Analysis of the performance of 17 algorithms from a systematic review: Influence of sensor position, analysed variable and computational approach in gait timing estimation from IMU measurements

Panebianco, Giulia Pacini; Bisi, Maria Cristina; Stagni, Rita; Fantozzi, Silvia

doi:10.1016/j.gaitpost.2018.08.025

Cited by 159 publications

(185 citation statements)

References 26 publications

Supporting

Mentioning

161

Contrasting

Order By: Relevance

“…The proliferation of studies using accelerometry (ACC) for gait analysis has been accompanied by an increasing number of methodologies seeking to detect IC and FC events as accurately as possible [11]- [17]. The performance of these gait event detection (GED) algorithms can vary greatly depending on sensor location, computational approach, and cadence [10], [18], [19]. Furthermore, not all methods provide a means for FC detection [16], [20], [21], limiting the number of temporal gait parameters which can be derived.…”

Section: Introductionmentioning

confidence: 99%

“…To improve FC detection, it has been suggested that shankmounted ACC may provide more signal information which could be used to identify instances of FC [19]. The choice of sensor location is an important factor to consider in study design with regard to ease of application, subject compliance and the range activities that can be detected [22], [23].…”

Section: Introductionmentioning

confidence: 99%

“…For monitoring tasks of daily living, waist mounted accelerometers placed near the centre of mass are often preferable as sensor positioning is relatively simple and energy expenditure from several activities can be estimated with a single sensor [22]- [24]. However, comparative studies have shown that waist mounted sensors perform worse than shank mounted sensors for detecting ICs and FCs [10], [19], potentially influencing the accuracy of temporal gait cycle parameters obtained in situations where gait variability is increased. As shank mounted accelerometers are closer to the point of IC and FC, they may intrinsically capture information integrated in the ACC signal that could be absent when recorded at the waist [19].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Gait Event Detection From Accelerometry Using the Teager–Kaiser Energy Operator

Flood

O’Callaghan

Lowery

2020

IEEE Trans. Biomed. Eng.

View full text Add to dashboard Cite

A novel method based on the application of the Teager-Kaiser Energy Operator is presented to estimate instances of initial contact (IC) and final contact (FC) from accelerometry during gait. The performance of the proposed method was evaluated against four existing gait event detection (GED) methods under three walking conditions designed to capture the variance of gait in real-world environments. Methods: A symmetric discrete approximation of the Teager-Kaiser energy operator was used to capture simultaneous amplitude and frequency modulations of the shank acceleration signal at IC and FC during flat treadmill walking, inclined treadmill walking, and flat indoor walking. Accuracy of estimated gait events were determined relative to gait events detected using force-sensitive resistors. The performance of the proposed algorithm was assessed against four established methods by comparing mean-absolute error, sensitivity, precision and F1-score values. Results: The proposed method demonstrated high accuracy for GED in all walking conditions, yielding higher F1-scores (IC: >0.98, FC: >0.9) and lower mean-absolute errors (IC: <0.018s, FC: <0.039s) than other methods examined. Estimated ICs from shank-based methods tended to exhibit unimodal distributions preceding the forcesensitive resistor estimated ICs, whereas estimated gait events for waist-based methods had quasi-uniform random distributions and lower accuracy. Conclusion: Compared to established gait event detection methods, the proposed method yielded comparably high accuracy for IC detection, and was more accurate than all other methods examined for FC detection. Significance: The results support the use of the Teager-Kaiser Energy Operator for accurate automated GED across a range of walking conditions.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Gait Event Detection From Accelerometry Using the Teager–Kaiser Energy Operator

Flood

O’Callaghan

Lowery

2020

IEEE Trans. Biomed. Eng.

View full text Add to dashboard Cite

show abstract

“…Moreover, they are light, inexpensive, non-invasive and easy to use, which potentiates their implementation in clinical settings (8)(9)(10)(11). To allow for routine use of IMUs, algorithms for automated step detection and computation of gait features of interest have been developed (12). However, few algorithms have been validated in individuals with progressive MS (pMS) (13), and individuals with severe disease are often not served well (14,15).…”

Section: Introductionmentioning

confidence: 99%

“…Early works on the topic often used single IMUs and failed to adapt to different types of cohorts (19). More recent work with bilateral lower limb sensors has provided promising results for moderate to severe conditions but is still rare in the literature (12,20). For these reasons, the use of templates or several techniques based on machine learning (18,21) or Dynamic Time Warping (22)(23)(24) has been advocated in several articles (25,26) as a way to automatically learn the characteristics of a cohort.…”

Section: Introductionmentioning

confidence: 99%

Personalized Template-Based Step Detection From Inertial Measurement Units Signals in Multiple Sclerosis

et al. 2020

View full text Add to dashboard Cite

Background: Objective gait assessment is key for the follow-up of patients with progressive multiple sclerosis (pMS). Inertial measurement units (IMUs) provide reliable and yet easy quantitative gait assessment in routine clinical settings. However, to the best of our knowledge, no automated step-detection algorithm performs well in detecting severely altered pMS gait.Method: This article elaborates on a step-detection method based on personalized templates tested against a gold standard. Twenty-two individuals with pMS and 10 young healthy subjects (HSs) were instructed to walk on an electronic walkway wearing synchronized IMUs. Templates were derived from the IMU signals by using Initial and Final Contact times given by the walkway. These were used to detect steps from other gait trials of the same individual (intra-individual template-based detection, IITD) or another participant from the same group (pMS or HS) (intra-group template-based detection, IGTD). All participants were seen twice with a 6-month interval, with two measurements performed at each visit. Performance and accuracy metrics were computed, along with a similarity index (SId), which was computed as the mean distance between detected steps and their respective closest template.Results: For HS participants, both the IITD and the IGTD algorithms had precision and recall of 1.00 for detecting steps. For pMS participants, precision and recall ranged from 0.94 to 1.00 for IITD and 0.85 to 0.95 for IGTD depending on the level of disability. The SId was correlated with performance and the accuracy of the result. An SId threshold of 0.957 (IITD) and 0.963 (IGTD) could rule out decreased performance (F-measure ≤ 0.95), with negative predictive values of 0.99 and 0.96 with the IITD and IGTD algorithms. Also, the SId computed with the IITD and IGTD algorithms could distinguish individuals showing changes at 6-month follow-up.Conclusion: This personalized step-detection method has high performance for detecting steps in pMS individuals with severely altered gait. The algorithm can be Vienne-Jumeau et al. PersonalizedStep Detection in Multiple Sclerosis self-evaluating with the SI, which gives a measure of the confidence the clinician can have in the detection. What is more, the SId can be used as a biomarker of change in disease severity occurring between the two measurement times.

show abstract

A Systematic Study of the Influence of Various User Specific and Environmental Factors on Wearable Human Body Capacitance Sensing

Bian

Lukowicz

2022

Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering

View full text Add to dashboard Cite

Analysis of the performance of 17 algorithms from a systematic review: Influence of sensor position, analysed variable and computational approach in gait timing estimation from IMU measurements

Cited by 159 publications

References 26 publications

Gait Event Detection From Accelerometry Using the Teager–Kaiser Energy Operator

Gait Event Detection From Accelerometry Using the Teager–Kaiser Energy Operator

Personalized Template-Based Step Detection From Inertial Measurement Units Signals in Multiple Sclerosis

A Systematic Study of the Influence of Various User Specific and Environmental Factors on Wearable Human Body Capacitance Sensing

Contact Info

Product

Resources

About