Aims. The surgical target for optimal implant positioning in robotic-assisted total knee arthroplasty remains the subject of ongoing discussion. One of the proposed targets is to recreate the knee’s functional behaviour as per its pre-diseased state. The aim of this study was to optimize implant positioning, starting from mechanical alignment (MA), toward restoring the pre-diseased status, including ligament strain and kinematic patterns, in a patient population. Methods. We used an active appearance model-based approach to segment the preoperative CT of 21 osteoarthritic patients, which identified the osteophyte-free surfaces and estimated cartilage from the segmented bones; these geometries were used to construct patient-specific musculoskeletal models of the pre-diseased knee. Subsequently, implantations were simulated using the MA method, and a previously developed optimization technique was employed to find the optimal implant position that minimized the root mean square deviation between pre-diseased and postoperative ligament strains and kinematics. Results. There were evident biomechanical differences between the simulated patient models, but also trends that appeared reproducible at the population level. Optimizing the implant position significantly reduced the maximum observed strain root mean square deviations within the cohort from 36.5% to below 5.3% for all but the anterolateral ligament; and concomitantly reduced the kinematic deviations from 3.8 mm (SD 1.7) and 4.7° (SD 1.9°) with MA to 2.7 mm (SD 1.4) and 3.7° (SD 1.9°) relative to the pre-diseased state. To achieve this, the femoral component consistently required translational adjustments in the anterior, lateral, and proximal directions, while the tibial component required a more posterior slope and varus rotation in most cases. Conclusion. These findings confirm that MA-induced biomechanical alterations relative to the pre-diseased state can be reduced by optimizing the implant position, and may have implications to further advance pre-planning in robotic-assisted surgery in order to restore pre-diseased knee function. Cite this article: Bone Joint J 2024;106-B(11):1231–1239

Aims. The Oxford Shoulder Score (OSS) is a 12-item measure commonly used for the assessment of shoulder surgeries. This study explores whether computerized adaptive testing (CAT) provides a shortened, individually tailored questionnaire while maintaining test accuracy. Methods. A total of 16,238 preoperative OSS were available in the National Joint Registry (NJR) for England, Wales, Northern Ireland, the Isle of Man, and the States of Guernsey dataset (April 2012 to April 2022). Prior to CAT, the foundational item response theory (IRT) assumptions of unidimensionality, monotonicity, and local independence were established. CAT compared sequential item selection with stopping criteria set at standard error (SE) < 0.32 and SE < 0.45 (equivalent to reliability coefficients of 0.90 and 0.80) to full-length patient-reported outcome measure (PROM) precision. Results. Confirmatory factor analysis (CFA) for unidimensionality exhibited satisfactory fit with root mean square standardized residual (RSMSR) of 0.06 (cut-off ≤ 0.08) but not with comparative fit index (CFI) of 0.85 or Tucker-Lewis index (TLI) of 0.82 (cut-off > 0.90). Monotonicity, measured by H value, yielded 0.482, signifying good monotonic trends. Local independence was generally met, with Yen’s Q3 statistic > 0.2 for most items. The median item count for completing the CAT simulation with a SE of 0.32 was 3 (IQR 3 to 12), while for a SE of 0.45 it was 2 (IQR 2 to 6). This constituted only 25% and 16%, respectively, when compared to the 12-item full-length questionnaire. Conclusion. Calibrating IRT for the OSS has resulted in the development of an efficient and shortened CAT while maintaining accuracy and reliability. Through the reduction of redundant items and implementation of a standardized measurement scale, our study highlights a promising approach to alleviate time burden and potentially enhance compliance with these widely used outcome measures. Cite this article: Bone Joint Res 2024;13(8):392–400

When a knee flex deeply, the posterior side of thigh and calf contact. The contact force is unignorable to estimate the load acting on a knee because the force generates extensional moment on the knee, and the moment might be about 20–80% of the flexional moment generated by a floor reacting force. Besides, the thigh-calf contact force varies so much even if the posture or the test subject are the same that it is hard to use the average value to estimate the knee load. We have assumed that the force might change not only by the individual physical size but also by a slight change of the posture, especially the angle of the upper body. Therefore we tried to create the estimation equation for the thigh-calf contact force using both anthropometric sizes and posture angles as parameters. The objective posture was kneeling, both plantarflexing and dorsiflexing the ankle joint. Test subjects were 10 healthy males. They were asked to sit on a floor with kneeling, and to tilt their upper body forward and backward. The estimation equations were created as the linear combinations of the parameters, determining the coefficient as to minimize the root mean square errors. We used the parameters as explanatory variables which could be divided into posture parameters and individual parameters. Posture parameters included the angle of upper body, thigh and lower thigh. Individual parameters included height, weight, axial and circumferential lengths of thigh and lower thigh. The magnitude of the force was normalized by a body weight, and the acting position was expressed by the moment arm length around a knee joint and normalized by a height. As a result, the adjusted coefficient of determination improved and the root mean square error decreased when using both posture and individual parameters, though there were large errors when neglecting either parameters. The accuracy decreased little when using the same equation for plantarflexed and dorsiflexed kneeling in magnitude. The relation of measured and estimated values of the magnitude and acting position, using the common equation with all the parameters. It might be because the difference of the postures could be described by the inclination angle of a thigh. In both postures, the magnitude of a thigh-calf contact force was mainly affected by the posture and acting position by the individual parameters. When calculating the knee joint load, the errors would be about 8.59 Nm on the knee moment and 290 N on the knee load when using just an average, and they would decrease to 2.23 Nm and 74 N respectively using the estimation equation

