Application of deep learning for automated diagnosis and classification of hip dysplasia on plain radiographs.

BACKGROUND: Hip dysplasia is a condition where the acetabulum is too shallow to support the femoral head and is commonly considered a risk factor for hip osteoarthritis. The objective of this study was to develop a deep learning model to diagnose hip dysplasia from plain radiographs and classify dysplastic hips based on their severity. METHODS: We collected pelvic radiographs of 571 patients from two single-center cohorts and one multicenter cohort. The radiographs were split in half to create hip radiographs (n = 1022). One orthopaedic surgeon and one resident assessed the radiographs for hip dysplasia on either side. We used the center edge (CE) angle as the primary diagnostic criteria. Hips with a CE angle < 20°, 20° to 25°, and > 25° were labeled as dysplastic, borderline, and normal, respectively. The dysplastic hips were also classified with both Crowe and Hartofilakidis classification of dysplasia. The dataset was divided into train, validation, and test subsets using 80:10:10 split-ratio that were used to train two deep learning models to classify images into normal, borderline and (1) Crowe grade 1-4 or (2) Hartofilakidis grade 1-3. A pre-trained on Imagenet VGG16 convolutional neural network (CNN) was utilized by performing layer-wise fine-turning. RESULTS: Both models struggled with distinguishing between normal and borderline hips. However, achieved high accuracy (Model 1: 92.2% and Model 2: 83.3%) in distinguishing between normal/borderline vs. dysplastic hips. The overall accuracy of Model 1 was 68% and for Model 2 73.5%. Most misclassifications for the Crowe and Hartofilakidis classifications were +/- 1 class from the correct class. CONCLUSIONS: This pilot study shows promising results that a deep learning model distinguish between normal and dysplastic hips with high accuracy. Future research and external validation are warranted regarding the ability of deep learning models to perform complex tasks such as identifying and classifying disorders using plain radiographs. LEVEL OF EVIDENCE: Diagnostic level IV.

The use of deep learning enables high diagnostic accuracy in detecting syndesmotic instability on weight-bearing CT scanning.

PURPOSE: Delayed diagnosis of syndesmosis instability can lead to significant morbidity and accelerated arthritic change in the ankle joint. Weight-bearing computed tomography (WBCT) has shown promising potential for early and reliable detection of isolated syndesmotic instability using 3D volumetric measurements. While these measurements have been reported to be highly accurate, they are also experience-dependent, time-consuming, and need a particular 3D measurement software tool that leads the clinicians to still show more interest in the conventional diagnostic methods for syndesmotic instability. The purpose of this study was to increase accuracy, accelerate analysis time, and reduce interobserver bias by automating 3D volume assessment of syndesmosis anatomy using WBCT scans. METHODS: A retrospective study was conducted using previously collected WBCT scans of patients with unilateral syndesmotic instability. One-hundred and forty-four bilateral ankle WBCT scans were evaluated (48 unstable, 96 control). We developed three deep learning models for analyzing WBCT scans to recognize syndesmosis instability. These three models included two state-of-the-art models (Model 1-3D Convolutional Neural Network [CNN], and Model 2-CNN with long short-term memory [LSTM]), and a new model (Model 3-differential CNN LSTM) that we introduced in this study. RESULTS: Model 1 failed to analyze the WBCT scans (F1 score = 0). Model 2 only misclassified two cases (F1 score = 0.80). Model 3 outperformed Model 2 and achieved a nearly perfect performance, misclassifying only one case (F1 score = 0.91) in the control group as unstable while being faster than Model 2. CONCLUSIONS: In this study, a deep learning model for 3D WBCT syndesmosis assessment was developed that achieved very high accuracy and accelerated analytics. This deep learning model shows promise for use by clinicians to improve diagnostic accuracy, reduce measurement bias, and save both time and expenditure for the healthcare system. LEVEL OF EVIDENCE: II.

In Vivo Length Changes Between the Attachments of the Medial Patellofemoral Complex Fibers in Knees With Anatomic Risk Factors for Patellar Instability.

BACKGROUND: Medial patellofemoral complex (MPFC) reconstruction plays an important role in the surgical treatment of patellar instability. Anatomic reconstruction is critical in re-creating the native function of the ligament, which includes minimizing length changes that occur in early flexion. Anatomic risk factors for patellar instability such as trochlear dysplasia, patella alta, and increased tibial tuberosity to trochlear groove (TT-TG) distance have been shown to influence the function of the MPFC graft in cadaveric studies, but the native length change patterns of the MPFC fibers in knees with anatomic risk factors have not been described. PURPOSE: To describe the in vivo length changes of the MPFC fibers in knees with anatomic risk factors for patellar instability and identify the optimal attachment sites for MPFC reconstruction. STUDY DESIGN: Controlled laboratory study. METHODS: Dynamic computed tomography imaging was performed on the asymptomatic knee in patients with contralateral patellar instability. Three-dimensional digital knee models were created to assess knees between 0° and 50° of flexion in 10° increments. MPFC fiber lengths were calculated at each flexion angle between known anatomic attachment points on the extensor mechanism (quadriceps tendon, MPFC midpoint [M], and patella) and femur (1, 2, and 3, representing the proximal to distal femoral footprint). Changes in MPFC fiber length were compared for each condition and assessed for their relationships to morphologic risk factors (trochlear depth, Caton Deschamps Index [CDI], and TT-TG distance). RESULTS: In 22 knees, native MPFC fibers were found to be longer at 0° than at 20° to 50° of flexion. Length changes observed between 0° and 50° increased with the number of risk factors present. In the central fibers of the MPFC (M-2), 1.7% ± 3.1% length change was noted in knees with no anatomic risk factors, which increased to 5.6% ± 4.6%, 17.0% ± 6.4%, and 26.7% ± 6.8% in the setting of 1, 2, and 3 risk factors, respectively. Nonanatomic patella-based attachments were more likely to demonstrate unfavorable length change patterns, in which length was greater at 50° than 0°. In patellar attachments, an independent relationship was found between increasing length changes and TT-TG distance, while in quadriceps tendon attachments, a trend toward a negative relationship between length changes and CDI was noted. All configurations demonstrated a strong relationship between percentage change in length and number of morphologic risk factors present, with the greatest influence found in patella-based attachments. CONCLUSION: The MPFC fibers demonstrated increased length changes in knees when a greater number of morphological risk factors for patellar instability were present, which worsened in the setting of nonanatomic configurations. This suggests that the function of the intact MPFC in patients with anatomic risk factors may not reflect previously described findings in anatomically normal knees. Further studies are needed to understand the pathoanatomy related to these changes, as well as the implications for graft placement and assessment of length changes during MPFC reconstruction techniques. CLINICAL RELEVANCE: MPFC length change patterns vary based on the number of morphologic risk factors for patellar instability present and should be considered during reconstructive procedures.

Redefining Trochlear Dysplasia: Normal Thresholds Vary by Measurement Technique, Landmarks, and Sex.

BACKGROUND: Trochlear dysplasia is a known risk factor for patellar instability. Multiple radiographic measurements exist to assess trochlear morphology, but the optimal measurement technique and threshold for instability are unknown. PURPOSE: To describe the optimal measurements and thresholds for trochlear dysplasia on magnetic resonance imaging (MRI) that can identify knees with patellar instability in male and female patients. STUDY DESIGN: Cohort study (diagnosis); Level of evidence, 3. METHODS: Knee MRI scans of patients with patellar instability were compared with those of age- and sex-matched controls. Measurements of the sulcus angle, lateral trochlear inclination (LTI), and trochlear depth were performed on axial images using bony and cartilaginous landmarks. Receiver operating characteristic curve analysis was performed, with the area under the curve (AUC) describing the accuracy of each diagnostic test. Optimal cutoff values were calculated to distinguish between knees with and without patellar instability. AUC and cutoff values were reported for each measurement as well as for male and female subgroups. RESULTS: A total of 238 knee MRI scans were included in this study (138 female, 100 male; age range, 18-39 years). Trochlear depth measurements had the greatest diagnostic value, with AUCs of 0.79 and 0.82 on bone and cartilage, respectively. All measurements (sulcus angle, LTI, trochlear depth) on bone and cartilage had an AUC ≥0.7 (range, 0.70-0.86), with optimal cutoff values of 145° (bone) and 154° (cartilage) for the sulcus angle, 17° (bone) and 13° (cartilage) for LTI, and 4 mm (bone) and 3 mm (cartilage) for trochlear depth. Optimal cutoff values in female patients varied from those in male patients for all measurements except for cartilaginous trochlear depth. CONCLUSION: Normal thresholds for trochlear dysplasia varied based on the use of bony versus cartilaginous landmarks. Cartilaginous trochlear depth measurements had the greatest ability to identify knees with patellar instability. Furthermore, optimal cutoff values for all measurements except for cartilaginous trochlear depth differed between female and male patients. These findings suggest that sex-specific parameters of normal values may be needed in the assessment of risk factors for patellofemoral instability.

