Towards Real-Time Detection of Freezing of Gait Using Wavelet Transform on Wireless Accelerometer Data

Abstract

Injuries associated with fall incidences continue to pose a significant burden to persons with Parkinson’s disease (PD) both in terms of human suffering and economic loss. Freezing of gait (FOG), which is one of the symptoms of PD, is a common cause of falls in this population. Although a significant amount of work has been performed to characterize/detect FOG using both qualitative and quantitative methods, there remains paucity of data regarding real-time detection of FOG, such as the requirements for minimum sensor nodes, sensor placement locations, and appropriate sampling period and update time. Here, the continuous wavelet transform (CWT) is employed to define an index for correctly identifying FOG. Since the CWT method uses both time and frequency components of a waveform in comparison to other methods utilizing only the frequency component, we hypothesized that using this method could lead to a significant improvement in the accuracy of FOG detection. We tested the proposed index on the data of 10 PD patients who experience FOG. Two hundred and thirty seven (237) FOG events were identified by the physiotherapists. The results show that the index could discriminate FOG in the anterior–posterior axis better than other two axes, and is robust to the update time variability. These results suggest that real time detection of FOG may be realized by using CWT of a single shank sensor with window size of 2 s and update time of 1 s (82.1% and 77.1% for the sensitivity and specificity, respectively). Although implicated, future studies should examine the utility of this method in real-time detection of FOG.

Towards Real-Time Detection of Gait Events on Different Terrains Using Time-Frequency Analysis and Peak Heuristics Algorithm

“…Continuous wavelet transform (CWT) is a widely used time-frequency analysis tool that effectively captures the general characteristics of the signal under observation [27]. Furthermore, CWT with a proper mother wavelet has shown robustness in detecting gait events [20,27,28,29]. The Morlet mother wavelet was adopted in this study to investigate the time-frequency relationship between the gait event and gait cycle [20].…”

Comparison of User Models Based on GMM-UBM and I-Vectors for Speech, Handwriting, and Gait Assessment of Parkinson’s Disease Patients

“…We computed the continuous wavelet transform with a Gaussian wavelet. The feature set is formed with the energy content in 8 frequency bands from the scalogram, three spectral centroids, the energy in the in the 1st, 2nd, and 3rd quartiles of the spectrum, the energy content in the locomotor band (0.5-3 Hz), the energy content in the freeze band (3)(4)(5)(6)(7)(8), and the freeze index, which is the ratio between the energy in the locomotor and freeze bands [15,16].…”

The turning and barrier course reveals gait parameters for detecting freezing of gait and measuring the efficacy of deep brain stimulation

“…Objective detection of freezing behavior using the TBC Many automated FOG detection algorithms have been reported (Coste et al, 2014;Kim et al, 2015;Moore et al, 2013Moore et al, , 2008Rezvanian and Lockhart, 2016;Zach et al, 2015), but are consistently limited by small datasets (Silva de Lima et al, 2017). Standardized datasets are needed, so that more comprehensive algorithms that cater to the extensive variance seen in Parkinson's impaired gait and FOG can be tested.…”

“…Abbreviations FOG = freezing of gait; DBS = deep brain stimulation; STN = subthalamic nucleus; TBC = Turning and Barrier Course; FW = forward walking; Inertial Measurement Unit = IMU; FOG-Q = Freezing of Gait Questionnaire; LOOCV = leave-one-out cross validation; AUROC = Area Under Receiver Operator Curve; UPDRS = Unified Parkinson's Disease Rating Scale assessments, and clinic-based measurements (Barthel et al, 2016). The authors concluded that there is no "unique methodological tool that encompasses the entire complexity of FOG" and "further development of such an assessment tool requires understanding and thorough analysis of the specific FOG characteristics" (Barthel et al, 2016).Several studies have employed wearable inertial sensors in a variety of different tasks, such as turning 360 degrees in place for two minutes, walking around cones, or walking during dual tasking, to monitor, detect and predict FOG (Coste et al, 2014;Khemani et al, 2015;Kim et al, 2015;Kwon et al, 2014;Palmerini et al, 2017;Rezvanian and Lockhart, 2016;Silva de Lima et al, 2017;Zach et al, 2015). These tasks have improved the detection resolution of FOG but are either not representative of real-world environments or still require a clinical rater to detect freezing episodes, and cannot objectively measure gait impairment, such as arrhythmicity, that is correlated with FOG (Anidi et al, 2018;Hausdorff, 2009;Nantel et al, 2011;Plotnik and Hausdorff, 2008;Syrkin-Nikolau et al, 2017).…”

“…Several studies have employed wearable inertial sensors in a variety of different tasks, such as turning 360 degrees in place for two minutes, walking around cones, or walking during dual tasking, to monitor, detect and predict FOG (Coste et al, 2014;Khemani et al, 2015;Kim et al, 2015;Kwon et al, 2014;Palmerini et al, 2017;Rezvanian and Lockhart, 2016;Silva de Lima et al, 2017;Zach et al, 2015). These tasks have improved the detection resolution of FOG but are either not representative of real-world environments or still require a clinical rater to detect freezing episodes, and cannot objectively measure gait impairment, such as arrhythmicity, that is correlated with FOG (Anidi et al, 2018;Hausdorff, 2009;Nantel et al, 2011;Plotnik and Hausdorff, 2008;Syrkin-Nikolau et al, 2017).…”