Subject-specific regression equations to estimate lower spinal loads during symmetric and asymmetric static lifting

Determining whether biomechanical variables that describe common 'safe lifting' cues are associated with low back loads

Lift technique training programs have been implemented to help reduce injury risk, but the underlying content validity of cues used within these programs is not clear. The objective of this study was to determine whether biomechanical variables, that commonly used lifting cues aim to elicit, are associated with resultant low back extensor moment exposures. A sample of 72 participants were recruited to perform 10 repetitions of a floor-to-waist height barbell lift while whole-body kinematics and ground reaction forces were collected. Kinematic, kinetic, and energetic variables representative of characteristics commonly targeted by lifting cues were calculated as predictor variables, while peak and cumulative low back moments were calculated as dependent measures. Multiple regression revealed that 56.6-59.2% of variance in low back moments was explained by predictor variables. From these regression models, generating motion with the legs (both greater hip and knee work), minimizing the horizontal distance of the body to the load, maintaining a stable body position, and minimizing lift time were associated with lower magnitudes of low back moments. These data support that using cues targeting these identified variables may be more effective at reducing peak low back moment exposures via lift training.

Machine learning applications in spine biomechanics

Spine biomechanics is at a transformation with the advent and integration of machine learning and computer vision technologies. These novel techniques facilitate the estimation of 3D body shapes, anthropometrics, and kinematics from as simple as a single-camera image, making them more accessible and practical for a diverse range of applications. This study introduces a framework that merges these methodologies with traditional musculoskeletal modeling, enabling comprehensive analysis of spinal biomechanics during complex activities from a single camera. Additionally, we aim to evaluate their performance and limitations in spine biomechanics applications. The real-world applications explored in this study include assessment in workplace lifting, evaluation of whiplash injuries in car accidents, and biomechanical analysis in professional sports. Our results demonstrate potential and limitations of various algorithms in estimating body shape, kinematics, and conducting in-field biomechanical analyses. In industrial settings, the potential to utilize these new technologies for biomechanical risk assessments offers a pathway for preventive measures against back injuries. In sports activities, the proposed framework provides new opportunities for performance optimization, injury prevention, and rehabilitation. The application in forensic domain further underscores the wide-reaching implications of this technology. While certain limitations were identified, particularly in accuracy of predictions, complex interactions, and external load estimation, this study demonstrates their potential for advancement in spine biomechanics, heralding an optimistic future in both research and practical applications.

Multi-task artificial neural networks and their extrapolation capabilities to predict full-body 3D human posture during one- and two-handed load-handling activities

Machine-learning based human posture-prediction tools can potentially be robust alternatives to motion capture measurements. Existing posture-prediction approaches are confined to two-handed load-handling activities performed at heights below 120 cm from the floor and to predicting a limited number of body-joint coordinates/angles. Moreover, the extrapolating power of these tools beyond the range of the input dataset they were trained for (e.g., for underweight, overweight, or left-handed individuals) has not been investigated. In this study, we trained/validated/tested two posture-prediction (for full-body joint coordinates and angles) artificial neural networks (ANNs) using both 70%/15%/15% random-hold-out and leave-one-subject-out methods, based on a comprehensive kinematic dataset of forty-one full-body skin markers collected from twenty right-handed normal-weight (BMI = 18-26 kg/m2) subjects. Subjects performed 204 one- and two-handed unloaded activities at different vertical (0 to 180 cm from the floor) and horizontal (up to 60 cm lateral and/or anterior) destinations. Subsequently, the extrapolation capability of the trained/validated/tested ANNs was evaluated using data collected from fifteen additional subjects (unseen by the ANNs); three individuals in five groups: underweight, overweight, obese, left-handed individuals, and subjects with a hand-load. Results indicated that the ANNs predicted body joint coordinates and angles during various activities with errors of ∼ 25 mm and ∼ 10°, respectively; considerable improvements when compared to previous posture-prediction ANNs. Extrapolation errors of our ANNs generally remained within the error range of existing ANNs with obesity and being left-handed having, respectively, the most and least compromising effects on their accuracy. These easy-to-use ANNs appear, therefore, to be robust alternatives to common posture-measurement approaches.