The Influence of Static Load and Sideways Impact Fall on Extramedullary Bone Plates Used to Treat Intertrochanteric Femoral Fracture: A Preclinical Strength Assessment.

Hip fracture accounts for a large number of hospitalizations, thereby causing substantial economic burden. Majority (> 90%) of all hip fractures are associated to sideways fall. Studies on sideways fall usually involve loading at quasi-static or at constant displacement rate, which neglects the physics of actual fall. Understanding femur resonance frequency and associated mode shapes excited by dynamic loads is also critical. Two commercial extramedullary implants, proximal femoral locking plate (PFLP) and variable angle dynamic hip screw (VA-DHS), were chosen to carry out the preclinical assessments on a simulated Evans-I type intertrochanteric fracture. In this study, we hypothesized that the behavior of the implant depends on the loading types-axial static and transverse impact-and a rigid implanted construct will absorb less impact energy for sideways fall. The in silico models were validated using experimental measurements of full-field strain data obtained from a 2D digital image correlation (DIC) study. Under peak axial load of 3 kN, PFLP construct predicted greater axial stiffness (1.07 kN/mm) as opposed to VA-DHS (0.85 kN/mm), although the former predicted slightly higher proximal stress shielding. Further, with greater mode 2 frequency, PFLP predicted improved performance in resisting bending due to sideways fall as compared to the other implant. Overall, the PFLP implanted femur predicted the least propensity to adverse stress intensities, suggesting better structural rigidity and higher capacity in protecting the fractured femur against fall.

Does bone penetration of cement differ by cement type and application time-point?

Intrusion of cement into bone is often considered an indirect indicator for implant stability in cemented joint replacement procedures. However, the influence of cement type (different viscosities/manufacturers) and application time-point on penetration of cements continues to be debated. This study aimed to quantify the effect of cement type and application time-point on the depth of penetration using porcine tibial specimens. Four different bone cements were applied to 60 resected porcine cadaveric tibias at three time-points within the working window (1, 2, and 3 min after dough time). Penetration was measured using computed tomography, utilizing two rigorous methods from the literature and a newly proposed volumetric method. Application time-point had a strong influence on the thickness of the cement layer above the resected tibia (0.25, 0.49, 0.73 mm at the three time-points). No significant variation in penetration depth metrics with cement type or application time-point was found, except percentage area covered by cement at 2 mm depth. This metric was significantly different between 1 and 3-minute time-points (12% and 6% respectively). Time-point of application had a significant influence on thickness of pure cement layer over resected bone. However, penetration depth was not significantly affected by cement type or application time-point. The clinical significance of these findings is that it may be better to apply cement relatively soon after dough time to avoid excessively thick cement mantle between implant and bone. Further, the choice of cement type may have minimal impact on cement penetration, indicating that long standing principles of good cement application maybe more important.

Don't Hold Your Breath-A Video-Based Study of Procedural Intervals During Pediatric Rapid Sequence Intubation.

OBJECTIVE: Develop a framework for data collection to determine the contributions of both laryngoscopy and tube delivery intervals to the apneic period in unsuccessful and successful attempts among patients undergoing rapid sequence intubation (RSI) in a pediatric emergency department (PED). DESIGN: This was a retrospective, observational study of RSI. SETTING: An academic PED. PATIENTS: A consecutive sample of all intubations attempts of first provider physicians performing RSI in the shock trauma suite over a 10-month period in 2018-2019. MEASUREMENT AND MAIN RESULTS: Data were collected by structured video review. The main outcome was the duration of the laryngoscopy and tube delivery intervals per attempt. We compared interval duration between successful and unsuccessful attempts, adjusting for age, accounting for repeated measures, and clustering by provider. There were 69 patients with 89 total intubation attempts. Sixty-three patients were successfully intubated by the first provider (91%). Pediatric emergency medicine fellows performed 54% of the attempts. The median duration of the apneic period per attempt was longer in unsuccessful attempts (57 vs 44 seconds; median of difference, -10.5; 95% confidence interval [CI], -17.0 to -4.0). The duration of laryngoscopy was similar (18 vs 13 seconds; median of difference, -3.5; 95% CI, -8.0 to 1.0), but tube delivery was longer in unsuccessful attempts (25.5 vs. 11 seconds; median of difference, -12.5; 95% CI, -17.0 to -4.0). These results did not change when adjusting for age or clustering by provider. CONCLUSIONS: We successfully developed a specific, time-based framework for the contributors to prolonged apnea in RSI. Prolonged tube delivery accounted for more of the apneic period. Future studies and improvement should focus on problems during tube delivery in the PED.

