µCT images are commonly analysed to assess changes in bone density and architecture in preclinical murine models. Several platforms provide automated analysis of bone architecture parameters from volumetric regions of interest (ROI). However, segmentation of the regions of subchondral bone to create the volumetric ROIs remains a manual and time-consuming task. This study aimed to develop and evaluate automated pipelines for trabecular bone architecture analysis of mouse proximal tibia subchondral bone.

A segmented dataset involving 62 knees (healthy and arthritic) from 10-week male C57BL/6 mice were used to train a U-Net type architecture, with µCT scans (downsampled) input that output segmentation and bone volume density (BV/TV) of the subchondral trabecular bone. Segmentations were upsampled and used in tandem with the original scans (10µ) as input for architecture analysis along with the thresholded trabecular bone. The analysis considered the manually and U-Net segmented ROIs using two available pipelines: the ITKBoneMorphometry library and CTan (SKYSCAN). The analyses included: bone volume (BV), total volume (TV), BV/TV, trabecular number (TbN), trabecular thickness (TbTh), trabecular separation (TbSp), and bone surface density (BSBV).

There was good agreement for bone measures between the manual and U-Net pipelines utilizing ITK (R=0.88-0.98) and CTan (R=0.91-0.98). ITK and CTan showed good agreement for BV, TV, BV/TV, TbTh and BSBV (R=0.9-0.98). However, a limited agreement was seen between TbN (R=0.73) and TbSb (R=0.59) due to methodological differences in how spacing is evaluated.

This U-Net/ITK pipeline seamlessly automated both segmentation and quantification of the proximal tibia subchondral bone. This automated pipeline allows the analysis of large volumes of data, and its open-source nature may enable the standardization of stereologic analysis of trabecular bone across different research groups.

Total knee and hip arthroplasty (TKA and THA) are the most commonly performed surgical procedures, the costs of which constitute a significant healthcare burden. Improving access to care for THA/TKA requires better efficiency. It is hypothesized that this may be possible through a two-stage approach that utilizes prediction of surgical time to enable optimization of operating room (OR) schedules.

Data from 499,432 elective unilateral arthroplasty procedures, including 302,490 TKAs, and 196,942 THAs, performed from 2014-2019 was extracted from the American College of Surgeons (ACS) National Surgical and Quality Improvement (NSQIP) database. A deep multilayer perceptron model was trained to predict duration of surgery (DOS) based on pre-operative clinical and biochemical patient factors. A two-stage approach, utilizing predicted DOS from a held out “test” dataset, was utilized to inform the daily OR schedule. The objective function of the optimization was the total OR utilization, with a penalty for overtime. The scheduling problem and constraints were simulated based on a high-volume elective arthroplasty centre in Canada. This approach was compared to current patient scheduling based on mean procedure DOS. Approaches were compared by performing 1000 simulated OR schedules.

The predict then optimize approach achieved an 18% increase in OR utilization over the mean regressor. The two-stage approach reduced overtime by 25-minutes per OR day, however it created a 7-minute increase in underutilization. Better objective value was seen in 85.1% of the simulations.

With deep learning prediction and mathematical optimization of patient scheduling it is possible to improve overall OR utilization compared to typical scheduling practices. Maximizing utilization of existing healthcare resources can, in limited resource environments, improve patient's access to arthritis care by increasing patient throughput, reducing surgical wait times and in the immediate future, help clear the backlog associated with the COVID-19 pandemic.

Physiotherapy is a critical element in successful conservative management of low back pain (LBP). The aim of this study was to develop and evaluate a system with wearable inertial sensors to objectively detect sitting postures and performance of unsupervised exercises containing movement in multiple planes (flexion, extension, rotation).

A set of 8 inertial sensors were placed on 19 healthy adult subjects. Data was acquired as they performed 7 McKenzie low-back exercises and 3 sitting posture positions. This data was used to train two models (Random Forest (RF) and XGBoost (XGB)) using engineered time series features. In addition, a convolutional neural network (CNN) was trained directly on the time series data. A feature importance analysis was performed to identify sensor locations and channels that contributed most to the models. Finally, a subset of sensor locations and channels was included in a hyperparameter grid search to identify the optimal sensor configuration and the best performing algorithm(s) for exercise classification. Models were evaluated using F1-score in a 10-fold cross validation approach.

The optimal hardware configuration was identified as a 3-sensor setup using lower back, left thigh, and right ankle sensors with acceleration, gyroscope, and magnetometer channels. The XBG model achieved the highest exercise (F1=0.94±0.03) and posture (F1=0.90±0.11) classification scores. The CNN achieved similar results with the same sensor locations, using only the accelerometer and gyroscope channels for exercise classification (F1=0.94±0.02) and the accelerometer channel alone for posture classification (F1=0.91±0.03).

This study demonstrates the potential of a 3-sensor lower body wearable solution (e.g. smart pants) that can identify proper sitting postures and exercises in multiple planes, suitable for low back pain. This technology has the potential to improve the effectiveness of LBP rehabilitation by facilitating quantitative feedback, early problem diagnosis, and possible remote monitoring.

