cohort and two case-control studies were eligible for review. Biomechanical hindfoot characteristics were positively affected (n=4), except in rigid subtalar joint (n=1) or talar tilt (n=1) deformity. Patient symptoms and/or radiographic alignment at the level of the hindfoot did also improve after knee osteotomy (n=13), except in case of a small pre-operative lateral distal tibia- and hip knee ankle (HKA) angulation or in case of a large HKA correction (>14.5°). Additionally, a pre-existent hindfoot deformity (>15.9°) was associated with undercorrection of lower limb alignment following knee osteotomy. The mean QUACS score was 61.3% (range: 46–69%) and mean MINORS score was 9.2 out of 16 (range 6–12) for non-comparative and 16.5 out of 24 (range 15–18) for comparative studies. Osteotomies performed to correct knee deformity have also an impact on biomechanical and clinical outcomes of the hindfoot. In general, these are reported to be beneficial, but several parameters were identified that are associated with newly onset – or deterioration of hindfoot symptoms following knee osteotomy. Further prospective studies are warranted to assess how diagnostic and therapeutic algorithms based on the identified criteria could be implemented to optimize the overall outcome of knee osteotomy. Remark: Aline Van Oevelen and Arne Burssens contributed equally to this work

Neurological complications in oncological and degenerative spine surgery represent one of the most feared risks of these procedures. Multimodal intraoperative neurophysiological monitoring (IONM) mainly uses methods to detect changes in the patient's neurological status in a timely manner, thus allowing actions that can reverse neurological deficits before they become irreversible. The utopian goal of spinal surgery is the absence of neurological complications while the realistic goal is to optimize the responses to changes in neuromonitoring such that permanent deficits occur less frequently as possible. In 2014, an algorithm was proposed in response to changes in neuromonitoring for deformity corrections in spinal surgery. There are several studies that confirm the positive impact that a checklist has on care. The proposed checklist has been specifically designed for interventions on stable columns which is significantly different from oncological and degenerative surgery. The goal of this project is to provide a checklist for oncological and degenerative spine surgery to improve the quality of care and minimize the risk of neurological deficit through the optimization of clinical decision-making during periods of intraoperative stress or uncertainty. After a literature review on risk factors and recommendations for responding to IONM changes, 3 surveys were administered to 8 surgeons with experience in oncological and degenerative spine surgery from 5 hospitals in Italy. In addition, anesthesiologists, intraoperative neuro-monitoring teams, operating room nurses participated. The members participated in the optimization and final drafting of the checklist. The authors reassessed and modified the checklist during 3 meetings over 9 months, including a clinical validation period using a modified Delphi process. A checklist containing 28 items to be considered in responding to the changes of the IONM was created. The checklist was submitted for inclusion in the new recommendations of the Italian Society of Clinical Neurophysiology (SINC) for intraoperative neurophysiological monitoring. The final checklist represents the consensus of a group of experienced spine surgeons. The checklist includes the most important and high-performance items to consider when responding to IONM changes in patients with an unstable spine. The implementation of this checklist has the potential to improve surgical outcomes and patient safety in the field of spinal surgery

Abstract. Objectives. Spinal disorders such as back pain incur a substantial societal and economic burden. Unfortunately, there is lack of understanding and treatment of these disorders are further impeded by the inability to assess spinal forces in vivo. The aim of this project is to address this challenge by developing and testing a novel image-driven approach that will assess the forces in an individual's spine in vivo by incorporating information acquired from multimodal imaging (magnetic resonance imaging (MRI) and biplane X-rays) in a subject-specific model. Methods. Magnetic resonance and biplane X-ray imaging are used to capture information about the anatomy, tissues, and motion of an individual's spine as they perform a range of everyday activities. This information is then utilised in a subject-specific computational model based on the finite element method to predict the forces in their spine. The project is also utilising novel machine learning algorithms and in vitro, six-axis mechanical testing on human, porcine and bovine samples to develop and test the modelling methods rigorously. Results & Discussion. MRI sequences have been identified that provide high-quality image data and information on different tissue types which will be used to predict subject-specific disc properties. In-vivo protocols to capture motion analysis, EMG muscle activity, and video X-rays of the spine have been designed with planned data collection of 15 healthy volunteers. Preliminary modelling work has evaluated potential machine learning approaches and quantified the sensitivity of the models developed to material properties. Conclusion. The development and testing of these image-driven subject-specific spine models will provide a new tool for determining forces in the spine. It will also provide new tools for measuring and modelling spine movement and quantifying the properties of the spinal tissues. Acknowledgments. Funding from the EPSRC: EP/V036602/1 (Meakin, Holsgrove & Javadi) and EP/V032275/1 (Holt & Williams). 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

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