Synthesis and Characterization of Carbon Nanotube-Doped Thermoplastic Nanocomposites for the Additive Manufacturing of Self-Sensing Piezoresistive Materials.

We present carbon nanotube (CNT)-reinforced polypropylene random copolymer (PPR) nanocomposites for the additive manufacturing of self-sensing piezoresistive materials via fused filament fabrication. The PPR/CNT feedstock filaments were synthesized through high shear-induced melt blending with controlled CNT loading up to 8 wt % to enable three-dimensional (3D) printing of nanoengineered PPR/CNT composites. The CNTs were found to enhance crystallinity (up to 6%) in PPR-printed parts, contributing to the overall CNT-reinforcement effect that increases both stiffness and strength (increases of 56% in modulus and 40% in strength at 8 wt % CNT loading). Due to electrical conductivity (∼10-4-10-1 S/cm with CNT loading) imparted to the PPR by the CNT network, multifunctional in situ strain and damage sensing in 3D-printed CNT/PPR bulk composites and lattice structures are revealed. A useful range of gauge factors (k) is identified for strain sensing (ks = 10.1-17.4) and damage sensing (kd = 20-410) across the range of CNT loadings for the 0° print direction. Novel auxetic re-entrant and S-unit cell lattices are printed, with multifunctionality demonstrated as strain- and damage-sensing in tension. The PPR/CNT multifunctional nanocomposite lattices demonstrated here exhibit tunable strain and damage sensitivity and have application in biomedical engineering for the creation of self-sensing patient-specific devices such as orthopedic braces, where the ability to sense strain (and stress) can provide direct information for optimization of brace design/fit over the course of treatment.

Defining and predicting radiographic knee osteoarthritis progression: a systematic review of findings from the osteoarthritis initiative.

PURPOSE: The purposes of this systematic review were to (1) identify the commonly used definitions of radiographic KOA progression, (2) summarize the important associative risk factors for disease progression based on findings from the OAI study and (3) summarize findings from radiographic KOA progression prediction modeling studies regarding the characterization of progression and outcomes. METHODS: A systematic review was performed by conducting a literature search of definitions, risk factors and predictive models for radiographic KOA progression that utilized data from the OAI database. Radiographic progression was further characterized into "accelerated KOA" and "typical progression," as defined by included studies. RESULTS: Of 314 studies identified, 41 studies were included in the present review. Twenty-eight (28) studies analyzed risk factors associated with KOA progression, and 13 studies created or validated prediction models or risk calculators for progression. Kellgren-Lawrence (KL) grade based on radiographs was most commonly used to characterize KOA progression (50%), followed by joint space width (JSW) narrowing (32%) generally over 48 months. Risk factors with the highest odds ratios (OR) for progression included periarticular bone mineral density (OR 10.40), any knee injury within 1 year (OR 9.22) and baseline bone mineral lesions (OR 7.92). Nine prediction modeling studies utilized both clinical and structural risk factors to inform their models, and combined models outperformed purely clinical or structural models. CONCLUSION: The cumulative evidence suggests that combinations of structural and clinical risk factors may be able to predict radiographic KOA progression, particularly in patients with accelerated progression. Clinically relevant and feasible prediction models and risk calculators may provide valuable decision-making support when caring for patients at risk of KOA progression, although standardization in modeling and variable identification does not yet exist.

Sustained Improvement in the Performance of Rapid Sequence Intubation Five Years after a Quality Improvement Initiative.

Many quality improvement interventions do not lead to sustained improvement, and the sustainability of healthcare interventions remains understudied. We conducted a time-series analysis to determine whether improvements in the safety of rapid sequence intubation (RSI) in our academic pediatric emergency department were sustained 5 years after a quality improvement initiative. METHODS: There were 3 study periods: baseline (April 2009-March 2010), improvement (July 2012-December 2013), and operational (January 2014-December 2018). All patients undergoing RSI were eligible. We collected data using a structured video review. We compared key processes and outcomes with statistical process control charts. RESULTS: We collected data for 615 of 643 (96%) patient encounters with RSI performed: 114 baseline (12 months), 105 improvement (18 months), and 396 operational (60 months). Key characteristics were similar, including patient age. Statistical process control charts indicated sustained improvement of all 6 key processes and the primary outcome measure (oxyhemoglobin desaturation) throughout the 5-year operational period. CONCLUSIONS: Improvements in RSI safety were sustained 5 years after a successful improvement initiative, with further improvement seen in several key processes. Further research is needed to elucidate the factors contributing to sustainability.

