Artificial Intelligence Algorithm Can Predict Lymph Node Malignancy from EBUS-TBNA Images for NSCLC.

INTRODUCTION: EBUS-TBNA for lung cancer staging is operator-dependent, resulting in high rates of non-diagnostic lymph node (LN) samples. We hypothesized that an artificial intelligence (AI) algorithm can consistently and reliably predict nodal metastases from the ultrasound images of LNs, when compared to pathology. METHODS: In this analysis of prospectively-recorded B-mode images of mediastinal LNs during EBUS-TBNA, we used transfer learning to build an end-to-end ensemble of three deep neural networks (ResNet152V2, InceptionV3, and DenseNet201). Model hyperparameters were tuned, and the optimal version(s) of each model trained using 80% of the images. A learned ensemble (multi-layer perceptron) of the optimal versions were applied to the remaining 20% of the images (Test Set). All predictions were compared to final pathology from nodal biopsies and/or surgical specimen. RESULTS: A total of 2,569 LN images from 773 patients were used. The Training Set included 2,048 LNs, of which 70.02% were benign and 29.98% were malignant on pathology. The Testing Set included 521 LNs, of which 70.06% were benign and 29.94% were malignant on pathology. The final ensemble model had an overall accuracy of 80.63% [95%Confidence Interval (CI):76.93%-83.97%], 43.23% sensitivity (95%CI:35.30%-51.41%), 96.91% specificity (95%CI:94.54%-98.45%), 85.90% positive predictive value (95%CI:76.81%-91.80%), 79.68% negative predictive value (95%CI:77.34%-81.83%), and AUC of 0.701 (95%CI:0.646-0.755) for malignancy. CONCLUSION: There now exists an AI algorithm which can identify nodal metastases based only on ultrasound images with good overall accuracy, specificity, and positive predictive value. Further optimization with larger sample sizes would be beneficial prior to clinical application.

Investigating the relationship between quantitative-based ultrasound and MRI estimations of rotator cuff fatty infiltration.

INTRODUCTION: Fatty infiltration (FI) of the rotator cuff has important clinical implications. Quantitatively estimating FI using ultrasound (US) has considerable benefits for assessing FI in a non-invasive, accessible manner. This research investigated whether FI of the supraspinatus (SS) and infraspinatus (IS), estimated using US was related to intramuscular fat fractions measured from magnetic resonance images (MRI). METHODS: Data from 12 healthy young adult participants were used for analysis. US images of the SS and IS were captured using multiple transducer placement techniques from which echogenicity of the muscle region was quantified. Shoulder MRI were captured from which SS and IS were manually segmented and intramuscular fat fractions calculated. Six upper limb strength exertions were performed, resisted by a hand dynamometer. RESULTS: IS and SS echogenicity explained a significant amount of variance in MRI fat fractions for certain body positions and transducer techniques. Echogenicity agreement was higher for IS than SS. Significant relationships were identified between strength exertions and both echogenicity and MRI muscle volume, but not MRI fat fraction. CONCLUSIONS: This research provides preliminary evidence showing that quantitative-based US methods can be used to estimate MRI calculated fat fractions for the rotator cuff.

Imaging of joint response to exercise with MRI and PET.

Imaging of the joint in response to loading stress may provide additional measures of joint structure and function beyond conventional, static imaging studies. Exercise such as running, stair climbing, and squatting allows evaluation of the joint response to larger loading forces than during weight bearing. Quantitative MRI (qMRI) may assess properties of cartilage and meniscus hydration and organization in vivo that have been investigated to assess the functional response of these tissues to physiological stress. [18F]sodium fluoride ([18F]NaF) interrogates areas of newly mineralizing bone and provides an opportunity to study bone physiology, including perfusion and mineralization rate, as a measure of joint loading stress. In this review article, methods utilizing quantitative MRI, PET, and hybrid PET-MRI systems for assessment of the joint response to loading from exercise in vivo are examined. Both methodology and results of various studies performed are outlined and discussed. Lastly, the technical considerations, challenges, and future opportunities for these approaches are addressed.

Investigating acute changes in osteoarthritic cartilage by integrating biomechanics and statistical shape models of bone: data from the osteoarthritis initiative.