Bone turnover and microdamage are impacted by skeletal metastases which can contribute to increased fracture risk. Treatments for metastatic disease may further impact bone quality. This study aimed to establish an understanding of microdamage accumulation and load to failure in healthy and osteolytic vertebrae following cancer treatment (stereotactic body radiotherapy (SBRT), zoledronic acid (ZA), or docetaxel (DTX)).

Forty-two 6-week old athymic female rats (Hsd:RH-Foxn1rnu, Envigo) were studied; 22 were inoculated with HeLa cervical cancer cells through intracardiac injection (day 0). Animals were randomly assigned to four groups: untreated (healthy=5, osteolytic=6), SBRT on day 14 (healthy=6, osteolytic=6), ZA on day 7 (healthy=4, osteolytic=5), and DTX on day 14 (healthy=5, osteolytic=5). Animals were euthanized on day 21. L1-L3 motion segments were compression loaded to failure and force-displacement data recorded. T13 vertebrae were stained with BaSO₄ and µCT imaged (90kVp, 44uA, 4.9µm) to visualize microdamage location and volume. Damage volume fraction (DV/BV) was calculated as the ratio of BaSO₄ to bone volume. Differences in mean load-to-failure were compared using three-way ANOVA (disease status, treatment, cells injected). Differences in mean DV/BV between treatment groups were compared using one-way ANOVA.

Treatment had a significant effect on load-to-failure (p=0.004) with ZA strengthening the healthy and osteolytic vertebrae. Reduced strength post SBRT seen in the metastatic (but not the healthy) group may be explained by greater tumor involvement secondary to higher cell injection concentrations. Untreated metastatic samples had higher DV/BV (16.25±2.54%) compared to all treatment groups (p<0.05) suggesting a benefit of treatment to bone quality.

Focal and systemic cancer treatments were shown to effect load-to-failure and microdamage accumulation in healthy and osteolytic vertebrae. Developing a better understanding of how treatments effect bone quality and mechanical stability is critical for effective management of patients with spinal metastases.

Access to health care, including physiotherapy, is increasingly occurring through virtual formats. At-home adherence to physical therapy programs is often poor and few tools exist to objectively measure low back physiotherapy exercise participation without the direct supervision of a medical professional. The aim of this study was to develop and evaluate the potential for performing automatic, unsupervised video-based monitoring of at-home low back physiotherapy exercises using a single mobile phone camera.

24 healthy adult subjects performed seven exercises based on the McKenzie low back physiotherapy program while being filmed with two smartphone cameras. Joint locations were automatically extracted using an open-source pose estimation framework. Engineered features were extracted from the joint location time series and used to train a support vector machine classifier (SVC). A convolutional neural network (CNN) was trained directly on the joint location time series data to classify exercises based on a recording from a single camera. The models were evaluated using a 5-fold cross validation approach, stratified by subject, with the class-balanced accuracy used as the performance metric.

Optimal performance was achieved when using a total of 12 pose estimation landmarks from the upper and lower body, with the SVC model achieving a classification accuracy of 96±4% and the CNN model an accuracy of 97±2%.

This study demonstrates the feasibility of using a smartphone camera and a supervised machine learning model to effectively assess at-home low back physiotherapy adherence. This approach could provide a low-cost, scalable method for tracking adherence to physical therapy exercise programs in a variety of settings.

Aims

An objective technological solution for tracking adherence to at-home shoulder physiotherapy is important for improving patient engagement and rehabilitation outcomes, but remains a significant challenge. The aim of this research was to evaluate performance of machine-learning (ML) methodologies for detecting and classifying inertial data collected during in-clinic and at-home shoulder physiotherapy exercise.

Excessive resident duty hours (RDH) are a recognized issue with implications for physician well-being and patient safety. A major component of the RDH concern is on-call duty. While considerable work has been done to reduce resident call workload, there is a paucity of research in optimizing resident call scheduling. Call coverage is scheduled manually rather than demand-based, which generally leads to over-scheduling to prevent a service gap. Machine learning (ML) has been widely applied in other industries to prevent such issues of a supply-demand mismatch. However, the healthcare field has been slow to adopt these innovations. As such, the aim of this study was to use ML models to 1) predict demand on orthopaedic surgery residents at a level I trauma centre and 2) identify variables key to demand prediction.

Daily surgical handover emails over an eight year (2012-2019) period at a level I trauma centre were collected. The following data was used to calculate demand: spine call coverage, date, and number of operating rooms (ORs), traumas, admissions and consults completed. Various ML models (linear, tree-based and neural networks) were trained to predict the workload, with their results compared to the current scheduling approach. Quality of models was determined by using the area under the receiver operator curve (AUC) and accuracy of the predictions. The top ten most important variables were extracted from the most successful model.

During training, the model with the highest AUC and accuracy was the multivariate adaptive regression splines (MARS) model, with an AUC of 0.78±0.03 and accuracy of 71.7%±3.1%. During testing, the model with the highest AUC and accuracy was the neural network model, with an AUC of 0.81 and accuracy of 73.7%. All models were better than the current approach, which had an AUC of 0.50 and accuracy of 50.1%. Key variables used by the neural network model were (descending order): spine call duty, year, weekday/weekend, month, and day of the week.