Postoperative complications: an observational study of trends in the United States from 2012 to 2018.

BACKGROUND: Postoperative complications continue to constitute a major issue for both the healthcare system and the individual patient and are associated with inferior outcomes and higher healthcare costs. The objective of this study was to evaluate the trends of postoperative complication rates over a 7-year period. METHODS: The NSQIP datasets from 2012 to 2018 were used to assess 30-day complication incidence rates including mortality rate following surgical procedures within ten surgical subspecialties. Multivariable logistic regression was used to associate complication rates with dataset year, while adjusting for relevant confounders. RESULTS: A total of 5,880,829 patients undergoing major surgery were included. Particularly the incidence rates of four complications were found to be decreasing: superficial SSI (1.9 to 1.3%), deep SSI (0.6 to 0.4%), urinary tract infection (1.6 to 1.2%) and patient unplanned return to the operating room (3.1 to 2.7%). Incidence rate for organ/space SSI exhibited an increase (1.1 to 1.5%). When adjusted, regression analyses indicated decreased odds ratios (OR) through the study period years for particularly deep SSI OR 0.92 [0.92-0.93], superficial SSI OR 0.94 [0.94-0.94] and acute renal failure OR 0.96 [0.95-0.96] as the predictor variable (study year) increased (p < 0.01). However, OR's for organ/space SSI 1.05 [1.05-1.06], myocardial infarction 1.01 [1.01-1.02] and sepsis 1.01 [1.01-1.02] increased slightly over time (all p < 0.01). CONCLUSIONS: Incidence rates for the complications exhibited a stable trend over the study period, with minor in or decreases observed.

Assessing the utility of deep neural networks in predicting postoperative surgical complications: a retrospective study.

BACKGROUND: Early detection of postoperative complications, including organ failure, is pivotal in the initiation of targeted treatment strategies aimed at attenuating organ damage. In an era of increasing health-care costs and limited financial resources, identifying surgical patients at a high risk of postoperative complications and providing personalised precision medicine-based treatment strategies provides an obvious pathway for reducing patient morbidity and mortality. We aimed to leverage deep learning to create, through training on structured electronic health-care data, a multilabel deep neural network to predict surgical postoperative complications that would outperform available models in surgical risk prediction. METHODS: In this retrospective study, we used data on 58 input features, including demographics, laboratory values, and 30-day postoperative complications, from the American College of Surgeons (ACS) National Surgical Quality Improvement Program database, which collects data from 722 hospitals from around 15 countries. We queried the entire adult (≥18 years) database for patients who had surgery between Jan 1, 2012, and Dec 31, 2018. We then identified all patients who were treated at a large midwestern US academic medical centre, excluded them from the base dataset, and reserved this independent group for final model testing. We then randomly created a training set and a validation set from the remaining cases. We developed three deep neural network models with increasing numbers of input variables and so increasing levels of complexity. Output variables comprised mortality and 18 different postoperative complications. Overall morbidity was defined as any of 16 postoperative complications. Model performance was evaluated on the test set using the area under the receiver operating characteristic curve (AUC) and compared with previous metrics from the ACS-Surgical Risk Calculator (ACS-SRC). We evaluated resistance to changes in the underlying patient population on a subset of the test set, comprising only patients who had emergency surgery. Results were also compared with the Predictive OpTimal Trees in Emergency Surgery Risk (POTTER) calculator. FINDINGS: 5 881 881 surgical patients, with 2941 unique Current Procedural Terminology codes, were included in this study, with 4 694 488 in the training set, 1 173 622 in the validation set, and 13 771 in the test set. The mean AUCs for the validation set were 0·864 (SD 0·053) for model 1, 0·871 (0·055) for model 2, and 0·882 (0·053) for model 3. The mean AUCs for the test set were 0·859 (SD 0·063) for model 1, 0·863 (0·064) for model 2, and 0·874 (0·061) for model 3. The mean AUCs of each model outperformed previously published performance metrics from the ACS-SRC, with a direct correlation between increasing model complexity and performance. Additionally, when tested on a subgroup of patients who had emergency surgery, our models outperformed previously published POTTER metrics. INTERPRETATION: We have developed unified prediction models, based on deep neural networks, for predicting surgical postoperative complications. The models were generally superior to previously published surgical risk prediction tools and appeared robust to changes in the underlying patient population. Deep learning could offer superior approaches to surgical risk prediction in clinical practice. FUNDING: The Novo Nordisk Foundation.