OBJECTIVE: This proof-of-principle study integrates joint reaction forces (JRFs) and bone shape to assess acute cartilage changes from walking and cycling. METHODS: Sixteen women with symptomatic knee osteoarthritis were recruited. Biomechanical assessment estimated JRFs during walking and cycling. Subsamples had magnetic resonance imaging (MRI) performed before and after a 25-min walking (n = 7) and/or cycling (n = 9) activity. MRI scans were obtained to assess cartilage shape and composition (T2 relaxation time). Bone shape was quantified using a statistical shape model built from 13 local participants and 100 MRI scans from the Osteoarthritis Initiative. Statistical parametric mapping quantified cartilage change and correlations between cartilage change with JRFs and statistical shape model features. RESULTS: Cartilage thickness (interior lateral, Δ - 0.10 mm) and T2 (medial, Δ - 4 ms) decreased on the tibial plateau. On the femur, T2 change depended on the activity. Greater tibiofemoral JRF was associated with more cartilage deformation on the lateral femoral trochlea after walking (r - 0.56). Knees more consistent with osteoarthritis showed smaller decreases in tibial cartilage thickness. DISCUSSION: Walking and cycling caused distinct patterns of cartilage deformation, which depended on knee JRFs and bone morphology. For the first time, these results show that cartilage deformation is dependent on bone shapes and JRFs in vivo.

Automatic knee cartilage and bone segmentation using multi-stage convolutional neural networks: data from the osteoarthritis initiative.

OBJECTIVES: Accurate and efficient knee cartilage and bone segmentation are necessary for basic science, clinical trial, and clinical applications. This work tested a multi-stage convolutional neural network framework for MRI image segmentation. MATERIALS AND METHODS: Stage 1 of the framework coarsely segments images outputting probabilities of each voxel belonging to the classes of interest: 4 cartilage tissues, 3 bones, 1 background. Stage 2 segments overlapping sub-volumes that include Stage 1 probability maps concatenated to raw image data. Using six fold cross-validation, this framework was tested on two datasets comprising 176 images [88 individuals in the Osteoarthritis Initiative (OAI)] and 60 images (15 healthy young men), respectively. RESULTS: On the OAI segmentation dataset, the framework produces cartilage segmentation accuracies (Dice similarity coefficient) of 0.907 (femoral), 0.876 (medial tibial), 0.913 (lateral tibial), and 0.840 (patellar). Healthy cartilage accuracies are excellent (femoral = 0.938, medial tibial = 0.911, lateral tibial = 0.930, patellar = 0.955). Average surface distances are less than in-plane resolution. Segmentations take 91 ± 11 s per knee. DISCUSSION: The framework learns to automatically segment knee cartilage tissues and bones from MR images acquired with two sequences, producing efficient, accurate quantifications at varying disease severities.

A new technique to evaluate the impact of running on knee cartilage deformation by region.

OBJECTIVES: When measuring changes in knee cartilage thickness in vivo after loading, mean values may not reflect local changes. The objectives of this investigation were: (1) use statistical parametric mapping (SPM) to determine regional deformation patterns of tibiofemoral cartilage in response to running; (2) quantify regional differences in cartilage thickness between males and females; and (3) explore the influence of sex on deformation. MATERIALS AND METHODS: Asymptomatic males (n = 15) and females (n = 15) had MRI imaging of their right knee before and after 15 min of treadmill running. Medial and lateral tibial, and medial and lateral weight-bearing femoral cartilage were segmented. SPM was completed on cartilage thickness maps to test the main effects of Running and Sex, and their interaction. F-statistic maps were thresholded; clusters above this threshold indicated significant differences. RESULTS: Deformation was observed in all four compartments; the lateral tibia had the largest area of deformation (p < 0.0001). Thickness differences between sexes were observed in all four compartments, showing females have thinner cartilage (p ≤ 0.009). The lateral tibia had small clusters indicating an interaction of sex on deformation (p ≤ 0.012). DISCUSSION: SPM identified detailed spatial information on tibiofemoral cartilage thickness differences observed after running, and between sexes and their interaction.