Successful estimation of postoperative PROMs prior to a joint replacement surgery is important in deciding the best treatment option for a patient. However, estimation of the outcome is associated with substantial noise around individual prediction. Here, we test whether a classifier neural network can be used to simultaneously estimate postoperative PROMs and uncertainty better than current methods. We perform Oxford hip score (OHS) estimation using data collected by the NJR from 249,634 hip replacement surgeries performed from 2009 to 2018. The root mean square error (RMSE) of the various methods are compared to the standard deviation of outcome change distribution to measure the proportion of the total outcome variability that the model can capture. The area under the curve (AUC) for the probability of the change score being above a certain threshold was also plotted. The proposed classifier NN had a better or equivalent RMSE than all other currently used models. The threshold AUC shows similar results for all methods close to a change score of 20 but demonstrates better accuracy of the classifier neural network close to 0 change and greater than 30 change, showing that the full probability distribution performed by the classifier neural network resulted in a significant improvement in estimating the upper and lower quantiles of the change score probability distribution. Consequently, probabilistic estimation as performed by the classifier NN is the most adequate approach to this problem, since the final score has an important component of uncertainty. This study shows the importance of uncertainty estimation to accompany postoperative PROMs prediction and presents a clinically-meaningful method for personalised outcome that includes such uncertainty estimation

Abstract. Rehabilitation exercise is critical for patients’ recovery after knee injury or post-surgery. Unfortunately, adherence to exercise is low due to a lack of positive feedback and poor self-motivation. Therefore, it is crucial to monitor their progress and provide supervision. Inertial measurement unit (IMUs) based sensing technology can provide remote patient monitoring functions. However, most current solutions only measure the range of knee motion in one degree of freedom. The current IMUs estimate the orientation-angle based on the integrated raw data, which might lack accuracy in measuring knee motion. This study aims to develop an IMU-based sensing system using the absolute measured orientation-angle to provide more accurate comprehensive monitoring by measuring the knee rotational angles. An IMU sensing system monitoring the knee joint angles, flexion/extension (FE), adduction/abduction (AA), and internal/external (IE) was developed. The accuracy and reliability of FE measurements were validated in human participants during squat exercise using measures including root mean square error (RMSE) and correlation coefficient. The RMSE of the three knee angles (FE, AA, and IE) were 0.82°, 0.26°, and 0.11°, which are acceptable for assessing knee motion. The FE measurement was validated in human participants and showed excellent accuracy (correlation coefficient of 0.99°). Further validation of AA and IE in human participants is underway. The sensing system showed the capability to estimate three knee rotation angles (FE, AA, and IE). It showed the potential to provide comprehensive continuous monitoring for knee rehabilitation exercises, which can also be used as a clinical assessment tool

Aims. The purpose of this study was to develop a personalized outcome prediction tool, to be used with knee arthroplasty patients, that predicts outcomes (lengths of stay (LOS), 90 day readmission, and one-year patient-reported outcome measures (PROMs) on an individual basis and allows for dynamic modifiable risk factors. Methods. Data were prospectively collected on all patients who underwent total or unicompartmental knee arthroplasty at a between July 2015 and June 2018. Cohort 1 (n = 5,958) was utilized to develop models for LOS and 90 day readmission. Cohort 2 (n = 2,391, surgery date 2015 to 2017) was utilized to develop models for one-year improvements in Knee Injury and Osteoarthritis Outcome Score (KOOS) pain score, KOOS function score, and KOOS quality of life (QOL) score. Model accuracies within the imputed data set were assessed through cross-validation with root mean square errors (RMSEs) and mean absolute errors (MAEs) for the LOS and PROMs models, and the index of prediction accuracy (IPA), and area under the curve (AUC) for the readmission models. Model accuracies in new patient data sets were assessed with AUC. Results. Within the imputed datasets, the LOS (RMSE 1.161) and PROMs models (RMSE 15.775, 11.056, 21.680 for KOOS pain, function, and QOL, respectively) demonstrated good accuracy. For all models, the accuracy of predicting outcomes in a new set of patients were consistent with the cross-validation accuracy overall. Upon validation with a new patient dataset, the LOS and readmission models demonstrated high accuracy (71.5% and 65.0%, respectively). Similarly, the one-year PROMs improvement models demonstrated high accuracy in predicting ten-point improvements in KOOS pain (72.1%), function (72.9%), and QOL (70.8%) scores. Conclusion. The data-driven models developed in this study offer scalable predictive tools that can accurately estimate the likelihood of improved pain, function, and quality of life one year after knee arthroplasty as well as LOS and 90 day readmission. Cite this article: Bone Joint J 2020;102-B(9):1183–1193