Patient-specific musculoskeletal models as a framework for comparing ACL function in unicompartmental versus bicruciate retaining arthroplasty.

Unicompartmental knee arthroplasty has been shown to provide superior functional outcomes compared to total knee arthroplasty and have motivated development of advanced implant designs including bicruciate retaining knee arthroplasty. However, few validated frameworks are available to directly compare the effect of implant design and surgical techniques on ligament function and joint kinematics. In the present study, the subject-specific lower extremity models were developed based on musculoskeletal modeling framework using force-dependent kinematics method, and validated against in vivo telemetric data. The experiment data of two subjects who underwent TKA were obtained from the SimTK "Grand Challenge Competition" repository, and integrated into the subject-specific lower extremity model. Five walking gait trials and three different knee implant models for each subject were used as partial inputs for the model to predict knee biomechanics for unicompartmental, bicruciate retaining, and total knee arthroplasty. The results showed no significant differences in the tibiofemoral contact forces or angular kinematic parameters between three groups. However, unicompartmental knee arthroplasty demonstrated significantly more posterior tibial location between 0% and 40% of the gait cycle (p < 0.017). Significant differences in range of tibiofemoral anterior/posterior translation and medial/lateral translation were also observed between unicompartmental and bicruciate retaining arthroplasty (p < 0.017). Peak values of anterior cruciate ligament forces differed between unicompartmental and bicruciate retaining arthroplasty from 10% to 30% of the gait cycle. Findings of this study indicate that unicompartmental and bicruciate retaining arthroplasty do not have identical biomechanics and point to the complementary role of anterior cruciate ligament and articular geometry in guiding knee function. Further, the patient-specific musculoskeletal model developed provides a reliable framework for assessing new implant designs, and effect of surgical techniques on knee biomechanics following arthroplasty.

Comparing the performance of a deep convolutional neural network with orthopedic surgeons on the identification of total hip prosthesis design from plain radiographs.

PURPOSE: A crucial step in the preoperative planning for a revision total hip replacement (THR) surgery is the accurate identification of the failed implant design, especially if one or more well-fixed/functioning components are to be retained. Manual identification of the implant design from preoperative radiographic images can be time-consuming and inaccurate, which can ultimately lead to increased operating room time, more complex surgery, and increased healthcare costs. METHOD: In this study, we present a novel approach to identifying THR femoral implants' design from plain radiographs using a convolutional neural network (CNN). We evaluated a total of 402 radiographs of nine different THR implant designs including, Accolade II (130 radiographs), Corail (89 radiographs), M/L Taper (31 radiographs), Summit (31 radiographs), Anthology (26 radiographs), Versys (26 radiographs), S-ROM (24 radiographs), Taperloc Standard Offset (24 radiographs), and Taperloc High Offset (21 radiographs). We implemented a transfer learning approach and adopted a DenseNet-201 CNN architecture by replacing the final classifier with nine fully connected neurons. Furthermore, we used saliency maps to explain the CNN decision-making process by visualizing the most important pixels in a given radiograph on the CNN's outcome. We also compared the CNN's performance with three board-certified and fellowship-trained orthopedic surgeons. RESULTS: The CNN achieved the same or higher performance than at least one of the surgeons in identifying eight of nine THR implant designs and underperformed all of the surgeons in identifying one THR implant design (Anthology). Overall, the CNN achieved a lower Cohen's kappa (0.78) than surgeon 1 (1.00), the same Cohen's kappa as surgeon 2 (0.78), and a slightly higher Cohen's kappa than surgeon 3 (0.76) in identifying all the nine THR implant designs. Furthermore, the saliency maps showed that the CNN generally focused on each implant's unique design features to make a decision. Regarding the time spent performing the implant identification, the CNN accomplished this task in ~0.06 s per radiograph. The surgeon's identification time varied based on the method they utilized. When using their personal experience to identify the THR implant design, they spent negligible time. However, the identification time increased to an average of 8.4 min (standard deviation 6.1 min) per radiograph when they used another identification method (online search, consulting with the orthopedic company representative, and using image atlas), which occurred in about 17% of cases in the test subset (40 radiographs). CONCLUSIONS: CNNs such as the one developed in this study can be used to automatically identify the design of a failed THR femoral implant preoperatively in just a fraction of a second, saving time and in some cases improving identification accuracy.

Natural language processing with deep learning for medical adverse event detection from free-text medical narratives: A case study of detecting total hip replacement dislocation.

