DeepFoG: An IMU-Based Detection of Freezing of Gait Episodes in Parkinson’s Disease Patients via Deep Learning

Bikias, Thomas; Iakovakis, Dimitrios; Hadjidimitriou, Stelios; Charisis, Vasileios; Hadjileontiadis, Leontios J.

doi:10.3389/frobt.2021.537384

Cited by 51 publications

(34 citation statements)

References 35 publications

(42 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This would allow for long-term recording, providing the large number of training events needed for algorithms to learn freezing signals in the individual patient in order to subsequently predict FOG in real time. Deep learning has already been deployed to automatically detect gait freezing in video recorded walks ( Hu et al, 2020 ) and also using real-time inertial measurements from wearable devices ( Bikias et al, 2021 ). One group has recently developed an algorithm for use in patients without any previous anomalous gait data, trained on reference accelerometer data from a small group of reference normal and anomalous gaits, identifying 87.4% of FOG onsets ( Bikias et al, 2021 ).…”

Section: What Approaches Could Help Us Identify a New Treatment?mentioning

confidence: 99%

“…Deep learning has already been deployed to automatically detect gait freezing in video recorded walks ( Hu et al, 2020 ) and also using real-time inertial measurements from wearable devices ( Bikias et al, 2021 ). One group has recently developed an algorithm for use in patients without any previous anomalous gait data, trained on reference accelerometer data from a small group of reference normal and anomalous gaits, identifying 87.4% of FOG onsets ( Bikias et al, 2021 ). Multi-modal measurements combining accelerometer and EEG readings are more accurate than single-modality measurement in detecting FOG events ( Wang et al, 2020 ), suggesting future systems may require integration of different inputs.…”

Section: What Approaches Could Help Us Identify a New Treatment?mentioning

confidence: 99%

See 1 more Smart Citation

Future Therapeutic Strategies for Freezing of Gait in Parkinson’s Disease

Cui

Lewis

2021

Front. Hum. Neurosci.

View full text Add to dashboard Cite

Freezing of gait (FOG) is a common and challenging clinical symptom in Parkinson’s disease. In this review, we summarise the recent insights into freezing of gait and highlight the strategies that should be considered to improve future treatment. There is a need to develop individualised and on-demand therapies, through improved detection and wearable technologies. Whilst there already exist a number of pharmacological (e.g., dopaminergic and beyond dopamine), non-pharmacological (physiotherapy and cueing, cognitive training, and non-invasive brain stimulation) and surgical approaches to freezing (i.e., dual-site deep brain stimulation, closed-loop programming), an integrated collaborative approach to future research in this complex area will be necessary to systematically investigate new therapeutic avenues. A review of the literature suggests standardising how gait freezing is measured, enriching patient cohorts for preventative studies, and harnessing the power of existing data, could help lead to more effective treatments for freezing of gait and offer relief to many patients.

show abstract

Section: What Approaches Could Help Us Identify a New Treatment?mentioning

confidence: 99%

Section: What Approaches Could Help Us Identify a New Treatment?mentioning

confidence: 99%

Future Therapeutic Strategies for Freezing of Gait in Parkinson’s Disease

Cui

Lewis

2021

Front. Hum. Neurosci.

View full text Add to dashboard Cite

show abstract

“…The wrist IMU, though rated highly wearable, demonstrated poor performance (AUROC = 0.56). This finding may be limited by the four-participant dataset, as a recent convolutional neural network trained on wrist IMU data from eleven participants showed comparable performance to our best technical set (sensitivity = 0.83 and specificity = 0.88, (37)).…”

Section: Discussionmentioning

confidence: 73%

Assessing inertial measurement unit locations for freezing of gait detection and patient preference