This was the first study attempting to use ML to predict the service demand on orthopaedic surgery residents at a major level I trauma centre. Multiple ML models were shown to be more appropriate and accurate at predicting the demand on surgical residents as compared to the current scheduling approach. Future work should look to incorporate predictive models with optimization strategies to match scheduling with demand in order to improve resident well being and patient care.

Single level discectomy (SLD) is one of the most commonly performed spinal surgery procedures. Two key drivers of their cost-of-care are duration of surgery (DOS) and postoperative length of stay (LOS). Therefore, the ability to preoperatively predict SLD DOS and LOS has substantial implications for both hospital and healthcare system finances, scheduling and resource allocation. As such, the goal of this study was to predict DOS and LOS for SLD using machine learning models (MLMs) constructed on preoperative factors using a large North American database.

The American College of Surgeons (ACS) National Surgical and Quality Improvement (NSQIP) database was queried for SLD procedures from 2014-2019. The dataset was split in a 60/20/20 ratio of training/validation/testing based on year. Various MLMs (traditional regression models, tree-based models, and multilayer perceptron neural networks) were used and evaluated according to 1) mean squared error (MSE), 2) buffer accuracy (the number of times the predicted target was within a predesignated buffer), and 3) classification accuracy (the number of times the correct class was predicted by the models). To ensure real world applicability, the results of the models were compared to a mean regressor model.

A total of 11,525 patients were included in this study. During validation, the neural network model (NNM) had the best MSEs for DOS (0.99) and LOS (0.67). During testing, the NNM had the best MSEs for DOS (0.89) and LOS (0.65). The NNM yielded the best 30-minute buffer accuracy for DOS (70.9%) and ≤120 min, >120 min classification accuracy (86.8%). The NNM had the best 1-day buffer accuracy for LOS (84.5%) and ≤2 days, >2 days classification accuracy (94.6%). All models were more accurate than the mean regressors for both DOS and LOS predictions.

We successfully demonstrated that MLMs can be used to accurately predict the DOS and LOS of SLD based on preoperative factors. This big-data application has significant practical implications with respect to surgical scheduling and inpatient bedflow, as well as major implications for both private and publicly funded healthcare systems. Incorporating this artificial intelligence technique in real-time hospital operations would be enhanced by including institution-specific operational factors such as surgical team and operating room workflow.

Bone turnover and the accumulation of microdamage are impacted by the presence of skeletal metastases which can contribute to increased fracture risk. Treatments for metastatic disease may further impact bone quality. The present study aims to establish a preliminary understanding of microdamage accumulation and load to failure in osteolytic vertebrae following stereotactic body radiotherapy (SBRT), zoledronic acid (ZA), or docetaxel (DTX) treatment.

Twenty-two six-week old athymic female rats (Hsd:RH-Foxn1rnu, Envigo, USA) were inoculated with HeLa cervical cancer cells through intracardiac injection (day 0). Institutional approval was obtained for this work and the ARRIVE guidelines were followed. Animals were randomly assigned to four groups: untreated (n=6), spine stereotactic body radiotherapy (SBRT) administered on day 14 (n=6), zoledronic acid (ZA) administered on day 7 (n=5), and docetaxel (DTX) administered on day 14 (n=5). Animals were euthanized on day 21. T13-L3 vertebral segments were collected immediately after sacrifice and stored in −20°C wrapped in saline soaked gauze until testing. µCT scans (µCT100, Scanco, Switzerland) of the T13-L3 segment confirmed tumour burden in all T13 and L2 vertebrae prior to testing. T13 was stained with BaSO₄ to label microdamage. High resolution µCT scans were obtained (90kVp, 44uA, 4W, 4.9µm voxel size) to visualize stain location and volume. Segmentations of bone and BaSO₄ were created using intensity thresholding at 3000HU (~736mgHA/cm³) and 10000HU (~2420mgHA/cm³), respectively. Non-specific BaSO₄ was removed from the outer edge of the cortical shell by shrinking the segmentation by 105mm in 3D. Stain volume fraction was calculated as the ratio of BaSO₄ volume to the sum of BaSO₄ and bone volume. The L1-L3 motion segments were loaded under axial compression to failure using a µCT compatible loading device (Scanco) and force-displacement data was recorded. µCT scans were acquired unloaded, at 1500µm displacement and post-failure. Stereological analysis was performed on the L2 vertebrae in the unloaded µCT scans. Differences in mean stain volume fraction, mean load to failure, and mean bone volume/total volume (BV/TV) were compared between treatment groups using one-way ANOVAs. Pearson's correlation between stain volume fraction and load to failure by treatment was calculated using an adjusted load to failure divided by BV/TV.

Stained damage fraction was significantly different between treatment groups (p=0.0029). Tukey post-hoc analysis showed untreated samples to have higher stain volume fraction (16.25±2.54%) than all treatment groups (p<0.05). The ZA group had the highest mean load to failure (195.60±84.49N), followed by untreated (142.33±53.08N), DTX (126.60±48.75N), and SBRT (95.50±44.96N), but differences did not reach significance (p=0.075). BV/TV was significantly higher in the ZA group (49.28±3.56%) compared to all others. The SBRT group had significantly lower BV/TV than the untreated group (p=0.018). Load divided by BV/TV was not significantly different between groups (p=0.24), but relative load to failure results were consistent (ZA>Untreated>DTX>SBRT). No correlations were found between stain volume fraction and load to failure.