GT3X+ accelerometer placement affects the reliability of step-counts measured during running and pedal-revolution counts measured during bicycling.

UNLABELLED: Accelerometers provide a measure of step-count. Reliability and validity of step-count and pedal-revolution count measurements by the GT3X+ accelerometer, placed at different anatomical locations, is absent in the literature. The purpose of this study was to investigate the reliability and validity of step and pedal-revolution counts produced by the GT3X+ placed at different anatomical locations during running and bicycling. Twenty-two healthy adults (14 men and 8 women) completed running and bicycling activity bouts (5 minutes each) while wearing 6 accelerometers: 2 each at the waist, thigh and shank. Accelerometer and video data were collected during activity. Excellent reliability and validity were found for measurements taken from accelerometers mounted at the waist and shank during running (Reliability: intraclass correlation (ICC) ≥ 0.99; standard error of measurement (SEM) ≤1.0 steps; VALIDITY: Pearson ≥ 0.99) and at the thigh and shank during bicycling (Reliability: ICC ≥ 0.99; SEM ≤1.0 revolutions; VALIDITY: Pearson ≥ 0.99). Excellent reliability was found between measurements taken at the waist and shank during running (ICC ≥ 0.98; SEM ≤1.6 steps) and between measurements taken at the thigh and shank during bicycling (ICC ≥ 0.99; SEM ≤1.0 revolutions). These data suggest that the GT3X+ can be used for measuring step-count during running and pedal-revolution count during bicycling. Only shank placement is recommended for both activities.

Interactions of cumulative load with biomarkers of cartilage turnover predict knee cartilage change over 2 years: data from the osteoarthritis initiative.

The purpose was to investigate relationships of cumulative load and cartilage turnover biomarkers with 2-year changes in cartilage in knee osteoarthritis. From participants with Kellgren-Lawrence (KL) grades of 1 to 3, cartilage thickness and transverse relaxation time (T2) were computed from 24-month (baseline) and 48-month magnetic resonance images. Cumulative load was the interaction term of the Physical Activity Scale for the Elderly (PASE) and body mass index (BMI). Serum cartilage oligomeric matrix protein (COMP) and the nitrated form of type II collagen (Coll2-1 NO2) were collected at baseline. Multiple regressions (adjusted for baseline age, KL grade, cartilage measures, pain, comorbidity) evaluated the relationships of cumulative load and biomarkers with 2-year changes. In 406 participants (63.7 (8.7) years), interactions of biomarkers with cumulative load weakly predicted 2-year cartilage changes: (i) COMP × cumulative load explained medial tibia thickness change (R2 increased 0.062 to 0.087, p < 0.001); (ii) Coll2-1 NO2 × cumulative load explained central medial femoral T2 change (R2 increased 0.177 to 0.210, p < 0.001); and (iii) Coll2-1 NO2 × cumulative load explained lateral tibia T2 change (R2 increased 0.166 to 0.188, p < 0.001). Moderate COMP or Coll2-1 NO2 at baseline appeared protective. High COMP or Coll2-1 NO2, particularly with high BMI and low PASE, associated with worsening cartilage. Moderate serum concentrations of cartilage turnover biomarkers, at high and low physical activity, associated with maintained cartilage outcomes over 2 years. In conclusion, high concentrations of cartilage turnover biomarkers, particularly with high BMI and low physical activity, associated with knee cartilage thinning and increasing T2 over 2 years. Key Points • Higher quality cartilage may be better able to tolerate a larger cumulative load than poor quality cartilage. • Among participants enrolled in the Osteoarthritis Initiative Biomarkers Consortium Project, a representation of cumulative load exposure and its interaction with cartilage turnover biomarkers were weakly related with 2-year change in knee cartilage. • These findings suggest that cartilage turnover is a factor that modifies the relationship between loading exposure and cartilage loss in knee OA.

ShapeMed-Knee: A Dataset and Neural Shape Model Benchmark for Modeling 3D Femurs.