There is currently no commercially available and clinically successful treatment for scapholunate interosseous ligament rupture, the latter leading to the development of hand-wrist osteoarthritis. We have created a novel biodegradable implant which fixed the dissociated scaphoid and lunate bones and encourages regeneration of the ruptured native ligament. To determine if scaphoid and lunate kinematics in cadaveric specimens were maintained during robotic manipulation, when comparing the native wrist with intact ligament and when the implant was installed. Ten cadaveric experiments were performed with identical conditions, except for implant geometry that was personalised to the anatomy of each cadaveric specimen. Each cadaveric arm was mounted upright in a six degrees of freedom robot using k-wires drilled through the radius, ulna, and metacarpals. Infrared markers were attached to scaphoid, lunate, radius, and 3rd metacarpal. Cadaveric specimens were robotically manipulated through flexion-extension and ulnar-radial deviation by ±40° and ±30°, respectively. The cadaveric scaphoid and lunate kinematics were examined with 1) intact native ligament, 2) severed ligament, 3) and installed implant. Digital wrist models were generated from computed tomography scans and included implant geometry, orientation, and location. Motion data were filtered and aligned relative to neutral wrist in the digital models of each specimen using anatomical landmarks. Implant insertion points in the scaphoid and lunate over time were then calculated using digital models, marker data, and inverse kinematics. Root mean squared distance was compared between severed and implant configurations, relative to intact. Preliminary data from five cadaveric specimens indicate that the implant reduced distance between scaphoid and lunate compared to severed configuration for all but three trials. Preliminary results indicate our novel implant reduced scapho-lunate gap caused by ligament transection. Future analysis will reveal if the implant can achieve wrist kinematics similar to the native intact wrist

This study aims to create a novel computational workflow for frontal plane laxity evaluation which combines a rigid body knee joint model with a non-linear implicit finite-element model wherein collateral ligaments are anisotropically modelled using subject-specific, experimentally calibrated Holzpfel-Gasser-Ogden (HGO) models. The framework was developed based on CT and MRI data of three cadaveric post-TKA knees. Bones were segmented from CT-scans and modelled as rigid bodies in a multibody dynamics simulation software (MSC Adams/view, MSC Software, USA). Medial collateral and lateral collateral ligaments were segmented based on MRI-scans and are modelled as finite elements using the HGO model in Abaqus (Simulia, USA). All specimens were submitted varus/valgus loading (0-10Nm) while being rigidly fixed on a testing bench to prevent knee flexion. In subsequent computer simulations of the experimental testing, rigid bodies kinematics and the associated soft-tissue force response were computed at each time step. Ligament properties were optimised using a gradient descent approach by minimising the error between the experimental and simulation-based kinematic response to the applied varus/valgus loads. For comparison, a second model was defined wherein collateral ligaments were modelled as nonlinear no-compression spring elements using the Blankevoort formulation. Models with subject-specific, experimentally calibrated HGO representations of the collateral ligaments demonstrated smaller root mean square errors in terms of kinematics (0.7900° +/− 0.4081°) than models integrating a Blankevoort representation (1.4704° +/− 0.8007°). A novel computational workflow integrating subject-specific, experimentally calibrated HGO predicted post-TKA frontal-plane knee joint laxity with clinically applicable accuracy. Generally, errors in terms of tibial rotation were higher and might be further reduced by increasing the interaction nodes between the rigid body model and the finite element software. Future work should investigate the accuracy of resulting models for simulating unseen activities of daily living

The spinopelvic alignment is often assessed via the Pelvic Incidence-Lumbar Lordosis (PI-LL) mismatch. Here we describe and validate a simplified method to evaluating the spinopelvic alignment through the L1-Pelvis angle (L1P). This method is set to reduce the operator error and make the on-film measurement more practicable. 126 standing lateral radiographs of patients presenting for Total Hip Arthroplasty were examined. Three operators were recruited to label 6 landmarks. One operator repeated the landmark selection for intra-operator analysis. We compare PI-LL mismatch obtained via the conventional method, and our simplified method where we estimate this mismatch using PI-LL = L1P - 90°. We also assess the method's reliability and repeatability. We found no significant difference (p > 0.05) between the PI-LL mismatch from the conventional method (mean 0.22° ± 13.6) compared to L1P method (mean 0.0° ± 13.1). The overall average normalised root mean square error (NRMSE) for PI-LL mismatch across all operators is 7.53% (mean -3.3° ± 6.0) and 6.5% (mean -2.9° ± 4.9) for the conventional and L1P method, respectively. In relation to intra-operator repeatability, the correlation coefficients are 0.87 for PI, 0.94 for LL, and 0.96 for L1P. NRMSE between the two measurement sets are PI: 9.96%, LL: 5.97%, and L1P: 4.41%. A similar trend is observed in the absolute error between the two sets of measurements. Results indicate an equivalence in PI-LL measurement between the methods. Reproducibility of the measurements and reliability between operators were improved. Using the L1P angle, the classification of the sagittal spinal deformity found in the literature translates to: normal L1P<100°, mild 100°<L1P<110°, and severe L1P>110°. Surgeons adopting our method should expect a small improvement in reliability and repeatability of their measurements, and a significant improvement of the assessment of the mismatch through the visualisation of the angle L1P

Successful estimation of postoperative PROMs prior to a joint replacement surgery is important in deciding the best treatment option for a patient. However, estimation of the outcome is associated with substantial noise around individual prediction. Here, we test whether a classifier neural network can be used to simultaneously estimate postoperative PROMs and uncertainty better than current methods. We perform Oxford hip score (OHS) estimation using data collected by the NJR from 249,634 hip replacement surgeries performed from 2009 to 2018. The root mean square error (RMSE) of the various methods are compared to the standard deviation of outcome change distribution to measure the proportion of the total outcome variability that the model can capture. The area under the curve (AUC) for the probability of the change score being above a certain threshold was also plotted. The proposed classifier NN had a better or equivalent RMSE than all other currently used models. The standard deviation for the change score for the entire population was 9.93, which can be interpreted as the RMSE that would be achieved for a model that gives the same estimation for all patients regardless of the covariates. However, most of the variation in the postoperative OHS/OKS change score is not captured by the models, confirming the importance of accurate uncertainty estimation. The threshold AUC shows similar results for all methods close to a change score of 20 but demonstrates better accuracy of the classifier neural network close to 0 change and greater than 30 change, showing that the full probability distribution performed by the classifier neural network resulted in a significant improvement in estimating the upper and lower quantiles of the change score probability distribution. Consequently, probabilistic estimation as performed by the classifier NN is the most adequate approach to this problem, since the final score has an important component of uncertainty. This study shows the importance of uncertainty estimation to accompany postoperative PROMs prediction and presents a clinically-meaningful method for personalised outcome that includes such uncertainty estimation

Abstract. Objectives. Conventional approaches (including Tobit) do not accurately account for ceiling effects in PROMs nor give uncertainty estimates. Here, a classifier neural network was used to estimate postoperative PROMs prior to surgery and compared with conventional methods. The Oxford Knee Score (OKS) and the Oxford Hip Score (OHS) were estimated with separate models. Methods. English NJR data from 2009 to 2018 was used, with 278.655 knee and 249.634 hip replacements. For both OKS and OHS estimations, the input variables included age, BMI, surgery date, sex, ASA, thromboprophylaxis, anaesthetic and preoperative PROMs responses. Bearing, fixation, head size and approach were also included for OHS and knee type for OKS estimation. A classifier neural network (NN) was compared with linear or Tobit regression, XGB and regression NN. The performance metrics were the root mean square error (RMSE), maximum absolute error (MAE) and area under curve (AUC). 95% confidence intervals were computed using 5-fold cross-validation. Results. The classifier NN and regression NN had the best RMSE, both with the same scores of 8.59±0.04 for knee and 7.88±0.04 for hip. The classifier NN had the best MAE, with 6.73±0.03 for knee and 5.73±0.03 for hip. The Tobit model was second, with 6.86±0.03 for knee and 6.00±0.01 for hip. The classifier NN had the best AUC, with (68.7±0.4)% for knee and (73.9±0.3)% for hip. The regression NN was second, with (67.1±0.3)% for knee and (71.1±0.4)% for hip. The Tobit model had the best AUC among conventional approaches, with (66.8±0.3)% for knee and (71.0±0.4)% for hip. Conclusions. The proposed model resulted in an improvement from the current state-of-the-art. Additionally, it estimates the full probability distribution of the postoperative PROMs, making it possible to know not only the estimated value but also its uncertainty

Abstract. Objectives. Total hip arthroplasty (THA) procedures are physically demanding for surgeons. Repetitive mallet swings to impact a surgical handle (impactions), can lead to muscle fatigue, discomfort and injuries. The use of an automated surgical hammer may reduce fatigue and increase surgical efficiency. The aim of this study was to develop a method to quantify user's performance, by recording surface electromyography (sEMG), for automated and manual impactions. Methods. sEMG signals were recorded from eight muscle compartments (arm and back muscles) of an orthopaedic surgeon during repetitions of manual and automated impaction tasks, replicating femoral canal preparation (broaching) during a THA. Each task was repeated, randomly, four times manually and four times with the automated impaction device. The mechanical outcomes (broaching efficiency and broach advancement) were quantified by tracking the kinematics of the surgical instrumentation. Root mean square (RMS) values and median frequency (MDF) were calculated for each task to, respectively, investigate which muscles were mostly involved (higher RMS) in each task and to quantify the decrease in MDF, which is an indicator of muscle fatigue. Results. RMS for arm muscles was significantly higher (p-value=0.002) during manual impactions than during automated impactions and muscle fatigue was significantly reduced (p-value=0.011), for the same muscles, when the same tasks were performed with the automated surgical hammer. The time required to achieve the same mechanical outcome, in terms of broaching efficiency and broach advancement, was significantly reduced with the automated surgical hammer (p=0.019). Conclusions. Results from this study showed how with this methodology it was possible to discern muscle performance and fatigue, between impaction modalities. Moreover, the reduction in exposure time to automated impactions, could be a factor in muscle fatigue decrease. These results could therefore provide useful insights into the study of surgical ergonomic improvements, to reduce surgeons muscle fatigue and, potentially, injuries. Declaration of Interest. (a) fully declare any financial or other potential conflict of interest