Focal and systemic cancer treatments effect microdamage accumulation and load to failure in osteolytic vertebrae. Current testing of healthy controls will help to further separate the effects of the tumour and cancer treatments on bone quality.

Bone turnover and the accumulation of microdamage are impacted by the presence of skeletal metastases which can contribute to increased fracture risk. Treatments for metastatic disease may further impact bone quality. The present study aims to establish a preliminary understanding of microdamage accumulation and load to failure in osteolytic vertebrae following stereotactic body radiotherapy (SBRT), zoledronic acid (ZA), or docetaxel (DTX) treatment.

Twenty-two six-week old athymic female rats (Hsd:RH-Foxn1rnu, Envigo, USA) were inoculated with HeLa cervical cancer cells through intracardiac injection (day 0). Institutional approval was obtained for this work and the ARRIVE guidelines were followed. Animals were randomly assigned to four groups: untreated (n=6), spine stereotactic body radiotherapy (SBRT) administered on day 14 (n=6), zoledronic acid (ZA) administered on day 7 (n=5), and docetaxel (DTX) administered on day 14 (n=5). Animals were euthanized on day 21. T13-L3 vertebral segments were collected immediately after sacrifice and stored in −20°C wrapped in saline soaked gauze until testing. µCT scans (µCT100, Scanco, Switzerland) of the T13-L3 segment confirmed tumour burden in all T13 and L2 vertebrae prior to testing. T13 was stained with BaSO₄ to label microdamage. High resolution µCT scans were obtained (90kVp, 44uA, 4W, 4.9µm voxel size) to visualize stain location and volume. Segmentations of bone and BaSO₄ were created using intensity thresholding at 3000HU (~736mgHA/cm³) and 10000HU (~2420mgHA/cm³), respectively. Non-specific BaSO₄ was removed from the outer edge of the cortical shell by shrinking the segmentation by 105mm in 3D. Stain volume fraction was calculated as the ratio of BaSO₄ volume to the sum of BaSO₄ and bone volume. The L1-L3 motion segments were loaded under axial compression to failure using a µCT compatible loading device (Scanco) and force-displacement data was recorded. µCT scans were acquired unloaded, at 1500µm displacement and post-failure. Stereological analysis was performed on the L2 vertebrae in the unloaded µCT scans. Differences in mean stain volume fraction, mean load to failure, and mean bone volume/total volume (BV/TV) were compared between treatment groups using one-way ANOVAs. Pearson's correlation between stain volume fraction and load to failure by treatment was calculated using an adjusted load to failure divided by BV/TV.

Stained damage fraction was significantly different between treatment groups (p=0.0029). Tukey post-hoc analysis showed untreated samples to have higher stain volume fraction (16.25±2.54%) than all treatment groups (p<0.05). The ZA group had the highest mean load to failure (195.60±84.49N), followed by untreated (142.33±53.08N), DTX (126.60±48.75N), and SBRT (95.50±44.96N), but differences did not reach significance (p=0.075). BV/TV was significantly higher in the ZA group (49.28±3.56%) compared to all others. The SBRT group had significantly lower BV/TV than the untreated group (p=0.018). Load divided by BV/TV was not significantly different between groups (p=0.24), but relative load to failure results were consistent (ZA>Untreated>DTX>SBRT). No correlations were found between stain volume fraction and load to failure.

Focal and systemic cancer treatments effect microdamage accumulation and load to failure in osteolytic vertebrae. Current testing of healthy controls will help to further separate the effects of the tumour and cancer treatments on bone quality.

Total knee and hip arthroplasty (TKA and THA) are two of the highest volume and resource intensive surgical procedures. Key drivers of the cost of surgical care are duration of surgery (DOS) and postoperative inpatient length of stay (LOS). The ability to predict TKA and THA DOS and LOS has substantial implications for hospital finances, scheduling and resource allocation. The goal of this study was to predict DOS and LOS for elective unilateral TKAs and THAs using machine learning models (MLMs) constructed on preoperative patient factors using a large North American database.

The American College of Surgeons (ACS) National Surgical and Quality Improvement (NSQIP) database was queried for elective unilateral TKA and THA procedures from 2014-2019. The dataset was split into training, validation and testing based on year. Multiple conventional and deep MLMs such as linear, tree-based and multilayer perceptrons (MLPs) were constructed. The models with best performance on the validation set were evaluated on the testing set. Models were evaluated according to 1) mean squared error (MSE), 2) buffer accuracy (the number of times the predicted target was within a predesignated buffer of the actual target), and 3) classification accuracy (the number of times the correct class was predicted by the models). To ensure useful predictions, the results of the models were compared to a mean regressor.