Analyzing anatomic shapes of tissues and organs is pivotal for accurate disease diagnostics and clinical decision-making. One prominent disease that depends on anatomic shape analysis is osteoarthritis, which affects 30 million Americans. To advance osteoarthritis diagnostics and prognostics, we introduce ShapeMed-Knee, a 3D shape dataset with 9,376 high-resolution, medical-imaging-based 3D shapes of both femur bone and cartilage. Besides data, ShapeMed-Knee includes two benchmarks for assessing reconstruction accuracy and five clinical prediction tasks that assess the utility of learned shape representations. Leveraging ShapeMed-Knee, we develop and evaluate a novel hybrid explicit-implicit neural shape model which achieves up to 40% better reconstruction accuracy than a statistical shape model and implicit neural shape model. Our hybrid models achieve state-of-the-art performance for preserving cartilage biomarkers; they're also the first models to successfully predict localized structural features of osteoarthritis, outperforming shape models and convolutional neural networks applied to raw magnetic resonance images and segmentations. The ShapeMed-Knee dataset provides medical evaluations to reconstruct multiple anatomic surfaces and embed meaningful disease-specific information. ShapeMed-Knee reduces barriers to applying 3D modeling in medicine, and our benchmarks highlight that advancements in 3D modeling can enhance the diagnosis and risk stratification for complex diseases. The dataset, code, and benchmarks will be made freely accessible.

An Automated and Robust Tool for Musculoskeletal and Finite Element Modeling of the Knee Joint.

: To develop and assess an automatic and robust knee musculoskeletal finite element (MSK-FE) modeling pipeline. METHODS: Magnetic resonance images (MRIs) were used to train nnU-Net networks for auto-segmentation of knee bones (femur, tibia, patella, and fibula), cartilages (femur, tibia, and patella), menisci, and major knee ligaments. Two different MRI sequences were used to broaden applicability. Next, we created MSK-FE models of an unseen dataset using two MSK-FE modeling pipelines: template-based and auto-meshing. MSK models had personalized knee geometries with multi-degree-of-freedom elastic foundation contacts. FE models used fibril-reinforced poroviscoelastic swelling material models for cartilages and menisci. RESULTS: Volumes of knee bones, cartilages, and menisci did not significantly differ (p>0.05) across MRI sequences. MSK models estimated secondary knee kinematics during passive knee flexion tests consistent with in vivo and simulation-based values from the literature. Between the template-based and auto-meshing FE models, estimated cartilage mechanics often differed significantly (p<0.05), though differences were <15% (considering peaks during walking), i.e., <1.5 MPa for maximum principal stress, <1 percentage point for collagen fibril strain, and <3 percentage points for maximum shear strain. CONCLUSION: The template-based modeling provided a more rapid and robust tool than the auto-meshing approach, while the estimated knee biomechanics were comparable. Nonetheless, the auto-meshing approach might provide more accurate estimates in subjects with distinct knee irregularities, e.g., cartilage lesions. SIGNIFICANCE: The MSK-FE modeling tool provides a rapid, easy-to-use, and robust approach for investigating task- and person-specific mechanical responses of the knee cartilage and menisci, holding significant promise, e.g., in personalized rehabilitation planning.

Bicycle Set-Up Dimensions and Cycling Kinematics: A Consensus Statement Using Delphi Methodology.

Bicycle set-up dimensions and cycling kinematic data are important components of bicycle fitting and cyclist testing protocols. However, there are no guidelines on how bicycles should be measured and how kinematic data should be collected to increase the reliability of outcomes. This article proposes a consensus regarding bicycle set-up dimensions and recommendations for collecting cycling-related kinematic data. Four core members recruited panellists, prepared the document to review in each round for panellists, analysed the scores and comments of the expert panellists, reported the decisions and communicated with panellists. Fourteen experts with experience in research involving cycling kinematics and/or bicycle fitting agreed to participate as panellists. An initial list of 17 statements was proposed, rated using a five-point Likert scale and commented on by panellists in three rounds of anonymous surveys following a Delphi procedure. The consensus was agreed upon when more than 80% of the panellists scored the statement with values of 4 and 5 (moderately and strongly agree) with an interquartile range of less than or equal to 1. A consensus was achieved for eight statements addressing bicycle set-up dimensions (e.g. saddle height, saddle setback, etc.) and nine statements for cycling kinematic assessment (e.g. kinematic method, two-dimensional methodology, etc.). This consensus statement provides a list of recommendations about how bicycle set-up dimensions should be measured and the best practices for collecting cycling kinematic data. These recommendations should improve the transparency, reproducibility, standardisation and interpretation of bicycle measurements and cycling kinematic data for researchers, bicycle fitters and cycling related practitioners.

