Can Wearable Devices and Machine Learning Techniques Be Used for Recognizing and Segmenting Modified Physical Performance Test Items?

Zhang, Yiyuan; Wang, Xiangyu; Han, Pengxuan; Verschueren, Sabine; Chen, Wei; Vanrumste, Bart

doi:10.1109/tnsre.2022.3186616

Cited by 3 publications

(1 citation statement)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The second block is a prediction refinement block that contains multiple stages, each with multiple temporal convolutional layers, to refine the initial predictions and prevent oversegmentation errors [32]. Although MS-TCN enabled stateof-the-art activity recognition in various applications that deal with IMU data [33], [34], previous studies have shown that a many-to-one training strategy [35] enables improved generalization [35], [36]. Therefore, we train the prediction generation block with a many-to-one training scheme for our FOG detection model [9].…”

Section: B Sample-based Methodsmentioning

confidence: 99%

Automatic Detection and Assessment of Freezing of Gait Manifestations

Yang

Filtjens

Ginis

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

Freezing of gait (FOG) is an episodic and highly disabling symptom of Parkinson's disease (PD). Although described as a single phenomenon, FOG is not univocal and can express as different manifestations, such as trembling in place or complete akinesia. We aimed to analyze the utility of deep learning trained on inertial measurement unit data to classify FOG into both manifestations. We developed a temporal convolutional neural network, which we compared to three state-of-the-art FOG detection algorithms that were adapted to the FOG manifestation detection task. Next, we investigated its performance in distinguishing between the two manifestations and other forms of movement cessation (e.g., volitional stopping and sitting) based on gold-standard video annotations. Experiments were conducted on a dataset of twelve PD patients with FOG that completed a FOG-provoking protocol, including the timed-up-and-go and 360-degree turning-in-place tasks during ON and OFF anti-Parkinsonian medication. The results showed that our model enables accurate detection of FOG manifestations with an 11.43% higher F1 score than the second-best model. Assessment of FOG manifestation severity was moderately strong for trembling in place (Intra-class Correlation Coefficient (ICC)=0.64, [0.16,0.88]) and strong for complete akinesia (ICC=0.87, [0.63,0.96]). Remarkably, our results show that complete akinesia can be distinguished from volitional stopping. In conclusion, we established that FOG manifestations could be accurately detected and assessed with deep learning. Future work should establish whether these results hold firm for a more extensive and varied verification cohort.

show abstract

Section: B Sample-based Methodsmentioning

confidence: 99%

Automatic Detection and Assessment of Freezing of Gait Manifestations

Yang

Filtjens

Ginis

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

A data-driven approach to detect upper limb functional use during daily life in breast cancer survivors using wrist-worn sensors

Emmerzaal,

Filtjens,

Vets

et al. 2024

Sci Rep

View full text Add to dashboard Cite

To gain insights into the impact of upper limb (UL) dysfunctions after breast cancer treatment, this study aimed to develop a temporal convolutional neural network (TCN) to detect functional daily UL use in breast cancer survivors using data from a wrist-worn accelerometer. A pre-existing dataset of 10 breast cancer survivors was used that contained raw 3-axis acceleration data and simultaneously recorded video data, captured during four daily life activities. The input of our TCN consists of a 3-axis acceleration sequence with a receptive field of 243 samples. The 4 ResNet TCN blocks perform dilated temporal convolutions with a kernel of size 3 and a dilation rate that increases by a factor of 3 after each iteration. Outcomes of interest were functional UL use (minutes) and percentage UL use. We found strong agreement between the video and predicted data for functional UL use (ICC = 0.975) and moderately strong agreement for %UL use (ICC = 0.794). The TCN model overestimated the functional UL use by 0.71 min and 3.06%. Model performance showed good accuracy, f1, and AUPRC scores (0.875, 0.909, 0.954, respectively). In conclusion, using wrist-worn accelerometer data, the TCN model effectively identified functional UL use in daily life among breast cancer survivors.The upper limbs (ULs) are essential to our everyday life. To clean our houses, to complete work tasks, to drink, to eat, to scratch, to help or touch other people, etc. Usually, we are unaware of how much we use our ULs until we are unable to properly use them. Life events such as surgery or treatment for breast cancer are moments that might suddenly affect UL function.After breast cancer treatment, an estimated 30-50% of women suffer from persistent UL dysfunction 1 . This persistent dysfunction goes beyond the natural healing time of tissue after surgery or radiotherapy, and it is considered one of the most troublesome long-term complications after breast cancer treatment 1-3 . Women suffering from UL dysfunction are less able to perform their daily tasks, have an increased risk for chronic pain, and suffer from participation difficulties, which lead to a decrease in their quality of life 1,3 . This impact on everyday life is currently only assessed using questionnaires. While questionnaires are easily administered and provide valuable information on a person's perception of UL function, they have important drawbacks such as recall bias and self-presentation bias (i.e. disclose only what the person wants to or is consciously aware of). Next to questionnaires, clinic-based assessments of body function, e.g. UL range of motion and activity or functional task performance, can measure the capability of a patient but might be poor indicators of actual UL use in daily life 4-6 . Thus, assessment of behaviour in a natural setting is vital, to assess the real-world impact of surgery on UL use and to evaluate UL recovery 7 .To assess UL function, a few definitions need to be addressed. UL function is previously defined by the functional arm activity behavioura...

show abstract

Freezing of gait assessment with inertial measurement units and deep learning: effect of tasks, medication states, and stops

Yang

Filtjens

Ginis

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

Background: Freezing of gait (FOG) is an episodic and highly disabling symptom of Parkinson's Disease (PD). Traditionally, FOG assessment relies on time-consuming visual inspection of camera footage. Therefore, previous studies have proposed portable and automated solutions to annotate FOG. However, automated FOG assessment is challenging due to gait variability caused by medication effects and varying FOG-provoking tasks. Moreover, whether automated approaches can differentiate FOG from typical everyday movements, such as volitional stops, remains to be determined. To address these questions, we evaluated an automated FOG assessment model with deep learning (DL) based on inertial measurement units (IMUs). We assessed its performance trained on all standardized FOG-provoking tasks and medication states, as well as on specific tasks and medication states. Furthermore, we examined the effect of adding stopping periods on FOG detection performance. Methods: Twelve PD patients with self-reported FOG (mean age 69.33 +/- 6.28 years) completed a FOG-provoking protocol, including timed-up-and-go and 360-degree turning-in-place tasks in On/Off dopaminergic medication states with/without volitional stopping. IMUs were attached to the pelvis and both sides of the tibia and talus. A multi-stage temporal convolutional network was developed to detect FOG episodes. FOG severity was quantified by the percentage of time frozen (%TF) and the number of freezing episodes (#FOG). The agreement between the model-generated outcomes and the gold standard experts' video annotation was assessed by the intra-class correlation coefficient (ICC). Results: For FOG assessment in trials without stopping, the agreement of our model was strong (ICC(%TF) = 0.92 [0.68, 0.98]; ICC(#FOG) = 0.95 [0.72, 0.99]). Models trained on a specific FOG-provoking task could not generalize to unseen tasks, while models trained on a specific medication state could generalize to unseen states. For assessment in trials with stopping, the model trained on stopping trials made fewer false positives than the model trained without stopping (ICC(%TF) = 0.95 [0.73, 0.99]; ICC(#FOG) = 0.79 [0.46, 0.94]). Conclusion: A DL model trained on IMU signals allows valid FOG assessment in trials with/without stops containing different medication states and FOG-provoking tasks. These results are encouraging and enable future work investigating automated FOG assessment during everyday life.

show abstract

Can Wearable Devices and Machine Learning Techniques Be Used for Recognizing and Segmenting Modified Physical Performance Test Items?

Cited by 3 publications

References 35 publications

Automatic Detection and Assessment of Freezing of Gait Manifestations

Automatic Detection and Assessment of Freezing of Gait Manifestations

A data-driven approach to detect upper limb functional use during daily life in breast cancer survivors using wrist-worn sensors

Freezing of gait assessment with inertial measurement units and deep learning: effect of tasks, medication states, and stops

Contact Info

Product

Resources

About