A total of 499,432 patients (TKA 302,490; THA 196,942) were included. The MLP models had the best MSEs and accuracy across both TKA and THA patients. During testing, the TKA MSEs for DOS and LOS were 0.893 and 0.688 while the THA MSEs for DOS and LOS were 0.895 and 0.691. The TKA DOS 30-minute buffer accuracy and ≤120 min, >120 min classification accuracy were 78.8% and 88.3%, while the TKA LOS 1-day buffer accuracy and ≤2 days, >2 days classification accuracy were 75.2% and 76.1%. The THA DOS 30-minute buffer accuracy and ≤120 min, >120 min classification accuracy were 81.6% and 91.4%, while the THA LOS 1-day buffer accuracy and ≤2 days, >2 days classification accuracy were 78.3% and 80.4%. All models across both TKA and THA patients were more accurate than the mean regressors for both DOS and LOS predictions across both buffer and classification accuracies.

Conventional and deep MLMs have been effectively implemented to predict the DOS and LOS of elective unilateral TKA and THA patients based on preoperative patient factors using a large North American database with a high level of accuracy. Future work should include using operational factors to further refine these models and improve predictive accuracy. Results of this work will allow institutions to optimize their resource allocation, reduce costs and improve surgical scheduling.

The spine is a common site of metastasis. Complications include pathologic fracture, spinal cord compression, and neurological deficits. Vertebroplasty (VP) and Balloon Kyphoplasty (KP) are minimally invasive stabilization procedures used as a palliative treatment to improve mechanical stability, quality of life, and reduce pain. Photodynamic therapy (PDT) is a tumour-ablative modality that may complement mechanical stability afforded by VP/KP. This first-in-human study evaluates PDT safety when applied in conjunction with VP/KP.

This dose escalation trial involved one light only control group and four light-drug doses (50,100,150,200J;n=6) delivered at 150mW from a 690nm diode laser by 800-micron optical fibers prior to KP/VP. Patients eligible for VP/KP in treating pathologic fracture or at-risk lesions at a single level were recruited. Exclusion criteria included spinal canal compromise or neurologic impairment. PDT is a two-step binary therapy of systemic drug followed by intravertebral light activation. Light was applied via bone trochar prior to cementation. This study used a benzoporphyrin derivative monoacid (BPD-MA), Verteporfin (VisudyneTm), as the photosensitizer drug in the therapy. Drug/light safety, neurologic safety, generic (SF-36), and disease-specific outcomes (VAS, EORTC-QLQ-BM22, EORTC-QLQ-C15-PAL) were recorded through six weeks. Phototoxicity and the side effects of the BPD-MA were also examined following PDT use.

Thirty (10 male, 20 female) patients were treated (13 KP, 17 VP). The average age was 61 and significantly different between genders (Male 70yrs vs. Female 57yrs: p 0.05), and tumour status (lytic vs. mixed blastic/lytic: p>0.05). In most cases, fluence rates were similar throughout PDT treatment time, indicating a relatively stable treatment. Twelve (40%) of patients experienced complications during the study, none of which were attributed to PDT therapy. This included two kyphoplasty failures due to progression of disease, one case of shingles, one ankle fracture, one prominent suture, one case of constipation due to a lung lesion, one case of fatigue, and five patients experienced pain that was surgically related or preceded therapy.

Vertebral PDT appears safe from pharmaceutical and neurologic perspectives. KP/VP failure rate is broadly in line with reported values and PDT did not compromise efficacy. The 50J group demonstrated an improved response. Ongoing study determining safe dose range and subsequent efficacy studies are necessary.

Rotator cuff tears are the most common cause of shoulder disability, affecting 10% of the population under 60 and 40% of those aged 70 and above. Massive irreparable rotator cuff tears account for 30% of all tears and their management continues to be an orthopaedic challenge. Traditional surgical techniques, that is, tendon transfers are performed to restore shoulder motion, however, they result in varying outcomes of stability and complications. Superior capsular reconstruction (SCR) is a novel technique that has shown promise in restoring shoulder function, albeit in limited studies. To date, there has been no biomechanical comparison between these techniques. This study aims to compare three surgical techniques (SCR, latissimus dorsi tendon transfer and lower trapezius tendon transfer) for irreparable rotator cuff tears with respect to intact cuff control using a clinically relevant biomechanical outcome of rotational motion.

Eight fresh-frozen shoulder specimens with intact rotator cuffs were tested. After dissection of subcutaneous tissue and muscles, each specimen was mounted on a custom shoulder testing apparatus and physiologic loads were applied using a pulley setup. Under 2.2 Nm torque loading maximum internal and external rotation was measured at 0 and 60 degrees of glenohumeral abduction. Repeat testing was conducted after the creation of the cuff tear and subsequent to the three repair techniques. Repeated measures analysis with paired t-test comparisons using Sidak correction was performed to compare the rotational range of motion following each repair technique with respect to each specimen's intact control. P-values of 0.05 were considered significant.