Investigating the relationship between physical activity and self-reported outcomes in adults with rotator cuff related shoulder pain.

PURPOSE: Rotator cuff related shoulder pain is common among older adults. While exercise is often recommended for this condition, the relationship between physical activity levels and self-reported outcomes is unclear. This study investigated whether self-reported outcomes (pain, function, depressive symptoms, self-efficacy, general mental and physical health) relate to physical activity, and whether those who participated in shoulder-specific exercise had better self-reported outcomes than those who performed whole-body exercise. METHODS: Forty-six participants with rotator cuff related shoulder pain participated from which 35 had complete datasets. Questionnaires were used to assess physical activity, pain, physical function, general health, and self-efficacy. Physical activity levels were also measured using an accelerometer. RESULTS: Neither pain nor other self-reported outcomes were related to subjective or objective physical activity levels. Participants that completed shoulder-specific exercise had significantly higher exercise self-efficacy than those who completed nonspecific exercise (P = .01; d = 0.91). CONCLUSION: A significant relationship between pain or self-reported outcomes and physical activity was not identified. Those who self-reported regularly exercising their injured shoulder had higher exercise self-efficacy than those who did not. These findings have clinical implications, suggesting that strategies to boost exercise self-efficacy may be important for older adults with rotator cuff related shoulder pain.

Muscle-driven simulations and experimental data of cycling.

Muscle-driven simulations have provided valuable insights in studies of walking and running, but a set of freely available simulations and corresponding experimental data for cycling do not exist. The aim of this work was to develop a set of muscle-driven simulations of cycling and to validate them by comparison with experimental data. We used direct collocation to generate simulations of 16 participants cycling over a range of powers (40-216 W) and cadences (75-99 RPM) using two optimization objectives: a baseline objective that minimized muscle effort and a second objective that additionally minimized tibiofemoral joint forces. We tested the accuracy of the simulations by comparing the timing of active muscle forces in our baseline simulation to timing in experimental electromyography data. Adding a term in the objective function to minimize tibiofemoral forces preserved cycling power and kinematics, improved similarity between active muscle force timing and experimental electromyography, and decreased tibiofemoral joint reaction forces, which better matched previously reported in vivo measurements. The musculoskeletal models, muscle-driven simulations, simulation software, and experimental data are freely shared at https://simtk.org/projects/cycling_sim for others to reproduce these results and build upon this research.

Muscle Architecture and Subcutaneous Fat Measurements of Rectus Femoris and Vastus Lateralis at Optimal Length Aided by a Novel Ultrasound Transducer Attachment.

Purpose: This cross-sectional study determines the sensitivity of muscle architecture and fat measurements of the rectus femoris (RF) and vastus lateralis (VL) muscles from ultrasound images acquired with varying transducer tilt, using a novel transducer attachment, in healthy adults. Secondary objectives were to estimate intrarater and interrater reliability of image measurement and acquisition, respectively. Methods: Thirty healthy adults participated (15 women and 15 men; 25 [SD 2.5] y). Ultrasound image acquisition was conducted by two raters at different transducer tilts relative to the skin: estimated perpendicular, and five measured angles (80°, 85°, 90°, 95°, 100°) using the transducer attachment. Muscle thickness (MT), subcutaneous fat thickness (FT), pennation angle (PA), and fascicle length (FL) were measured. Sensitivity and reliability were assessed using intra-class correlation coefficients (ICCs) and standard error of measurements (SEMs). Results: MT and FT for RF and VL were not sensitive to transducer tilt. However, PA and FL were sensitive to transducer tilt. MT and FT for both muscles showed high ICCs and low SEMs for intrarater and interrater reliability. For PA of both muscles, standardizing transducer tilt improved interrater ICCs and lowered SEMs. Conclusion: MT and FT measurements of RF and VL acquired at 60° knee flexion are robust to varying transducer tilt angles. PA measurements benefit from standardizing transducer tilt.