INTRODUCTION. To assess and compare the effect of new orthopedic surgical procedures, in vitro evaluation remains critical during the pre-clinical validation. Focusing on reconstruction surgery, the ability to restore normal kinematics and stability is thereby of primary importance. Therefore, several simulators have been developed to study the kinematics and create controlled boundary conditions. To simultaneously capture the kinematics in six degrees of freedom as outlined by Grood & Suntay, markers are often rigidly connected to the moving bone segments. The position of these markers can subsequently be tracked while their position relative to the bones is determined using computed tomography (CT) of the test specimen with the markers attached. Although this method serves as golden standard, it clearly lacks real-time feedback. Therefore, this paper presents the validation of a newly developed real-time framework to assess knee kinematics at the time of testing. MATERIALS & METHODS. A total of five cadaveric fresh frozen lower limb specimens have been used to quantitatively assess the difference between the golden standard, CT based, method and the newly developed real-time method. A schematic of the data flow for both methods. Prior to testing, both methods require a CT scan of the full lower limb. During the tests, the proximal femur and distal tibia are necessarily resected to fit the knees in the test setup, thus also removing the anatomical landmarks needed to evaluate their mechanical axis. Subsequently, a set of three passive markers are rigidly attached to the femur and tibia, referred to as M3F and M3T respectively. For the CT based method, the marker positions are captured during the tests and a second CT scan is eventually performed to link the marker positions to the knee anatomy. Using in-house developed software, this allowed to offline evaluate the knee kinematics in six degrees of freedom by combining both CT datasets with the tracked marker positions. For the newly developed real-time method, a calibration procedure is first performed. This calibration aims to link the position of the 3D reconstructed bone and landmarks with the attached markers. A set of bone surface points is therefore registered. These surface points are obtained by tracking the position of a pen while touching the bone surface. The pen's position is thereby tracked by three rigidly attached markers, denoted M3P. The position of the pen tip is subsequently calculated from the known pen geometry. The iterative closest point (ICP) algorithm is then used to match the 3D reconstructed bone to the registered surface points. Two types of 3D reconstructions have therefore been considered. First, the original reconstructions were used, obtained from the CT data. Second, a modified reconstruction was used. This modification accounted for the finite radius (r = 1.0 mm) of the registration pen, by shifting the surface nodes 1.0 mm along the direction of the outer surface normal. During the tests, the positions of the femur and tibia markers are tracked and streamed in real-time to an in-house developed, Matlab based software framework (MathWorks Inc., Natick, Massachussets, USA). This software framework simultaneously calculates the bone positions and knee kinematics in six degrees of freedom, displaying this information to the surgeons and operators. To assess the accuracy, all knee specimens have been subjected to passive flexion-extension movement ranging from 0 to 120 degrees of flexion. For each degree of freedom, the average root mean square (RMS) difference between both measurement methods has been evaluated during this movement. In addition, the distribution of the registered surface points has been assessed along the principal directions of the uniformly meshed 3D reconstructions (average mesh size of 1.0 mm). RESULTS. The root mean square difference between both measurements indicates a strong dependency on the variance of the registered points. This dependency is particularly pronounced when using the original 3D reconstructions in combination with the ICP algorithm, with an R. 2. = 0.76 and 0.85 for the translational and rotational degrees of freedom respectively. When using the modified 3D reconstructions, which compensates for the finite radius of the marker tip, this dependency becomes negligible (R. 2. = 0.10 and 0.05). Using this modified 3D reconstruction, the average difference between both measurements is also reduced to an average value of 1.20 degrees and 1.47 mm. DISCUSSION. The difference in kinematic parameters between both measurement techniques is an order of magnitude lower than the claimed accuracy of the motion tracking cameras. However, the difference is in line with the inter- and intra- observer variability when identifying bony landmarks around the knee. Since these landmarks are essential to calculate knee kinematics, it is understood that the proposed real-time system is sufficiently accurate to study these kinematics