BACKGROUND: Accurate and timely detection of medical adverse events (AEs) from free-text medical narratives can be challenging. Natural language processing (NLP) with deep learning has already shown great potential for analyzing free-text data, but its application for medical AE detection has been limited. METHOD: In this study, we developed deep learning based NLP (DL-NLP) models for efficient and accurate hip dislocation AE detection following primary total hip replacement from standard (radiology notes) and non-standard (follow-up telephone notes) free-text medical narratives. We benchmarked these proposed models with traditional machine learning based NLP (ML-NLP) models, and also assessed the accuracy of International Classification of Diseases (ICD) and Current Procedural Terminology (CPT) codes in capturing these hip dislocation AEs in a multi-center orthopaedic registry. RESULTS: All DL-NLP models outperformed all of the ML-NLP models, with a convolutional neural network (CNN) model achieving the best overall performance (Kappa = 0.97 for radiology notes, and Kappa = 1.00 for follow-up telephone notes). On the other hand, the ICD/CPT codes of the patients who sustained a hip dislocation AE were only 75.24% accurate. CONCLUSIONS: We demonstrated that a DL-NLP model can be used in largescale orthopaedic registries for accurate and efficient detection of hip dislocation AEs. The NLP model in this study was developed with data from the most frequently used electronic medical record (EMR) system in the U.S., Epic. This NLP model could potentially be implemented in other Epic-based EMR systems to improve AE detection, and consequently, quality of care and patient outcomes.

Leveraging subject-specific musculoskeletal modeling to assess effect of anterior cruciate ligament retaining total knee arthroplasty during walking gait.

Bi-cruciate retaining total knee arthroplasty has several potential advantages including improved anteroposterior knee stability compared to contemporary posterior cruciate-retaining total knee arthroplasty. However, few studies have explored whether there is significant differences of knee biomechanics following bi-cruciate retaining total knee arthroplasty compared to posterior cruciate-retaining total knee arthroplasty. In the present study, subject-specific lower extremity musculoskeletal multi-body dynamics models for bi-cruciate retaining, bi-cruciate retaining without anterior cruciate ligament, and posterior cruciate-retaining total knee arthroplasty were developed based on the musculoskeletal modeling framework using force-dependent kinematics method and validated against in vivo telemetric data. The experiment data of two subjects who underwent total knee arthroplasty were obtained for the SimTK "Grand Challenge Competition" repository, and integrated into the musculoskeletal model. Five walking gait trials for each subject were used as partial inputs for the model to predict the knee biomechanics for bi-cruciate retaining, bi-cruciate retaining without anterior cruciate ligament, and posterior cruciate-retaining total knee arthroplasty. The results revealed significantly greater range of anterior/posterior tibiofemoral translation, and significantly more posterior tibial location during the early phase of gait and more anterior tibial location during the late phase of gait were found in bi-cruciate retaining total knee arthroplasty without anterior cruciate ligament when compared to the bi-cruciate retaining total knee arthroplasty. No significant differences in tibiofemoral contact forces, rotations, translations, and ligament forces between bi-cruciate retaining and posterior cruciate-retaining total knee arthroplasty during normal walking gait, albeit slight differences in range of tibiofemoral internal/external rotation and anterior/posterior translation were observed. The present study revealed that anterior cruciate ligament retention has a positive effect on restoring normal knee kinematics in bi-cruciate retaining total knee arthroplasty. Preservation of anterior cruciate ligament in total knee arthroplasty and knee implant designs interplay each other and both contribute to restoring normal knee kinematics in different types of total knee arthroplasty. Further evaluation of more demanding activities and subject data from patients with bi-cruciate retaining and posterior cruciate-retaining total knee arthroplasty via musculoskeletal modeling may better highlight the role of the anterior cruciate ligament and its stabilizing influence.

Detecting total hip replacement prosthesis design on plain radiographs using deep convolutional neural network.

Identifying the design of a failed implant is a key step in the preoperative planning of revision total joint arthroplasty. Manual identification of the implant design from radiographic images is time-consuming and prone to error. Failure to identify the implant design preoperatively can lead to increased operating room time, more complex surgery, increased blood loss, increased bone loss, increased recovery time, and overall increased healthcare costs. In this study, we present a novel, fully automatic and interpretable approach to identify the design of total hip replacement (THR) implants from plain radiographs using deep convolutional neural network (CNN). CNN achieved 100% accuracy in the identification of three commonly used THR implant designs. Such CNN can be used to automatically identify the design of a failed THR implant preoperatively in just a few seconds, saving time and improving the identification accuracy. This can potentially improve patient outcomes, free practitioners' time, and reduce healthcare costs.