Validation of an EMG submaximal method to calibrate a novel dynamic EMG-driven musculoskeletal model of the trunk: Effects on model estimates

BACKGROUND

Multijoint EMG-assisted optimization models are reliable tools to predict muscle forces as they account for inter- and intra-individual variations in activation. However, the conventional method of normalizing EMG signals using maximum voluntary contractions (MVCs) is problematic and introduces major limitations. The sub-maximal voluntary contraction (SVC) approaches have been proposed as a remedy, but their performance against the MVC approach needs further validation particularly during dynamic tasks.

METHODS

To compare model outcomes between MVC and SVC approaches, nineteen healthy subjects performed a dynamic lifting task with two loading conditions.

RESULTS

Results demonstrated that these two approaches produced highly correlated results with relatively small absolute and relative differences (<10 %) when considering highly-aggregated model outcomes (e.g. compression forces, stability indices). Larger differences were, however, observed in estimated muscle forces. Although some model outcomes, e.g. force of abdominal muscles, were statistically different, their effect sizes remained mostly small (η_G² ≤ 0.13) and in a few cases moderate (η_G² ≤ 0.165).

CONCLUSION

The findings highlight that the MVC calibration approach can reliably be replaced by the SVC approach when the true MVC exertion is not accessible due to pain, kinesiophobia and/or the lack of proper training.

Computational lumbar spine models: A literature review

BACKGROUND

Computational spine models of various types have been employed to understand spine function, assess the risk that different activities pose to the spine, and evaluate techniques to prevent injury. The areas in which these models are applied has expanded greatly, potentially beyond the appropriate scope of each, given their capabilities. A comprehensive understanding of the components of these models provides insight into their current capabilities and limitations.

METHODS

The objective of this review was to provide a critical assessment of the different characteristics of model elements employed across the spectrum of lumbar spine modeling and in newer combined methodologies to help better evaluate existing studies and delineate areas for future research and refinement.

FINDINGS

A total of 155 studies met selection criteria and were included in this review. Most current studies use either highly detailed Finite Element models or simpler Musculoskeletal models driven with in vivo data. Many models feature significant geometric or loading simplifications that limit their realism and validity. Frequently, studies only create a single model and thus can't account for the impact of subject variability. The lack of model representation for certain subject cohorts leaves significant gaps in spine knowledge. Combining features from both types of modeling could result in more accurate and predictive models.

INTERPRETATION

Development of integrated models combining elements from different model types in a framework that enables the evaluation of larger populations of subjects could address existing voids and enable more realistic representation of the biomechanics of the lumbar spine.

Computational biomechanics of human knee joint in stair ascent: Muscle-ligament-contact forces and comparison with level walking

About a third of knee joint disorders originate from the patellofemoral (PF) site that makes stair ascent a difficult activity for patients. A detailed finite element model of the knee joint is coupled to a lower extremity musculoskeletal model to simulate the stance phase of stair ascent. It is driven by the mean of measurements on the hip-knee-ankle moments-angles as well as ground reaction forces reported in healthy individuals. Predicted muscle activities compare well to the recorded electromyography data. Peak forces in quadriceps (3.87 BW, body weight, at 20% instance in our 607 N subject), medial hamstrings (0.77 BW at 20%), and gastrocnemii (1.21 BW at 80%) are estimated. Due to much greater flexion angles-moments in the first half of stance, large PF contact forces (peak of 3.1 BW at 20% stance) and stresses (peak of 4.83 MPa at 20% stance) are estimated that exceed their peaks in level walking by fourfold and twofold, respectively. Compared with level walking, ACL forces diminish in the first half of stance but substantially increase later in the second half (peak of 0.76 BW at 75% stance). Under nearly similar contact forces at 20% of stance, the contact stress on the tibiofemoral (TF) medial plateau reaches a peak (9.68 MPa) twice that on the PF joint suggesting the vulnerability of both joints. Compared with walking, stair ascent increases peak ACL force and both peak TF and PF contact stresses. Reductions in the knee flexion moment and/or angle appear as a viable strategy to mitigate internal loads and pain.

Influence of Load Knowledge on Biomechanics of Asymmetric Lifting

BACKGROUND

Load knowledge has been identified as a factor affecting the risk of low back pain (LBP) during symmetric lifting. However, the effects of load knowledge in asymmetric lifting tasks have not been reported yet. The purpose of this study was to investigate the load knowledge influence on lifting biomechanics in asymmetric lifting tasks; Methods: Twenty-four male adults were recruited to complete a psychophysical lifting capacity test and a simulated asymmetric lifting task. The lifting task was set with load knowledge of 'no knowledge' (NK), 'weight known' (WK), 'fragile material known' (FK), and 'weight and fragile material known' (WFK) for different lifting load weights. Trunk kinematics and kinetics were collected and analyzed; Results: When fragility information was presented, trunk sagittal flexion acceleration, lateral flexion velocity and acceleration, and average lateral bending moment were significantly lowered at the deposit phase. Lifting a high load weight was found to significantly increase low back sagittal bending moment at the lifting phase and low back moments of all three dimensions at the deposit phase; Conclusions: The decrease of trunk kinematic load suggests that providing material fragility information to workers in asymmetric lifting tasks would be effective in reducing their risk of LBP.

Evaluating stability of human spine in static tasks: a combined in vivo-computational study

Various interpretations and parameters have been proposed to assess spinal stability such as antagonist muscle coactivity, trunk stiffness and spinal buckling load; however, the correlation between these parameters remains unknown. We evaluated spinal stability during different tasks while changing the external moment and load height and investigated likely relationships between different EMG- and model-based parameters (e.g., EMG coactivity ratio, trunk stiffness, force coactivity ratio) and stability margins. EMG and kinematics of 40 young healthy subjects were recorded during various quasi-static tasks. Muscle forces, trunk stiffness and stability margins were calculated by a nonlinear subject-specific EMG-assisted-optimization musculoskeletal model of the trunk. The load elevation and external moment increased muscle activities and trunk stiffness while all stability margins (i.e., buckling loads) decreased. The force coactivity ratio was strongly correlated with the hand-load stability margin (i.e., additional weight in hands to initiate instability; R² = 0.54) demonstrating the stabilizing role of abdominal muscles. The total trunk stiffness (Pearson's r = 0.96) and the sum of EMGs of back muscles (Pearson's r = 0.65) contributed the most to the T1 stability margin (i.e., additional required load at T1 for instability/buckling). Force coactivity ratio and trunk stiffness can be used as alternative spinal stability metrics.

Computational modelling of the scoliotic spine: A literature review

Scoliosis is a deformity of the spine that in severe cases requires surgical treatment. There is still disagreement among clinicians as to what the aim of such treatment is as well as the optimal surgical technique. Numerical models can aid clinical decision-making by estimating the outcome of a given surgical intervention. This paper provided some background information on the modelling of the healthy spine and a review of the literature on scoliotic spine models, their validation, and their application. An overview of the methods and techniques used to construct scoliotic finite element and multibody models was given as well as the boundary conditions used in the simulations. The current limitations of the models were discussed as well as how such limitations are addressed in non-scoliotic spine models. Finally, future directions for the numerical modelling of scoliosis were addressed.

Comparison of different lifting analysis tools in estimating lower spinal loads - Evaluation of NIOSH criterion

Excessive loads on the human spine is recognized as a risk factor for back injuries/pain. Various lifting analysis tools such as musculoskeletal models, regression equations and NIOSH (National Institute for Occupational Safety and Health) lifting equation (NLE) have been proposed to evaluate and mitigate associated risks during manual material handling activities. Present study aims to compare predicted spinal loads from 5 different lifting analysis tools as well as to critically evaluate the NIOSH recommended weight limit (RWL). Spinal loads were estimated under different symmetric/asymmetric lifting tasks in which hand-load mass at each task was set based on RWL from NLE. Estimated intradiscal pressures (IDPs) of various tools were also compared with in vivo measurements. We compared RWL by NLE versus our estimations of RWL calculated from our regression equations using biomechanical criteria (compression <3400 N with/without shear <1000, 1250 or 1500 N). Our regression equations followed by OpenSim, AnyBody, simple polynomial and 3DSSPP satisfactorily predicted L4-L5 IDPs. Lifting analysis tools estimated comparable spinal compression forces (mean Pearson's r = 0.80; standard deviation of relative difference = 26%) while in shear, differences were greater (mean Pearson's r = 0.68; standard deviation of relative difference = 56%). NLE estimations of RWL were conservative in comparison with our estimations for lean individuals (BMI < 25 kg/m²) when compression <3400 N and shear <1250 N were considered as the biomechanical criteria. For heavier individuals, however, NLE estimations of RWL generated spinal compression >3400 N (NIOSH biomechanical safety threshold) as well as shear >1000 N. Although RWLs estimated by NLE was body weight independent, body weight substantially altered RWLs estimated from our regression equations. For improved estimation of the risk of injury, more accurate failure criteria for spinal segments are essential.