Introduction & Aims. In other medical fields, smart implantable devices are enabling decentralised monitoring of patients and early detection of disease. Despite research-focused smart orthopaedic implants dating back to the 1980s, such implants have not been adopted into regular clinical practice. The hardware footprint and commercial cost of components for sensing, powering, processing, and communicating are too large for mass-market use. However, a low-cost, minimal-modification solution that could detect loosening and infection would have considerable benefits for both patients and healthcare providers. This proof-of-concept study aimed to determine if loosening/infection data could be monitored with only two components inside an implant: a single-element sensor and simple communication element. Methods. The sensor and coil were embedded onto a representative cemented total knee replacement. The implant was then cemented onto synthetic bone using polymethylmethacrylate (PMMA). Wireless measurements for loosening and infection were then made across different thicknesses of porcine tissue to characterise the sensor's accuracy for a range of implantation depths. Loosening was simulated by taking measurements before and after compromising the implant-cement interface, with fluid influx simulated with phosphate-buffered saline solution. Elevated temperature was used as a proxy for infection, with the sensor calibrated wirelessly through 5 mm of porcine tissue across a temperature range of 26–40°C. Results. Measurements for loosening and infection could be acquired simultaneously with a duration of 4 s per measurement. For loosening, the debonded implant-cement interface was detectable up to 10 mm with 95% confidence. For temperature, the sensor was calibrated with a root mean square error of 0.19°C at 5 mm implantation depth and prediction intervals of ±0.38°C for new measurements with 95% confidence. Conclusions. This study has demonstrated that with only two onboard electrical components, it is possible to wirelessly measure cement debonding and elevated temperature on a smart implant. With further development, this minimal hardware/cost approach could enable mass-market smart arthroplasty implants

Abstract. Objectives. Develop a methodology to assess the long term mechanical behavior of intervertebral discs by utilizing novel sequential state testing. Methods. Bovine functional spinal units were sequentially mechanically tested in (1) native (n=8), (2) degenerated (n=4), and (3) treated states (n=4). At stage (2), artificial degeneration was created using rapid enzymatic degeneration, followed by a 24 hour hold period under static load at 42°C. At stage (3), nucleus augmentation treatments were injected with a hydrogel or a ‘sham’ (water, chondroitin sulfate) injection. The mechanical protocol employed applied a static load hold period followed by cyclic compressive loading between ∼350 and 750 N at 1 Hz. 1000 cycles were applied at each stage, and the final test on each specimen was extended up to 20000 cycles. To verify if test time can be reduced, functions were fitted using stiffness data up to 100, 1000, 2500, 5000, 10000 and 20000 cycles. Linear regression for the native specimens comparing the stiffness at various cycles to the stiffness at 20000 cycles was completed. Results. Independent of the disc state, as the number of cycles increased, the hysteresis decreased and the stiffness increased. The degenerated specimen stiffness was greater than the healthy and treated stiffness and the degenerate hysteresis loops were smaller. A mathematical model was found to successfully predict the high cycle behaviour of the disc reaching a root mean squared (RMS) error below 10% when using 5000 or more cycles. The linear regression gave a RMS error below 7.5% at 1000 cycles. Conclusions. A method was developed to consistently determine intervertebral disc mechanics through sequential testing. A shortened cyclic testing period was shown to be viable as a method to reduce preliminary test time for novel hydrogels, compared to currently literature. The methodology permits rapid preliminary assessment of intervertebral disc mechanics and treatments. 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

The angle of acetabular inclination is an important measurement in total hip replacement (THR) procedures. Determining the acetabular component orientation intra-operatively remains a challenge. An increasing number of innovators have described techniques and devices to achieve it. This paper describes a mechanical inclinometer design to measure intra-operative acetabular cup inclination. Then, the mechanical device is tested to determine its accuracy. The aim was to design an inclinometer to measure inclination without existing instrumentation modification. The device was designed to meet the following criteria: 1. measure inclination with acceptable accuracy (+/− 5o); 2. easy to use intra-operatively (handling & visualization); 3. adaptable and useable with majority of instrumentation kits without modification; 4. sterilizable by all methods; 5. robust/reusable. The prototype device was drafted by computer aided design (CAD) software. Then a prototype was constructed using a 3D printer to establish the final format. The final device was CNC machined from SAE 304 stainless steel. The design uses an eccentrically weighted flywheel mounted on two W16002-2RS ball bearings pressed into symmetrical housing components. The weighted wheel is engraved with calibrated markings relative to its mass centre. Device functioning is dependent on gravity maintaining the weighted wheel in a fixed orientation while the housing can adapt to the calibration allowing for determining the corresponding measurement. The prototype device accuracy was compared to a digital device. A digital protractor was used to create an angle. The mechanical inclinometer (user blinded to digital reading) was used to determine the angle and compared to the digital reading. The accuracy of the device compared to the standard freehand technique was assessed using a saw bone pelvis fixed in a lateral decubitus position. 18 surgeons (6 expert, 6 intermediate, 6 novice) were asked to place an uncemented acetabular cup in a saw bone pelvis to a target of 40 degrees. First freehand then using the inclinometer. The inclination was determined using a custom-built inertial measurement unit with the user blinded to the result. Comparison between the mechanical and digital devices showed that the mechanical device had an average error of −0.2, a standard deviation of 1.5, and range −3.3 to 2.6. The average root mean square error was 1.1 with a standard deviation of 0.9. Comparison of the inclinometer to the freehand technique showed that with the freehand component placement 50% of the surgeons were outside the acceptable range of 35–45 degrees. The use of the inclinometer resulted all participants to achieve placement within the acceptable range. It was noted that expert surgeons were more accurate at achieving the target inclination when compared to less experienced surgeons. This work demonstrates that the design and initial testing of a mechanical inclinometer is suitable for use in determining the acetabular cup inclination in THR. Experimental testing showed that the device is accurate to within acceptable limits and reliably improved the accuracy of uncemented cup implantation in all surgeons