Objectif : étude transversale pour déterminer la sensibilité de l'architecture musculaire et des mesures lipidiques du muscle droit antérieur de la cuisse (MDAC) et du muscle vaste externe (MVE) à partir des images échographiques acquises chez des adultes en santé par diverses inclinaisons du transducteur, au moyen d'un nouveau dispositif. Les objectifs secondaires consistaient à évaluer la fiabilité intraévaluateurs et interévaluateurs des mesures et de l'acquisition des images, respectivement. Méthodologie: au total, 30 adultes en santé ont participé (15 femmes et 15 hommes de 25 [ÉT 2,5 ans]). Deux évaluateurs ont acquis des images échographiques à des inclinaisons différentes du transducteur par rapport à la peau : mesure perpendiculaire estimative et mesure à cinq angles (80°, 85°, 90°, 95°, 100°) au moyen du dispositif du transducteur. Ils ont mesuré l'épaisseur des muscles (ÉM), l'épaisseur de la graisse sous-cutanée (ÉG), l'angle de pennation (AP) et la longueur des fascicules (LF). Ils ont aussi évalué la sensibilité et la fiabilité au moyen de coefficients de corrélation intraclasse (CCI) et de l'écart-type des mesures (ÉTM). Résultats: l'ÉM et l'ÉG du MDAC et du MVE n'étaient pas sensibles à l'inclinaison du transducteur, mais l'AP et la LF l'étaient. La fiabilité intraévaluateur et interévaluateur de l'ÉM et de l'ÉG des deux muscles présentait un CCI élevé et un ÉTM faible. Pour ce qui est de l'AP des deux muscles, la standardisation de l'inclinaison du transducteur améliorait la CCI et réduisait l'ÉTM interévaluateurs. Conclusion: les mesures de l'ÉM et de l'ÉG du MDAC et du MVE acquises à une flexion du genou de 60° sont probantes à des angles d'inclinaison variables du transducteur. Les mesures de l'AP tirent profit d'une inclinaison du transducteur standardisée.

Investigating the reliability and validity of subacromial space measurements using ultrasound and MRI.

BACKGROUND: While ultrasound (US) measures of the subacromial space (SAS) have demonstrated excellent reliability, measurements are typically captured by experts with extensive ultrasound experience. Further, the agreement between US measured SAS width and other imaging modalities has not been explored. This research evaluated the agreement between SAS measures captured by novice and expert raters and between US and magnetic resonance imaging (MRI). This study also evaluated the effect of US transducer tilt on measured SAS. METHODS: Nine men and nine women participated in this study. US images were captured by a novice and expert with the participant in both seated and supine positions. An inclinometer was fixed to the US probe to measure transducer tilt. SAS width was measured in real time from freeze framed images. MRI images were captured, and the humerus and acromion manually segmented. The SAS width was measured using a custom algorithm. RESULTS: Intraclass correlation coefficients (ICCs) between novice and expert raters were 0.74 and 0.63 for seated and supine positions, respectively. Intra-rater agreement was high for both novice (ICC = 0.83-0.84) and expert (ICC ≥ 0.94) raters. Agreement between US and MRI was poor (ICC = 0.21-0.49) but linearly related. CONCLUSIONS: Moderate agreement between novice and expert raters was demonstrated, while the agreement between US and MRI was poor. High intra-rater reliability within each rater suggests that US measures of the SAS may be completed by a novice with introductory training.

A scoping review of portable sensing for out-of-lab anterior cruciate ligament injury prevention and rehabilitation.