At 0° abduction, internal rotation increased after the tear (intact: 39.6 ± 13.6° vs. tear: 80.5 ± 47.7°, p=0.019). Internal rotation was higher following SCR (52.7 ± 12.9°, intact - SCR 95% CI: −25.28°,-0.95°, p=0.034), trapezius transfer (74.2 ± 25.3°, intact – trapezius transfer: 95% CI: −71.1°, 1.81°, p=0.064), and latissimus transfer (83.5 ± 52.1°, intact – latissimus transfer: 95% CI: −118.3°, 30.5°, p=0.400) than in intact controls. However, internal rotation post SCR yielded the narrowest estimate range close to intact controls. At 60° abduction, internal rotation increased after the tear (intact: 38.7 ± 14.4° vs. tear: 49.5 ± 13°, p=0.005). Internal rotation post SCR did not differ significantly from intact controls (SCR: 49.3 ± 10.1°, intact – SCR: 95% CI: −28°, 6.91°, p=0.38). Trapezius transfer showed a trend toward significantly higher internal rotation (65.7 ± 21.1°, intact – trapezius transfer: 95% CI: −55.7°, 1.7°, p=0.067), while latissimus transfer yielded widely variable rotation angle (65.7 ± 38°, intact – latissimus transfer: 95% CI: −85.9°, 31.9°, p=0.68). There were no significant differences in external rotation for any technique at 0° or 60° abduction.

Preliminary evaluation in this cadaveric biomechanical study provides positive evidence in support of use of SCR as a less morbid surgical option than tendon transfers. The cadaveric nature of this study limits the understanding of the motion to post-operative timepoint and the results herein are relevant for otherwise normal shoulders only. Further clinical evaluation is warranted to understand the long-term outcomes related to shoulder function and stability post SCR.

Spinal stenosis is a condition resulting in the compression of the neural elements due to narrowing of the spinal canal. Anatomical factors including enlargement of the facet joints, thickening of the ligaments, and bulging or collapse of the intervertebral discs contribute to the compression. Decompression surgery alleviates spinal stenosis through a laminectomy involving the resection of bone and ligament. Spinal decompression surgery requires appropriate planning and variable strategies depending on the specific situation. Given the potential for neural complications, there exist significant barriers to residents and fellows obtaining adequate experience performing spinal decompression in the operating room. Virtual teaching tools exist for learning instrumentation which can enhance the quality of orthopaedic training, building competency and procedural understanding. However, virtual simulation tools are lacking for decompression surgery. The aim of this work was to develop an open-source 3D virtual simulator as a teaching tool to improve orthopaedic training in spinal decompression.

A custom step-wise spinal decompression simulator workflow was built using 3D Slicer, an open-source software development platform for medical image visualization and processing. The procedural steps include multimodal patient-specific loading and fusion of Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) data, bone threshold-based segmentation, soft tissue segmentation, surgical planning, and a laminectomy and spinal decompression simulation. Fusion of CT and MRI elements was achieved using Fiducial-Based Registration which aligned the scans based on manually placed points allowing for the identification of the relative position of soft and hard tissues. Soft tissue segmentation of the spinal cord, the cerebrospinal fluid, the cauda equina, and the ligamentum flavum was performed using Simple Region Growing Segmentation (with manual adjustment allowed) involving the selection of structures on T1 and/or T2-weighted scans. A high-fidelity 3D model of the bony and soft tissue anatomy was generated with the resulting surgical exposure defined by labeled vertebrae simulating the central surgical incision. Bone and soft tissue resecting tools were developed by customizing manual 3D segmentation tools. Simulating a laminectomy was enabled through bone and ligamentum flavum resection at the site of compression. Elimination of the stenosis enabled decompression of the neural elements simulated by interpolation of the undeformed anatomy above and below the site of compression using Fill Between Slices to reestablish pre-compression neural tissue anatomy.

The completed workflow allows patient specific simulation of decompression procedures by staff surgeons, fellows and residents. Qualitatively, good visualization was achieved of merged soft tissue and bony anatomy. Procedural accuracy, the design of resecting tools, and modeling of the impact of bone and ligament removal was found to adequately encompass important challenges in decompression surgery.

This software development project has resulted in a well-characterized freely accessible tool for simulating spinal decompression surgery. Future work will integrate and evaluate the simulator within existing orthopaedic resident competency-based curriculum and fellowship training instruction. Best practices for effectively teaching decompression in tight areas of spinal stenosis using virtual simulation will also be investigated in future work.

Quantitative assessment of metastatic involvement of the bony spine is important for assessing disease progression and treatment response. Quantification of metastatic involvement is challenging as tumours may appear as osteolytic (bone resorbing), osteoblastic (bone forming) or mixed. This investigation aimed to develop an automated method to accurately segment osteoblastic lesions in a animal model of metastatically involved vertebrae, imaged with micro computed tomography (μCT).

Radiomics seeks to apply standardized features extracted from medical images for the purpose of decision-support as well as diagnosis and treatment planning. Here we investigate the application of radiomic-based features for the delineation of osteoblastic vertebral metastases. Osteoblastic lesions affect bone deposition and bone quality, resulting in a change in the texture of bony material physically seen through μCT imaging. We hypothesize that radiomics based features will be sensitive to changes in osteoblastic lesion bone texture and that these changes will be useful for automating segmentation.