Evaluation of pull-off strength and seating displacement of sleeved ceramic revision heads in modular hip arthroplasty.

Corrosion in revision total hip arthroplasty can be mitigated using a ceramic head on a well-fixed in situ stem, but concerns of their early failure because of any surface defects on in situ stem necessitates the use of a titanium sleeve, which furnishes a factory-finish surface. These sleeves are manufactured in different sizes allowing neck-length adjustment. The strength of the taper junction of non-sleeved primary heads is well-investigated, but the influence of an interposed titanium sleeve on achieving a secure taper lock is unclear. Therefore, this study aimed to investigate the pull-off strength and seating displacement of revision ceramic heads and titanium taper sleeves. Two different head diameters and two different taper adapter sleeve offset lengths were mated with trunnions at two different impaction forces. The seating displacement and pull-off force was recorded for each specimen. Profilometry of the grooved outer surfaces of the sleeve and trunnion was done before and after testing to analyze the change in surface roughness. The influence of head diameter, sleeve offset, and impaction force on seating displacement and pull-off force was analyzed using analysis of covariance. Pull-off forces for 6 kN assembly force were approximately three times those for 2 kN. The head diameter did not have a significant effect on the measured parameters. Compared with short offset length sleeves, extra-long increased seating displacement by 31% and reduced pull-off forces by 15%. While sleeves of different offset lengths permit control of neck length, surgeons must be careful of the impact of this choice on the stability of implant. © 2019 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 38:1523-1528, 2020.

Developing Resident-Sensitive Quality Measures: Engaging Stakeholders to Inform Next Steps.

OBJECTIVE: Despite the need for quality measures relevant to the work residents complete, few attempts have been made to address this gap. Resident-sensitive quality measures (RSQMs) can help fill this void. This study engaged resident and supervisor stakeholders to develop and inform next steps in creating such measures. METHODS: Two separate nominal group techniques (NGTs), one with residents and one with faculty and fellow supervisors, were used to generate RSQMs for 3 specific illnesses (asthma, bronchiolitis, and closed head injury) as well as general care for the pediatric emergency department. Two separate Delphi processes were then used to prioritize identified RSQMs. The measures produced by each group were compared side by side, illuminating similarities and differences that were explored through focus groups with residents and supervisors. These focus groups also probed future settings in which to develop RSQMs. RESULTS: In the NGT and Delphi groups, residents and supervisors placed considerable focus on measures in 3 areas across the illnesses of interest: 1) appropriate medication dosing, 2) documentation, and 3) information provided at patient discharge. Focus groups highlighted hospital medicine and general pediatrics as priority areas for developing future RSQMs but also noted contextual variables that influence the application of similar measures in different settings. Residents and supervisors had both similar as well as unique insights into developing RSQMs. CONCLUSIONS: This study continues to pave the path forward in developing future RSQMs by exploring specific settings, measures, and stakeholders to consider when undertaking this work.

Quantification of Adhesion Force of Bacteria on the Surface of Biomaterials: Techniques and Assays.

Biomaterials associated infection (BAI) has been identified as one of the leading causes of failure of bioimplants. A failed implant requires revision surgery, which is about 20 times costlier and more painful than primary surgery. Infection starts from initial attachment of bacteria onto the surface of biomaterials followed by colonization and biofilm formation. Once a biofilm is developed the bacteria become resistant toward antibiotics. On account of microbial cell development, their metabolic activity and viability are strongly affected by the adhesion. Hence a thorough investigation warrants an in-depth understanding of the interfacial adhesion. Several methods such as plate-and-wash assay, spinning-disc assay, centrifugation assay, step-pressure technique, optical tweezers, atomic force microscopy (AFM) and nanoindentation are used for the measurement of the bacterial adhesion. Most of the aforementioned techniques are nonquantitative and provide only approximate values of adhesion forces. Techniques such as AFM and nanoindentation can quantify a wide range of force 10 pN to 1 µN and 1 nN to 10 µN respectively, and hence they are particularly useful for exact quantification of the adhesion force as well as adhesion strength of bacterial cells on various surfaces of biomaterials. In this review, we present a comparative study of the techniques available to measure the bacterial adhesion force and strength, discuss the use of AFM in adhesion force quantification in detail and conclude by hypothesizing that the AFM technique has an edge over other techniques for quantification of bacterial adhesion force.