Anterior cruciate ligament (ACL) injury and ACL reconstruction (ACLR) surgery are common. Laboratory-based biomechanical assessment can evaluate ACL injury risk and rehabilitation progress after ACLR; however, lab-based measurements are expensive and inaccessible to most people. Portable sensors such as wearables and cameras can be deployed during sporting activities, in clinics, and in patient homes. Although many portable sensing approaches have demonstrated promising results during various assessments related to ACL injury, they have not yet been widely adopted as tools for out-of-lab assessment. The purpose of this review is to summarize research on out-of-lab portable sensing applied to ACL and ACLR and offer our perspectives on new opportunities for future research and development. We identified 49 original research articles on out-of-lab ACL-related assessment; the most common sensing modalities were inertial measurement units, depth cameras, and RGB cameras. The studies combined portable sensors with direct feature extraction, physics-based modeling, or machine learning to estimate a range of biomechanical parameters (e.g., knee kinematics and kinetics) during jump-landing tasks, cutting, squats, and gait. Many of the reviewed studies depict proof-of-concept methods for potential future clinical applications including ACL injury risk screening, injury prevention training, and rehabilitation assessment. By synthesizing these results, we describe important opportunities that exist for clinical validation of existing approaches, using sophisticated modeling techniques, standardization of data collection, and creation of large benchmark datasets. If successful, these advances will enable widespread use of portable-sensing approaches to identify ACL injury risk factors, mitigate high-risk movements prior to injury, and optimize rehabilitation paradigms.

Equations to Prescribe Bicycle Saddle Height based on Desired Joint Kinematics and Bicycle Geometry.

Overuse knee injuries are common in bicycling and are often attributed to poor bicycle-fit. Bicycle-fit for knee health focuses on setting saddle height to elicit a minimum knee flexion angle of 25-40°. Equations to predict saddle height include a single input, resulting in a likely suboptimal bicycle-fit. The purpose of this work was to develop an equation to predict saddle height from anthropometrics, bicycle geometry, and user-defined joint kinematics.Methods: Forty healthy adults (17 women, 23 men; mean (SD): 28.6 (7.2) years; 24.2 (2.6) kg/m2) participated. Kinematic analyses were conducted for 18 three-minute bicycling bouts including all combinations of 3 horizontal and 3 vertical saddle positions, and 2 crank arm lengths. For both minimum and maximum knee flexion, predictors were identified using Least Absolute Shrinkage and Selection Operator (LASSO) regression, and final models were fit using linear regression. Secondary analyses determined if saddle height equations were sex dependent.Results: The equation to predict saddle position from minimum knee flexion angle (R2=0.97; root mean squared error (RMSE) = 1.15 cm) was: Saddle height (cm) = 7.41 + 0.82(inseam cm) - 0.1(minimum knee flexion °) + 0.003(inseam cm)(seat tube angle °). The maximum knee flexion equation (R2=0.97; RMSE=1.15 cm) was: Saddle height (cm) = 41.63 + 0.78(inseam cm) - 0.25(maximum knee flexion °) + 0.002(inseam cm)(seat tube angle °). The saddle height equations were not dependent on sex.Conclusions: These equations provide a novel, practical strategy for bicycle-fit that accounts for rider anthropometrics, bicycle geometry and user-defined kinematics.Highlights This work developed simple equations to prescribed bicycle saddle height that elicits desired knee kinematics.Separate equations are presented for prescribing minimum or maximum knee flexion angle.Equations can be generalized to riders of both sexes, and a breadth of anthropometrics and ages.

An Artificial Intelligence Algorithm to Predict Nodal Metastasis in Lung Cancer.

BACKGROUND: Endobronchial ultrasound (EBUS) has features that allow a high accuracy for predicting lymph node (LN) malignancy. However their clinical application remains limited because of high operator dependency. We hypothesized that an artificial intelligence algorithm (NeuralSeg; NeuralSeg Ltd, Hamilton, Ontario, Canada) is capable of accurately identifying and predicting LN malignancy based on EBUS images. METHODS: In the derivation phase EBUS images were segmented twice by an endosonographer and used as controls in 5-fold cross-validation training of NeuralSeg. In the validation phase the algorithm was tested on new images it had not seen before. Logistic regression and receiver operator characteristic curves were used to determine NeuralSeg's capability of discrimination between benign and malignant LNs, using pathologic specimens as the gold standard. RESULTS: Two hundred ninety-eight LNs from 140 patients were used for derivation and 108 LNs from 47 patients for validation. In the derivation cohort NeuralSeg was able to predict malignant LNs with an accuracy of 73.8% (95% confidence interval [CI], 68.4%-78.7%). In the validation cohort NeuralSeg had an accuracy of 72.9% (95% CI, 63.5%-81.0%), specificity of 90.8% (95% CI, 81.9%-96.2%), and negative predictive value of 75.9% (95% CI, 71.5%-79.9%). NeuralSeg showed higher diagnostic discrimination during validation compared with derivation (c-statistic = 0.75 [95% CI, 0.65-0.85] vs 0.63 [95% CI, 0.54-0.72], respectively). CONCLUSIONS: NeuralSeg is able to accurately rule out nodal metastasis and can possibly be used as an adjunct to EBUS when nodal biopsy is not possible or inconclusive. Future work to evaluate the algorithm in a clinical trial is required.