O’Day

Lee²,

Seagers³

et al. 2021

Preprint

View full text Add to dashboard Cite

Background: Freezing of gait, a common symptom of Parkinson's disease, presents as sporadic episodes in which an individual's feet suddenly feel stuck to the ground. Inertial measurement units (IMUs) promise to enable at-home monitoring and personalization of therapy, but there is a lack of consensus on the number and location of IMUs for detecting freezing of gait. The purpose of this study was to assess IMU sets in the context of both freezing of gait detection performance and patient preference. Methods: Sixteen people with Parkinson's disease were surveyed about sensor preferences. Raw IMU data from seven people with Parkinson's disease, wearing up to eleven sensors, were used to train convolutional neural networks to detect freezing of gait. Models trained with data from different sensor sets were assessed for technical performance; a best technical set and minimal IMU set were identified. Clinical utility was assessed by comparing model- and human-rater-determined percent time freezing and number of freezing events. Results: The best technical set consisted of three IMUs (lumbar and both ankles, AUROC = 0.83), all of which were rated highly wearable. The minimal IMU set consisted of a single ankle IMU (AUROC = 0.80). Correlations between these models and human raters were good to excellent for percent time freezing (ICC = 0.93, 0.89) and number of freezing events (ICC = 0.95, 0.86), for the best technical set and minimal IMU set, respectively. Conclusions: Several IMU sets consisting of three IMUs or fewer were highly rated for both technical performance and wearability, and more IMUs did not necessarily perform better in FOG detection. We openly share our data and software to further the adoption of a general, open-source model that uses raw signals and a standard sensor set for at-home monitoring of freezing of gait.

show abstract

“…A 2-layer 1D CNN FOG detection model achieved slightly better sensitivity (83%) and lower specificity (88%) in a LOPO cross validation using acceleration data from a wrist mounted IMU [ 49 ], than the new 2-layer LSTM model in this paper. The CNN model used windows-based classification of data from 11 participants (184 FOG episodes) who froze, compared to our data from 7 participants (361 FOG episodes) who froze.…”

Section: Discussionmentioning

confidence: 99%

Prediction and detection of freezing of gait in Parkinson’s disease from plantar pressure data using long short-term memory neural-networks

Shalin

Pardoel

Lemaire

et al. 2021

J NeuroEngineering Rehabil

View full text Add to dashboard Cite

Background Freezing of gait (FOG) is a walking disturbance in advanced stage Parkinson’s disease (PD) that has been associated with increased fall risk and decreased quality of life. Freezing episodes can be mitigated or prevented with external intervention such as visual or auditory cues, activated by FOG prediction and detection systems. While most research on FOG detection and prediction has been based on inertial measurement unit (IMU) and accelerometer data, plantar-pressure data may capture subtle weight shifts unique to FOG episodes. Different machine learning algorithms have been used for FOG detection and prediction; however, long short-term memory (LSTM) deep learning methods hold an advantage when dealing with time-series data, such as sensor data. This research aimed to determine if LSTM can be used to detect and predict FOG from plantar pressure data alone, specifically for use in a real-time wearable system. Methods Plantar pressure data were collected from pressure-sensing insole sensors worn by 11 participants with PD as they walked a predefined freeze-provoking path. FOG instances were labelled, 16 features were extracted, and the dataset was balanced and normalized (z-score). The resulting datasets were classified using long short-term memory neural-network models. Separate models were trained for detection and prediction. For prediction models, data before FOG were included in the target class. Leave-one-freezer-out cross validation was used for model evaluation. In addition, the models were tested on all non-freezer data to determine model specificity. Results The best FOG detection model had 82.1% (SD 6.2%) mean sensitivity and 89.5% (SD 3.6%) mean specificity for one-freezer-held-out cross validation. Specificity improved to 93.3% (SD 4.0%) when ignoring inactive state data (standing) and analyzing the model only on active states (turning and walking). The model correctly detected 95% of freeze episodes. The best FOG prediction method achieved 72.5% (SD 13.6%) mean sensitivity and 81.2% (SD 6.8%) mean specificity for one-freezer-held-out cross validation. Conclusions Based on FOG data collected in a laboratory, the results suggest that plantar pressure data can be used for FOG detection and prediction. However, further research is required to improve FOG prediction performance, including training with a larger sample of people who experience FOG.

show abstract

DeepFoG: An IMU-Based Detection of Freezing of Gait Episodes in Parkinson’s Disease Patients via Deep Learning

Cited by 51 publications

References 35 publications

Future Therapeutic Strategies for Freezing of Gait in Parkinson’s Disease

Future Therapeutic Strategies for Freezing of Gait in Parkinson’s Disease

Assessing inertial measurement unit locations for freezing of gait detection and patient preference

Prediction and detection of freezing of gait in Parkinson’s disease from plantar pressure data using long short-term memory neural-networks

Contact Info

Product

Resources

About