Determining anatomical frames via inertial motion capture: A survey of methods

Vitali, Rachel V.; Perkins, Noel C.

doi:10.1016/j.jbiomech.2020.109832

Cited by 57 publications

(37 citation statements)

References 119 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To do this using IMU sensors, it is necessary to determine the sensor orientation with respect to the segment orientation in a global reference system (sensor-to-segment alignment). A variety of methods can be adopted for this purpose, all of which are dependent on sensor fusion algorithms and determining the initial sensor-to-segment alignment via the implementation of calibration postures or functional movements that the participant must execute [ 5 ]. Acceptable levels of within- and between-participant, and within- and between-tester repeatability, can only be achieved with appropriate sensor fusion algorithm application and accurate and reliable calibration procedures [ 6 , 7 , 8 ].…”

Section: Introductionmentioning

confidence: 99%

A Comparison of Three Neural Network Approaches for Estimating Joint Angles and Moments from Inertial Measurement Units

Mundt

Johnson

Potthast

et al. 2021

Sensors

View full text Add to dashboard Cite

The application of artificial intelligence techniques to wearable sensor data may facilitate accurate analysis outside of controlled laboratory settings—the holy grail for gait clinicians and sports scientists looking to bridge the lab to field divide. Using these techniques, parameters that are difficult to directly measure in-the-wild, may be predicted using surrogate lower resolution inputs. One example is the prediction of joint kinematics and kinetics based on inputs from inertial measurement unit (IMU) sensors. Despite increased research, there is a paucity of information examining the most suitable artificial neural network (ANN) for predicting gait kinematics and kinetics from IMUs. This paper compares the performance of three commonly employed ANNs used to predict gait kinematics and kinetics: multilayer perceptron (MLP); long short-term memory (LSTM); and convolutional neural networks (CNN). Overall high correlations between ground truth and predicted kinematic and kinetic data were found across all investigated ANNs. However, the optimal ANN should be based on the prediction task and the intended use-case application. For the prediction of joint angles, CNNs appear favourable, however these ANNs do not show an advantage over an MLP network for the prediction of joint moments. If real-time joint angle and joint moment prediction is desirable an LSTM network should be utilised.

show abstract

Section: Introductionmentioning

confidence: 99%

A Comparison of Three Neural Network Approaches for Estimating Joint Angles and Moments from Inertial Measurement Units

Mundt

Johnson

Potthast

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…We mitigate these uncertainties by using MOCAP data to establish sensor-to-segment alignment and to manually correct misidentified footfalls. While this yields a method that is not truly "MOCAP-free", these topics (sensor-to-segment alignment and still period detection from IMU data alone) are themselves active areas of research for human applications [39][40][41].…”

Section: Plos Onementioning

confidence: 99%

Error-state Kalman filter for lower-limb kinematic estimation: Evaluation on a 3-body model

et al. 2021

Self Cite

View full text Add to dashboard Cite

Human lower-limb kinematic measurements are critical for many applications including gait analysis, enhancing athletic performance, reducing or monitoring injury risk, augmenting warfighter performance, and monitoring elderly fall risk, among others. We present a new method to estimate lower-limb kinematics using an error-state Kalman filter that utilizes an array of body-worn inertial measurement units (IMUs) and four kinematic constraints. We evaluate the method on a simplified 3-body model of the lower limbs (pelvis and two legs) during walking using data from simulation and experiment. Evaluation on this 3-body model permits direct evaluation of the ErKF method without several confounding error sources from human subjects (e.g., soft tissue artefacts and determination of anatomical frames). RMS differences for the three estimated hip joint angles all remain below 0.2 degrees compared to simulation and 1.4 degrees compared to experimental optical motion capture (MOCAP). RMS differences for stride length and step width remain within 1% and 4%, respectively compared to simulation and 7% and 5%, respectively compared to experiment (MOCAP). The results are particularly important because they foretell future success in advancing this approach to more complex models for human movement. In particular, our future work aims to extend this approach to a 7-body model of the human lower limbs composed of the pelvis, thighs, shanks, and feet.

show abstract

“…The ensuing process of key release may also be reflected in the recorded brain dynamics; however, none of this behavioral and psychophysiological information is typically recorded, neither in past nor still current practice. Growing appreciation of the importance of the embodied cognition perspective on mental life (Shapiro, 2019), new lightweight, low cost methods of recording details of brain activities and motor behavior of experiment participants (Casson, 2019) (Jas et al, 2021) (Vitali & Perkins, 2020), and emergence of the practice of recording both brain activity and behavior (as well as psychophysiology) at higher resolution (sometimes termed Mobile Brain/Body Imaging or MoBI) (Makeig et al, 2009), make development of a suitable data annotation framework ever more urgent.…”

Section: Documenting Participant Responsesmentioning

confidence: 99%

“…Though here we focus on MEEG applications, event annotation standards and practices essential for MEEG data analysis can be applied equally well to other types of neuroimaging time series data including fMRI. For example, growing appreciation of the importance of embodied cognition on mental life (Shapiro, 2019), new lightweight, low cost methods of recording details of brain activities and motor behavior of experiment participants (Casson, 2019) (Jas et al, 2021) (Vitali & Perkins, 2020), and emergence of the practice of recording both brain activity and behavior (as well as psychophysiology) at higher resolution in a broader range of tasks and task environments (often termed Mobile Brain/Body Imaging or MoBI) (Makeig et al, 2009), make development of a suitable and more comprehensive data annotation framework ever more urgent.…”

Section: Introductionmentioning

confidence: 99%

Capturing the nature of events and event context using Hierarchical Event Descriptors (HED)

Robbins

Truong

Appelhoff

et al. 2021

Preprint

View full text Add to dashboard Cite

Because of the central role that event-related data analysis plays in EEG and MEG (MEEG) experiments, choices about which events to report and how to annotate their full natures can significantly influence the reliability, reproducibility, and value of MEEG datasets for further analysis. Current, more powerful annotation strategies combine robust event description with details of experiment design and metadata in a human-readable as well as machine-actionable form, making event annotation relevant to the full range of neuroimaging and other time series data. This paper dissects the event design and annotation process using as a case study the well-known multi-subject, multimodal dataset of Wakeman and Henson (openneuro.org, ds000117) shared by its authors using Brain Imaging Data Structure (BIDS) formatting (bids.neuroimaging.io). We propose a set of best practices and guidelines for event handling in MEEG research, examine the impact of various design decisions, and provide a working template for organizing events in MEEG and other neuroimaging data. We demonstrate how annotations using the new third-generation formulation of the Hierarchical Event Descriptors (HED-3G) framework and tools (hedtags.org) can document events occurring during neuroimaging experiments and their interrelationships, providing machine-actionable annotation enabling automated both within- and across-study comparisons and analysis, and point to a more complete BIDS formatted, HED-3G annotated edition of the MEEG portion of the Wakeman and Henson dataset (OpenNeuro ds003645).

show abstract

Determining anatomical frames via inertial motion capture: A survey of methods

Cited by 57 publications

References 119 publications

A Comparison of Three Neural Network Approaches for Estimating Joint Angles and Moments from Inertial Measurement Units

A Comparison of Three Neural Network Approaches for Estimating Joint Angles and Moments from Inertial Measurement Units

Error-state Kalman filter for lower-limb kinematic estimation: Evaluation on a 3-body model

Capturing the nature of events and event context using Hierarchical Event Descriptors (HED)

Contact Info

Product

Resources

About