Hip and ankle kinematics are the most important predictors of knee joint loading during bicycling.

OBJECTIVES: To assess the effect of ankle, knee, and hip kinematics on patellofemoral and tibiofemoral joint reaction forces (JRF) during bicycling. Secondarily, to assess if sex, horizontal saddle position, or crank arm length were related to JRFs, after accounting for kinematics. DESIGN: Experimental cross-sectional study. METHODS: Forty healthy adults (mean (SD); 28.6 (7.2) years, 24.2 (2.6)kg/m2, 17 women) bicycled under 18 bicycling positions. One position used commercial guidelines and 17 randomly deviated from commercial. Resultant patellofemoral as well as compressive and shear tibiofemoral JRFs were calculated. Linear mixed-effects models with a random intercept of leg-nested-in-participant were used. RESULTS: Patellofemoral resultant forces were most sensitive to all joint kinematics (i.e., sensitivity was defined as the slope of single predictor models); all JRFs were least sensitive to minimum knee flexion. Tibiofemoral compression was predicted by minimum hip flexion and maximum ankle dorsiflexion (R2=0.90). Tibiofemoral shear (R2=0.86) and the resultant patellofemoral JRF (R2=0.90) were predicted by minimum hip flexion, maximum ankle dorsiflexion, minimum knee flexion, and the interaction between minimum hip flexion and minimum knee flexion. Adding sex as a factor improved fit of all models. This sex-effect was driven by differences in cycling intensity, reflected by the tangential crank arm force. Horizontal saddle position and crank arm length were not related to JRFs. CONCLUSIONS: Optimizing joint kinematics should be the primary goal of bicycle-fit. JRFs were least sensitive to the current gold standard for assessing bicycle-fit, minimum knee flexion. Bicycle-fit is of particular importance for those working at high intensities.

Daily cumulative load and body mass index alter knee cartilage response to running in women.

BACKGROUND: It is unknown whether a greater accumulation of knee load over a typical day is related to how cartilage responds to an acute bout of loading. This information may clarify the role of habitual activity on cartilage function. RESEARCH QUESTION: Is there a relationship between change in tibial and femoral cartilage thickness, volume, and T2 relaxation time following running with daily cumulative knee load in women? Secondarily, is there a relationship between cartilage change following running and the statistical interaction of body mass index (BMI) and daily steps? METHODS: Participants (n = 15) completed gait analyses and wore an accelerometer over a week. Daily cumulative knee load was the statistical interaction between tibial compressive joint reaction force (JRF) impulse with the average number of daily steps measured using accelerometry. Magnetic resonance imaging scans were acquired before and immediately after 15-min of treadmill running. Changes in tibial and femoral cartilage thickness, volume, and T2 relaxation time were calculated. Multiple linear regressions tested the associations of cartilage change outcomes with: baseline (thickness, volume, T2), JRF impulse, steps, and the interaction JRF impulse*steps. Secondarily, BMI was substituted for JRF impulse. RESULTS AND SIGNIFICANCE: Tibial volume change was explained by baseline volume, JRF impulse, steps, and JRF impulse*steps (R2 = 0.50, p = 0.013). Additionally, tibial volume change was explained by baseline volume, BMI, steps, and BMI*steps (R2 = 0.43, p = 0.002). Those who were more physically active with lower JRF impulse (or lower BMI) showed less change in tibial cartilage after a running exposure. This may suggest cartilage conditioning.