Journal: Sensors (Basel, Switzerland)

Article Title: Real-Time Gait Event Detection Based on Kinematic Data Coupled to a Biomechanical Model †

doi: 10.3390/s17040671

Figure Lengend Snippet: Musculoskeletal model and data processing. The generic musculoskeletal model (3DGaitModel2392) was scaled to the subject’s dimensions. The plug-in gait marker placement protocol was used for the data collection. Joint kinematics were obtained from the measured marker data using a Kalman Smoothing algorithm . In our study, joint kinematics were obtained offline (discontinuous box), but this can be substituted by any online method to obtain joint kinematics. The real-time processing, outside of the discontinuous box, consisted of feeding the joint kinematics back to the scaled model, extracting the information required by the respective algorithms from this model and performing the gait and stance phase detection.

Article Snippet: The generic “3DGaitModel2392” musculoskeletal model was scaled to the subject’s dimensions using Opensim 3.3 [ , ].

Techniques: Marker

Journal: Science Advances

Article Title: Late acquisition of erect hindlimb posture and function in the forerunners of therian mammals

doi: 10.1126/sciadv.adr2722

Figure Lengend Snippet: ( A ) Time-calibrated phylogeny of study taxa, showing musculoskeletal models of the hindlimb; silhouettes not to scale. Numbered nodes are as follows: 1, Amniota; 2, Sauria; 3, Metopophora; 4, Sphenacodontoidea; 5, Neotherapsida; 6, Theriodontia; 7, Eutheriodontia; 8, Eucynodontia; 9, Mammaliaformes; 10, Prototribosphenida. ( B ) Example FFS for the cynodont Massetognathus , showing total volume and force magnitudes in each cardinal direction. ( C ) Force spaces were computed across a range of limb postures, with the base set of postures spanning a continuum of adduction (sprawled to erect) and flexion (crouched to extended, based on limb reference length L ref ; see Materials and Methods), shown here with Massetognathus . Note that while 30 combinations of adduction and flexion potentially exist, certain ones will not be kinematically feasible for various taxa because of the constraints of limb segment proportions and joint axis orientations. ( D ) The base set of postures was pitched forward and back at the hip to generate the full suite of postures explored.

Article Snippet: Three-dimensional, digital musculoskeletal models of the pelvis and right hindlimb of each taxon were created in OpenSim 3.3 ( ) following standard protocols ( ).

Techniques:

Journal: Sensors (Basel, Switzerland)

Article Title: Learning to Ascend Stairs and Ramps: Deep Reinforcement Learning for a Physics-Based Human Musculoskeletal Model

doi: 10.3390/s22218479

Figure Lengend Snippet: The proposed DRL method for the dynamic optimization of the forward dynamics of a human musculoskeletal model during stairs or ramp ascent.

Article Snippet: The 3D lower-extremity physics-based human musculoskeletal model used in this study was developed in OpenSim 3.3 (model version number 3000) as an .osim file.

Techniques:

Journal: Sensors (Basel, Switzerland)

Article Title: Learning to Ascend Stairs and Ramps: Deep Reinforcement Learning for a Physics-Based Human Musculoskeletal Model

doi: 10.3390/s22218479

Figure Lengend Snippet: The physics-based human musculoskeletal model developed in this study. Figures from left to right: side view facing the right leg, front view, side view facing the left leg, and back view.

Article Snippet: The 3D lower-extremity physics-based human musculoskeletal model used in this study was developed in OpenSim 3.3 (model version number 3000) as an .osim file.

Techniques:

Journal: Sensors (Basel, Switzerland)

Article Title: Learning to Ascend Stairs and Ramps: Deep Reinforcement Learning for a Physics-Based Human Musculoskeletal Model

doi: 10.3390/s22218479

Figure Lengend Snippet: The state variables of the human musculoskeletal model.

Article Snippet: The 3D lower-extremity physics-based human musculoskeletal model used in this study was developed in OpenSim 3.3 (model version number 3000) as an .osim file.

Techniques: Plasmid Preparation

Journal: Sensors (Basel, Switzerland)

Article Title: Learning to Ascend Stairs and Ramps: Deep Reinforcement Learning for a Physics-Based Human Musculoskeletal Model

doi: 10.3390/s22218479

Figure Lengend Snippet: The reward obtained during the learning process of the human musculoskeletal model to ascend the stairs.

Article Snippet: The 3D lower-extremity physics-based human musculoskeletal model used in this study was developed in OpenSim 3.3 (model version number 3000) as an .osim file.

Techniques:

Journal: Sensors (Basel, Switzerland)

Article Title: Learning to Ascend Stairs and Ramps: Deep Reinforcement Learning for a Physics-Based Human Musculoskeletal Model

doi: 10.3390/s22218479

Figure Lengend Snippet: The reward obtained during the learning process of the human musculoskeletal model to ascend the ramp.

Article Snippet: The 3D lower-extremity physics-based human musculoskeletal model used in this study was developed in OpenSim 3.3 (model version number 3000) as an .osim file.

Techniques:

Journal: PLoS Computational Biology

Article Title: Computational modelling of muscle fibre operating ranges in the hindlimb of a small ground bird ( Eudromia elegans ), with implications for modelling locomotion in extinct species

doi: 10.1371/journal.pcbi.1008843

Figure Lengend Snippet: This is shown for walking (A) and running (B) trials. Grey regions denote the stance phase, white regions denote the swing phase. Note the strong contributions to external joint moment made by reserves acting about the abduction–adduction and long-axis rotation DOFs of the knee and ankle joints. Also illustrated are the kinematics of the left hindlimb (reversed for visualization) at various parts of the stride cycle.

Article Snippet: A high-fidelity 3-D musculoskeletal model of the pelvis and hindlimbs was developed for use in OpenSim 3.3 [ , ] ( and see ).

Techniques: