Data augmentation of wearable sensor data for parkinson’s disease monitoring using convolutional neural networks

Um, Terry Taewoong; Pfister, Franz; Pichler, Daniel; Endo, Satoshi; Lang, M.; Hirche, Sandra; Fietzek, Urban; Kulić, Dana

doi:10.1145/3136755.3136817

Cited by 516 publications

(368 citation statements)

References 14 publications

Supporting

Mentioning

366

Contrasting

Unclassified

Order By: Relevance

“…One shortcoming of the probabilistic models is that the movements are represented at a single level of movement abstraction, and it is difficult to implement probabilistic modeling at multiple levels of movement abstraction. [128] Kinect v2 S Hand-crafted MA Saraee et al [156] Kinect v2 S Hand-crafted MA Taylor et al [45] Inertial sensor I Hand-crafted MC Zhang et al [48] Inertial sensor I Cross-correlation function MC Lin and Kulić [148] Inertial sensor S None MS Chen et al [183] Inertial sensor I Hand-crafted MC Zhang et al [138] Inertial sensor I None MA Houmanfar et al [53] Inertial sensor I Hand-crafted + LASSO MA Msayib et al [184] Inertial sensor S Hand-crafted MA Um et al [185] Inertial sensor I None DA Um et al [186] Inertial sensor I None MC Um et al [46] Inertial Table 4. Summary of approaches for evaluation of rehabilitation movements.…”

Section: B) Probability Density Functionsmentioning

confidence: 99%

A review of computational approaches for evaluation of rehabilitation exercises

Liao

Vakanski

Xian

et al. 2020

Computers in Biology and Medicine

View full text Add to dashboard Cite

Recent advances in data analytics and computer-aided diagnostics stimulate the vision of patient-centric precision healthcare, where treatment plans are customized based on the health records and needs of every patient. In physical rehabilitation, the progress in machine learning and the advent of affordable and reliable motion capture sensors have been conducive to the development of approaches for automated assessment of patient performance and progress toward functional recovery. The presented study reviews computational approaches for evaluating patient performance in rehabilitation programs using motion capture systems. Such approaches will play an important role in supplementing traditional rehabilitation assessment performed by trained clinicians, and in assisting patients participating in home-based rehabilitation. The reviewed computational methods for exercise evaluation are grouped into three main categories: discrete movement score, rule-based, and template-based approaches.The review places an emphasis on the application of machine learning methods for movement evaluation in rehabilitation. Related work in the literature on data representation, feature engineering, movement segmentation, and scoring functions is presented. The study also reviews existing sensors for capturing rehabilitation movements and provides an informative listing of pertinent benchmark datasets. The significance of this paper is in being the first to provide a comprehensive review of computational methods for evaluation of patient performance in rehabilitation programs.

show abstract

Section: B) Probability Density Functionsmentioning

confidence: 99%

A review of computational approaches for evaluation of rehabilitation exercises

Liao

Vakanski

Xian

et al. 2020

Computers in Biology and Medicine

View full text Add to dashboard Cite

show abstract

“…For the training of BANet and of other architectures evaluated for comparison, we apply two augmentation techniques, both previously used in [15] [39]. The first technique is based on creating new instances by adding normalized Gaussian noise to the original data with 3 different standard deviations: 0.05, 0.1 and 0.15.…”

Section: Data Preparation and Experimental Settingsmentioning

confidence: 99%

Learning Temporal and Bodily Attention in Protective Movement Behavior Detection

Wang

Peng

Olugbade

et al. 2019

2019 8th International Conference on Affective Computing and Intelligent Interaction Workshops and Demos (ACIIW)

View full text Add to dashboard Cite

Fig. 1. The overview of the Body Attention Network. Each body part is described by the joint angle plus energy. Data collected from feet were noisy and hence not used in this work.Abstract-For people with chronic pain, the assessment of protective behavior during physical functioning is essential to understand their subjective pain-related experiences (e.g., fear and anxiety toward pain and injury) and how they deal with such experiences (avoidance or reliance on specific body joints), with the ultimate goal of guiding intervention. Advances in deep learning (DL) can enable the development of such intervention. Using the EmoPain MoCap dataset, we investigate how attention-based DL architectures can be used to improve the detection of protective behavior by capturing the most informative temporal and body configurational cues characterizing specific movements and the strategies used to perform them. We propose an end-to-end deep learning architecture named BodyAttentionNet (BANet). BANet is designed to learn temporal and bodily parts that are more informative to the detection of protective behavior. The approach addresses the variety of ways people execute a movement (including healthy people) independently of the type of movement analyzed.Through extensive comparison experiments with other state-of-the-art machine learning techniques used with motion capture data, we show statistically significant improvements achieved by using these attention mechanisms. In addition, the BANet architecture requires a much lower number of parameters than the state of the art for comparable if not higher performances.

show abstract

“…Similar to image augmentation methods used to increase the amount of data available for training CNNs [16], Um et al [17] recently proposed several data augmentation methods for wearable sensors, and showed that augmentation significantly improved classification performance (for inertial data). We adopted three techniques from [17] (rotation, scaling, and jittering) to improve our model's invariance to factors such as IMU placement/orientation, shoe type, and gait type. For each training sample, we applied a random SO(3) rotation R (applied to a and ω for all data points in the sample), a random scaling factor, s ∈ [0.92, 1.02], to simulate faster or slower movement, and added zero-mean Gaussian noise n (σ = 0.075) to each channel.…”

Section: Trainingmentioning

confidence: 99%

LSTM-Based Zero-Velocity Detection for Robust Inertial Navigation

Wagstaff

Kelly

2018

2018 International Conference on Indoor Positioning and Indoor Navigation (IPIN)

109

View full text Add to dashboard Cite

We present a method to improve the accuracy of a zero-velocity-aided inertial navigation system (INS) by replacing the standard zero-velocity detector with a long short-term memory (LSTM) neural network. While existing threshold-based zero-velocity detectors are not robust to varying motion types, our learned model accurately detects stationary periods of the inertial measurement unit (IMU) despite changes in the motion of the user. Upon detection, zero-velocity pseudo-measurements are fused with a dead reckoning motion model in an extended Kalman filter (EKF). We demonstrate that our LSTM-based zero-velocity detector, used within a zero-velocity-aided INS, improves zero-velocity detection during human localization tasks. Consequently, localization accuracy is also improved.Our system is evaluated on more than 7.5 km of indoor pedestrian locomotion data, acquired from five different subjects. We show that 3D positioning error is reduced by over 34% compared to existing fixed-threshold zero-velocity detectors for walking, running, and stair climbing motions. Additionally, we demonstrate how our learned zero-velocity detector operates effectively during crawling and ladder climbing. Our system is calibration-free (no careful threshold-tuning is required) and operates consistently with differing users, IMU placements, and shoe types, while being compatible with any generic zero-velocityaided INS.

show abstract

Data augmentation of wearable sensor data for parkinson’s disease monitoring using convolutional neural networks

Cited by 516 publications

References 14 publications

A review of computational approaches for evaluation of rehabilitation exercises

A review of computational approaches for evaluation of rehabilitation exercises

Learning Temporal and Bodily Attention in Protective Movement Behavior Detection

LSTM-Based Zero-Velocity Detection for Robust Inertial Navigation

Contact Info

Product

Resources

About