INTRODUCTION. Statistical shape models (SSM) have become a common tool to create reference models for design input and verification of total joint implants. In a recent discussion paper around Artificial Intelligence and Machine Learning, the FDA emphasizes the importance of independent test data [1]. A leave-one-out test is a standard way to evaluate the generalization ability of an SSM [2]; however, this test does not fulfill the independence requirement of the FDA. In this study, we constructed an SSM of the knee (femur and tibia). Next to the standard leave-one-out validation, we used an independent test set of patients from a different geographical region than the patients used to build the SSM. We assessed the ability of the SSM to predict the shapes of knees in this independent test set. METHODS. A dataset of 82 computed tomography (CT) scans of Caucasian patients (42 male, 40 female) from 11 different geographic locations in France, Germany, Austria, Italy and Australia were used as training set to make an SSM of the femur and tibia. A leave-one-out test was performed to assess the ability of the SSM to predict shapes within the training set. A test dataset of 4 CT scans of Caucasian patients from Russia were used for the validation. The SSM was fitted onto each of the femur and tibia shapes and the root mean square error (RMSE) was measured. RESULTS. The leave-one-out tests showed that the femur and tibia SSMs were able to predict patients in the input population with an RMSE of 0.59 ± 0.1 mm (average ± standard deviation) for the femur and 0.70 ± 0.1 mm for the tibia. The validation test showed that the femur and tibia SSMs were able to predict the shapes of the Russian patients with an RMSE 0.62 ± 0.1 mm for the femur and 0.71 ± 0.1 mm for the tibia. DISCUSSION. There were no significant differences in the ability of the SSM to predict femur and tibia shapes of patients in a new geographic region compared to the ability of the SSM to predict shapes within the training set. CONCLUSIONS. Based on this study, 11 different geographic locations in France, Germany, Austria, Italy and Australia provide a complete sample of the Caucasian population. Using an independent set of CT scans is a valuable tool to further validate the generalization ability of an SSM. For any figures or tables, please contact authors directly

INTRODUCTION. While computational models have been used for many years to contribute to pre-clinical, design phase iterations of total knee replacement implants, the analysis time required has limited the real-time use as required for other applications, such as in patient-specific surgical alignment in the operating room. In this environment, the impact of variation in ligament balance and implant alignment on estimated joint mechanics must be available instantaneously. As neural networks (NN) have shown the ability to appropriately represent dynamic systems, the objective of this preliminary study was to evaluate deep learning to represent the joint level kinetic and kinematic results from a validated finite element lower limb model with varied surgical alignment. METHODS. External hip and ankle boundary conditions were created for a previously-developed finite element lower limb model [1] for step down (SD), deep knee bend (DKB) and gait to best reproduce in-vivo loading conditions as measured on patients with the Innex knee (. orthoload.com. ) (Figure1). These boundary conditions were subsequently used as inputs for the model with a current fixed-bearing total knee replacement to estimate implant-specific kinetics and kinematics during activities of daily living. Implant alignments were varied, including variation of the hip-knee-ankle angle-±3°, the frontal plane joint line −7° to +5°, internal-external femoral rotation ±3°, and the tibial posterior slope 5° and 0°. Through varying these parameters a total of 2464 simulations were completed. A NN was created utilizing the NN toolbox in MATLAB. Sequence data inputs were produced from the alignment and the external boundary conditions for each activity cycle. Sequence outputs for the model were the 6 degree of freedom kinetics and kinematics, totaling 12 outputs. All data was normalized across the entire data set. Ten percent of the simulation runs were removed at random from the training set to be used for validation, leaving 2220 simulations for training and 244 for validation. A nine-layer bi-long short-term memory (LSTM) NN was created to take advantage of bi-LSTM layers ability to learn from past and future data. Training on the network was undertaken using an RMSprop solver until the root mean square error (RMSE) stopped reducing. Evaluation of NN quality was determined by the RMSE of the validation set. RESULTS. The trained NN was able to effectively estimate the validation data. Average RMSE over the kinetics of the validation data set was 140.7N/N∗m while the average RMSE over the kinematics of the validation data set was 4.47mm/deg (Figure 2,3–DKB, Gait shown). It is noted the error may be skewed by the larger magnitude kinetics and kinematics in the DKB activity as the average RMSE for just SD and gait was 85.9N/N∗m and 2.8mm/deg for the kinetics and kinematics, respectively. DISCUSSION. The accuracy of the generated NN indicates its potential for use in real-time modeling, and further work will explore additional changes in post-operative soft-tissue balance as well as scaling to patient-specific geometry

