Developing and validating an accelerometer-based algorithm with machine learning to classify physical activity after acquired brain injury

Honoré, Helene; Gade, Rikke; Nielsen, Jørgen Feldbæk; Mechlenburg, Inger

doi:10.1080/02699052.2021.1880026

Cited by 8 publications

(11 citation statements)

References 43 publications

(45 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…DT is another major approach for efficient classification, and it provides certain interpretability that is crucial in medical applications. Models like C4.5 and Random Forest (RF) showed competitive results in activity recognition and severity assessment [59], [77], [85], [93], [105]. Linear Discriminant Analysis (LDA), Naive Bayes (NB), k-nearest neighbor (k-NN), and shallow ANN are also used as typical ML techniques to build specific classifiers.…”

Section: B ML Based Methodsmentioning

confidence: 99%

IMU-Based Monitoring for Assistive Disease Diagnosis and Management of IoHT: A Review

Fan¹,

Yerebakan²,

Dai³

et al. 2022

Preprint

View full text Add to dashboard Cite

With the rapid development of Internet of Things (IoT) technologies, traditional disease diagnosis carried out in medical institutions can now be performed remotely at home or even ambient environments, yielding the concept of Internet of Health Things (IoHT). Among the diverse IoHT applications, inertial measurement units (IMUs) based systems play a significant role in the detection of diseases in many fields, such as neurological, musculoskeletal, and mental. However, traditional numerical interpretation methods have proven to be challenging to provide satisfying detection accuracies owing to the low quality of raw data especially under strong electromagnetic interference (EMI). To address this issue, in recent years, machine learning (ML) based techniques have been proposed to smartly map IMU captured data on disease detection and progress. After a decade of development, the combination of IMUs and ML algorithms for assistive disease diagnosis has become a hot topic, with an increasing number of studies reported yearly. To help readers comprehensively understand the fundamentals and state-of-the-art techniques, this article systematically reviews this field, by introducing and explaining relevant application scenarios, discussing challenges, and predicting foreseeable future trends. <br>

show abstract

Section: B ML Based Methodsmentioning

confidence: 99%

IMU-Based Monitoring for Assistive Disease Diagnosis and Management of IoHT: A Review

Fan¹,

Yerebakan²,

Dai³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The application of our algorithm was aimed at patients undergoing neurorehabilitation, and the training data collected in the development phase of this study were combined with the training data from a previous study [ 24 ], collected in a population of both healthy people and patients with acquired brain injury. The following method section only describes the data collected in this study.…”

Section: Methodsmentioning

confidence: 99%

Monitoring Physical Behavior in Rehabilitation Using a Machine Learning–Based Algorithm for Thigh-Mounted Accelerometers: Development and Validation Study

Skovbjerg¹,

Honoré²,

Mechlenburg³

et al. 2022

JMIR Bioinform Biotech

Self Cite

View full text Add to dashboard Cite

Background Physical activity is emerging as an outcome measure. Accelerometers have become an important tool in monitoring physical behavior, and newer analytical approaches of recognition methods increase the degree of details. Many studies have achieved high performance in the classification of physical behaviors through the use of multiple wearable sensors; however, multiple wearables can be impractical and lower compliance. Objective The aim of this study was to develop and validate an algorithm for classifying several daily physical behaviors using a single thigh-mounted accelerometer and a supervised machine-learning scheme. Methods We collected training data by adding the behavior classes—running, cycling, stair climbing, wheelchair ambulation, and vehicle driving—to an existing algorithm with the classes of sitting, lying, standing, walking, and transitioning. After combining the training data, we used a random forest learning scheme for model development. We validated the algorithm through a simulated free-living procedure using chest-mounted cameras for establishing the ground truth. Furthermore, we adjusted our algorithm and compared the performance with an existing algorithm based on vector thresholds. Results We developed an algorithm to classify 11 physical behaviors relevant for rehabilitation. In the simulated free-living validation, the performance of the algorithm decreased to 57% as an average for the 11 classes (F-measure). After merging classes into sedentary behavior, standing, walking, running, and cycling, the result revealed high performance in comparison to both the ground truth and the existing algorithm. Conclusions Using a single thigh-mounted accelerometer, we obtained high classification levels within specific behaviors. The behaviors classified with high levels of performance mostly occur in populations with higher levels of functioning. Further development should aim at describing behaviors within populations with lower levels of functioning.

show abstract

“…For arm use in daily life, there is no gold standard for comparison. Although uniaxial accelerometers show little disadvantage compared to the later developed triaxial accelerometer examining physical activity in preschool children or adolescents, future studies can use multiaxial accelerometers placed on all four limbs and around the waist on both children with CP and typically developing children, and use machine learning in the analyses process [37][38][39].…”

Section: Limitationsmentioning

confidence: 99%

Time-matched accelerometers on limbs and waist in children with CP give new insights into real-life activities after botulinum toxin treatment: A proof of concept study

Gantelius

Vikerfors

Edqvist

et al. 2023

PRM

View full text Add to dashboard Cite

PURPOSE: This study aimed to explore the feasibility of using time-matched uniaxial accelerometers for measuring movement in daily life in children with cerebral palsy (CP) before and after botulinum toxin injections. METHODS: This observational study of clinical care with a pre-post design was set in the home and school environment. Participants included eleven children (4–13 years of age) with CP (GMFCS I-III). The children wore uniaxial accelerometers (ActiGraph, model GT1M) for 4 days on both wrists, the right ankle and around the waist before, 3 weeks and 3 months after BoNT-A injections in the legs. Five children also got BoNT-A in the most affected arm. All injections were given according to clinical indications and routine. The accelerometers were all time-matched to define ambulation, arm swing, voluntary activity of arms, and bimanual activity. The feasibility of wearing accelerometers with this setup was evaluated. A linear mixed model was used for analysis of the percentage time and at which intensity the different activities were performed. The confidence interval demonstrated any difference between the dominant and non-dominant arm. RESULTS: Time-matching of accelerometers placed on both wrists, the waist, and one ankle is a feasible method of registering ambulation, arm swing during gait, and arm movements while not ambulating. Before injections, the children spent 5.6% of their time ambulating. This value declined to 3.9% at 3 months. Contrary to clinical goals, arm movement did not increase after injecting the most affected arm with BoNT-A, however, injections may have decreased mirror movements, which are often bothersome for the child. CONCLUSION: A time-matched 4-accelerometer set-up is feasible in children with cerebral palsy. A future study including time-matched multi-axial accelerometers on all four limbs, could provide important information on the effect of BoNT-A in daily life.

show abstract

Developing and validating an accelerometer-based algorithm with machine learning to classify physical activity after acquired brain injury

Abstract: Developing and validating an accelerometerbased algorithm with machine learning to classify physical activity after acquired brain injury. Brain Injury, 35(4), 460-467.

Cited by 8 publications

References 43 publications

IMU-Based Monitoring for Assistive Disease Diagnosis and Management of IoHT: A Review

IMU-Based Monitoring for Assistive Disease Diagnosis and Management of IoHT: A Review

Monitoring Physical Behavior in Rehabilitation Using a Machine Learning–Based Algorithm for Thigh-Mounted Accelerometers: Development and Validation Study

Time-matched accelerometers on limbs and waist in children with CP give new insights into real-life activities after botulinum toxin treatment: A proof of concept study

Contact Info

Product

Resources

About