Osteoblastic metastases were generated via intracardiac injection of human ZR-75-1 breast cancer cells into a preclinical athymic rat model (n=3). Four months post inoculation, ex-vivo μCT images (µCT100, Scanco) were acquired of each rodent spine focused on the metastatically involved third lumbar vertebra (L3) at 7µm/voxel and resampled to 34µm/voxel.

The trabecular bone within each vertebra was isolated using an atlas and level-set based segmentation approach previously developed by our group. Pyradiomics, an open source Radiomics library written in python, was used to calculate 3D image features at each voxel location within the vertebral bone. Thresholding of each radiomic feature map was used to isolate the osteoblastic lesions.

The utility of radiomic feature-based segmentation of osteoblastic bone tissue was evaluated on randomly selected 2D sagittal and axial slices of the μCT volume. Feature segmentations were compared to ground truth osteoblastic lesion segmentations by calculating the Dice Similarity Coefficient (DSC). Manually defined ground truth osteoblastic tumor segmentations on the μCT slices were informed by histological confirmation of the lesions.

The radiomic based features that best segmented osteoblastic tissue while optimizing computational time were derived from the Neighbouring Gray Tone Difference Matrix (NGTDM). Measures of coarseness yielded the best agreement with the manual segmentations (DSC=707%) followed by contrast, strength and complexity (DSC=6513%, 5428%, and 4826%, respectively).

This pilot study using a radiomic based approach demonstrates the utility of the NGTDM features for segmentation of vertebral osteoblastic lesions. This investigation looked at the utility of isolated features to segment osteoblastic lesions and found modest performance in isolation. In future work we will explore combining these features using machine learning based classifiers (i.e. decision forests, support vector machines, etc.) to improve segmentation performance.

Anatomic studies have demonstrated that bipolar glenoid and humeral bone loss have a cumulative impact on shoulder instability, and that these defects may engage in functional positions depending on their size, location, and orientation, potentially resulting in failure of stabilisation procedures. Determining which lesions pose a risk for engagement remains a challenge, with Itoi's 3DCT based glenoid track method and arthroscopic assessment being the accepted approaches at this time. The purpose of this study was to investigate the interaction of humeral and glenoid bone defects on shoulder engagement in a cadaveric model. Two alternative approaches to predicting engagement were evaluated; 1) CT scanning the shoulder in abduction and external rotation 2) measurement of Bankart lesion width and a novel parameter, the intact anterior articular angle (IAAA), on conventional 2D multi-plane reformats.

Hill-Sachs and Bony Bankart defects of varying size were created in 12 cadaveric upper limbs, producing 45 bipolar defect combinations. The shoulders were assessed for engagement using cone beam CT in various positions of function, from 30 to 90 degrees of both abduction and external rotation. The humeral and glenoid defects were characterised by measurement of their size, location, and orientation. The abduction external rotation scan and 2D IAAA approaches were compared to the glenoid track method for predicting engagement.

Engagement was predicted by Itoi's glenoid track method in 24 of 45 specimens (53%). The abduction external rotation CT scan performed at 60 degrees of glenohumeral abduction (corresponding to 90 degrees of abduction relative to the trunk) and 90 degrees of external rotation predicted engagement accurately in 43 of 45 specimens (96%), with sensitivity and specificity of 92% and 100% respectively. A logistic model based on Bankart width and IAAA provided a prediction accuracy of 89% with sensitivity and specificity of 91% and 87%. Inter-rater agreement was excellent (Kappa = 1) for classification of engagement on the abduction external rotation CT, and good (intraclass correlation = 0.73) for measurement of IAAA.

Bipolar lesions at risk for engagement can be identified using an abduction external rotation CT scan at 60 degrees of glenohumeral abduction and 90 degrees of external rotation, or by performing 2D measurements of Bankart width and IAAA on conventional CT multi-plane reformats. This information will be useful for peri-operative decision making around surgical techniques for shoulder stabilisation in the setting of bipolar bone defects.

Predictable fracture healing fails to occur in 5–10% of cases. This is particularly concerning among individuals with osteoporosis. With an increasing aging population, one in three women and one in five men above the age of 50 experience fragility fractures. As such, there is a critical need for an effective treatment option that could enhance fracture healing in osteoporotic bone. Lithium, the standard treatment for bipolar disorder, has been previously shown to improve fracture healing through modulation of the Wnt/beta-catenin pathway. We optimised the precise oral lithium administration parameters to improve mechanical strength and enhance healing of femoral fractures in healthy rats. A low dose of Lithium (20 mg/kg) administered seven days post fracture for a two week duration improved torsional strength by 46% at four weeks post fracture compared to non-treated animals. Application of lithium to enhance fracture healing in osteoporotic bone would have a significant healthcare impact and requires further study. Aim: To evaluate the efficacy of optimal lithium administration post fracture on quality of fracture healing in a rat osteoporotic model. Hypothesis: Lithium treatment in osteoporotic rats will improve the structural and mechanical properties of the healing bone despite the impaired nature of bone tissue.

Sprague Dawley female rats (∼350 g, age ∼3 months) were bilaterally ovariectomised and maintained for 3 months to establish the osteoporotic phenotype. A unilateral, closed mid-shaft femoral fracture was created using a weight-drop apparatus. At seven days post fracture, the treatment group received 20 mg/kg-wt lithium chloride via oral gavage daily for 14 days. The control group received an equivalent dose of saline. All animals were sacrificed at day 28 and the femurs harvested bilaterally. Treatment efficacy was evaluated based on torsional loading and stereologic analysis.

Lithium treatment positively impacted the healing femurs, with an average yield torque ∼1.25-fold higher than in the saline group (200±36 vs. 163±31 N-mm, p=0.15). Radiographically, the lithium-treated rats had a high level of restored periosteal continuity, larger bridging and intercortical callus at the fracture site. These hallmarks of healing were generally absent in the saline group. The Lithium group had significantly higher total volume (624±32 vs. 568±95 mm3), lower bone volume fraction (41±4 vs. 50±5%) and higher theoretical torsional rigidity (477±50 vs. 357±93 kN-mm2) compared to the saline group. Torsional strength and stereology values were similar for the contralateral femurs of the two groups.

Lithium was found to enhance fracture healing in osteoporotic bone under the dosing regimen optimised in healthy femora. This is promising data as treatment represents an easily translatable pharmacological intervention for fracture healing that may ultimately reduce the healthcare burden of osteoporotic fractures.

Strain is a robust indicator of bone failure initiation. Previous work has demonstrated the measurement of vertebral trabecular bone strain by Digital Volume Correlation (DVC) of µCT scan in both a loaded and an unloaded configuration. This project aims to improve previous strain measurement methods relying on image registration, improving resolution to resolve trabecula level strain and to improve accuracy by applying feature based registration algorithms to µCT images of vertebral trabecular bone to quantify strain. It is hypothesised that extracting reliable corresponding feature points from loaded and unloaded µCT scans can be used to produce higher resolution strain fields compared to DVC techniques.

The feature based strain calculation algorithm has two steps: 1) a displacement field is calculated by finding corresponding feature points identified in both the loaded and unloaded µCT scans 2) strain fields are calculated from the displacement fields. Two methods of feature point extraction, Scale Invariant Feature Transform (SIFT) and Skeletonisation, were applied to unloaded (fixed) and loaded (moving) µCT images of a rat tail vertebra. Spatially non-uniform displacement fields were generated by automatically matching corresponding feature points in the unloaded and loaded scans. The Thin Plate Spline method and a Moving Least Squares Meshless Method were both tested for calculating strain from the displacement fields. Verification of the algorithms was performed by testing against known artificial strain/displacement fields. A uniform and a linearly varying 2% compressive strain field were applied separately to an unloaded 2D sagittal µCT slice to simulate the moving image.

SIFT was unable to reliably match identified feature points leading to large errors in displacement. Skeletonisation generated a more accurate and precise displacement field. TPS was not tolerant to small displacement field errors, which resulted in inaccurate strain fields. The Meshless Methods proved much more resilient to displacement field errors. The combination of Skeletonisation with the Meshless Method resulted in best performance with an accuracy of −405µstrain and a detection limit of 1210µstrain at a strain resolution of 221.5µm. The DVC algorithm verified using the same validation test yielded a similar detection limit (1190µstrain), but with a lower accuracy for the same test (2370µstrain) for a lower resolution strain field (770µm) (Hardisty, 2009).

The Skeletonisation algorithm combined with the Meshless Method calculated strain at a higher resolution, but with a similar detection limit, to that of traditional DVC methods. Future improvements to this method include the implementation of subpixel feature point identification and adapting this method of strain measurement into a 3D domain. Ultimately, a hybrid DVC/feature registration algorithm may further improve the ability to measure trabecular bone strain using µCT based image registration.

Simulation is an effective adjunct to the traditional surgical curriculum, though access to these technologies is often limited and costly. The objectives of this work were to develop a freely accessible virtual pedicle screw simulator and to improve the clinical authenticity of the simulator through integration of low-cost motion tracking. The open-source medical imaging and visualisation software, 3D Slicer, was used as the development platform for the virtual simulation. 3D Slicer contains many features for quickly rendering and transforming 3D models of the bony spine anatomy from patient-specific CT scans. A step-wise pedicle screw insertion workflow module was developed which emulated typical pre-operative planning steps. This included taking anatomic measurements, identifying insertion landmarks, and choosing appropriate screw sizes. Monitoring of the surgeon's simulated tool was assessed with a low-cost motion tracking sensor in real-time. This allowed for the surgeon's physical motions to be tracked as they defined the virtual screw's insertion point and trajectory on the rendered anatomy. Screw insertion was evaluated based on bone density contact and cortical breaches. Initial surgeon feedback of the virtual simulator with integrated motion tracking was positive, with no noticeable lag and high accuracy between the real-world and virtual environments. The software yields high fidelity 3D visualisation of the complex geometry and the tracking enabled coordination of motion to small changes in both translational and angular positioning. Future work will evaluate the benefit of this simulation platform with use over the course of resident spine rotations to improve planning and surgical competency.

Summary

A novel bipolar cooled radiofrequency ablation probe, optimised for bone metastases applications, is shown in two preclinical models to offer a safe and minimally invasive treatment option that can ablate large tissue volumes and preserve the regenerative ability of bone.