Introduction. Existing studies report more accurate implant placement with robotic-assisted unicompartmental knee arthroplasty (UKA); however, surgeon experience has not always been accounted for. The purpose of this study was to compare the accuracy of an experienced, high-volume surgeon to published data on robotic-assisted UKA tibial component alignment. Methods. One hundred thirty-one consecutive manual UKAs performed by a single surgeon using a cemented, fixed bearing implant were radiographically reviewed by an independent reviewer to avoid surgeon bias. Native and tibial implant slope and coronal alignment were measured on pre- and postoperative lateral and anteroposterior radiographs, respectively. Manual targets were set within 2° of native tibial slope and 0 to 2° varus tibial component alignment. Deviations from target were calculated as root mean square (RMS) errors and were compared to robotic-assisted UKA data. Results. One hundred twenty-eight UKAs were analyzed. The proportion of manual UKAs within the target for tibial component alignment (66%) exceeded published values comparing robotic (58%) to manual (41%) UKA. RMS error for tibial component alignment (1.5°) was less than published RMS error rates in robotic UKAs (range 1.8 to 5°). Fifty-eight percent of study UKAs were within the surgeon's preoperative goal for tibial slope, closer to published findings of 80% for robotic UKAs vs. 22% of manual UKAs. RMS error for tibial slope in study UKAs (1.5°) was smaller than RMS error rates for tibial slope in robotic UKAs (range 1.6 to 1.9°). Conclusion. These data demonstrate that an experienced, high-volume surgeon's accuracy in manual UKA can meet or exceed robotic-assisted UKA. Therefore, a surgeon's experience and aptitude should be taken into account when determining the value of robotics in knee arthroplasty. Further, the relationship between implant position and patient outcomes, and consensus on ideal surgical targets for optimal survivorship need further elucidation

Introduction. Machine learning is a relatively novel method to orthopaedics which can be used to evaluate complex associations and patterns in outcomes and healthcare data. The purpose of this study is to utilize 3 different supervised machine learning algorithms to evaluate outcomes from a multi-center international database of a single shoulder prosthesis to evaluate the accuracy of each model to predict post-operative outcomes of both aTSA and rTSA. Methods. Data from a multi-center international database consisting of 6485 patients who received primary total shoulder arthroplasty using a single shoulder prosthesis (Equinoxe, Exactech, Inc) were analyzed from 19,796 patient visits in this study. Specifically, demographic, comorbidity, implant type and implant size, surgical technique, pre-operative PROMs and ROM measures, post-operative PROMs and ROM measures, pre-operative and post-operative radiographic data, and also adverse event and complication data were obtained for 2367 primary aTSA patients from 8042 visits at an average follow-up of 22 months and 4118 primary rTSA from 11,754 visits at an average follow-up of 16 months were analyzed to create a predictive model using 3 different supervised machine learning techniques: 1) linear regression, 2) random forest, and 3) XGBoost. Each of these 3 different machine learning techniques evaluated the pre-operative parameters and created a predictive model which targeted the post-operative composite score, which was a 100 point score consisting of 50% post-operative composite outcome score (calculated from 33.3% ASES + 33.3% UCLA + 33.3% Constant) and 50% post-operative composite ROM score (calculated from S curves weighted by 70% active forward flexion + 15% internal rotation score + 15% active external rotation). 3 additional predictive models were created to control for the time required for patient improvement after surgery, to do this, each primary aTSA and primary rTSA cohort was subdivided to only include patient data follow-up visits >20 months after surgery, this yielded 1317 primary aTSA patients from 2962 visits at an average follow-up of 50 months and 1593 primary rTSA from 3144 visits at an average follow-up of 42 months. Each of these 6 predictive models were trained using a random selection of 80% of each cohort, then each model predicted the outcomes of the remaining 20% of the data based upon the demographic, comorbidity, implant type and implant size, surgical technique, pre-operative PROMs and ROM measures inputs of each 20% cohort. The error of all 6 predictive models was calculated from the root mean square error (RMSE) between the actual and predicted post-op composite score. The accuracy of each model was determined by subtracting the percent difference of each RMSE value from the average composite score associated with each cohort. Results. For all patient visits, the XGBoost decision tree algorithm was the most accurate model for both aTSA & rTSA patients, with an accuracy of ∼89.5% for both aTSA and rTSA. However for patients with 20+ month visits only, the random forest decision tree algorithm was the most accurate model for both aTSA & rTSA patients, with an accuracy of ∼89.5% for both aTSA and rTSA. The linear regression model was the least accurate predictive model for each of the cohorts analyzed. However, it should be noted that all 3 machine learning models provided accuracy of ∼85% or better and a RMSE <12. (Table 1) Figures 1 and 2 depict the typical spread and RMSE of the actual vs. predicted total composite score associated with the 3 models for aTSA (Figure 1) and rTSA (Figure 2). Discussion. The results of this study demonstrate that multiple different machine learning algorithms can be utilized to create models that predict outcomes with higher accuracy for both aTSA and rTSA, for numerous timepoints after surgery. Future research should test this model on different datasets and using different machine learning methods in order to reduce over- and under-fitting model errors. For any figures or tables, please contact the authors directly