Among Parkinson’s disease (PD) symptoms, freezing of gait (FoG) is one of the most debilitating. To assess FoG, current clinical practice mostly employs repeated evaluations over weeks and months based on questionnaires, which may not accurately map the severity of this symptom. The use of a non-invasive system to monitor the activities of daily living (ADL) and the PD symptoms experienced by patients throughout the day could provide a more accurate and objective evaluation of FoG in order to better understand the evolution of the disease and allow for a more informed decision-making process in making adjustments to the patient’s treatment plan. This paper presents a new algorithm to detect FoG with a machine learning approach based on Support Vector Machines (SVM) and a single tri-axial accelerometer worn at the waist. The method is evaluated through the acceleration signals in an outpatient setting gathered from 21 PD patients at their home and evaluated under two different conditions: first, a generic model is tested by using a leave-one-out approach and, second, a personalised model that also uses part of the dataset from each patient. Results show a significant improvement in the accuracy of the personalised model compared to the generic model, showing enhancement in the specificity and sensitivity geometric mean (GM) of 7.2%. Furthermore, the SVM approach adopted has been compared to the most comprehensive FoG detection method currently in use (referred to as MBFA in this paper). Results of our novel generic method provide an enhancement of 11.2% in the GM compared to the MBFA generic model and, in the case of the personalised model, a 10% of improvement with respect to the MBFA personalised model. Thus, our results show that a machine learning approach can be used to monitor FoG during the daily life of PD patients and, furthermore, personalised models for FoG detection can be used to improve monitoring accuracy.

Among Parkinson’s disease (PD) motor symptoms, freezing of gait (FOG) may\ud be the most incapacitating. FOG episodes may result in falls and reduce patients’\ud quality of life. Accurate assessment of FOG would provide objective information\ud to neurologists about the patient’s condition and the symptom’s characteristics,\ud while it could enable non-pharmacologic support based on rhythmic\ud cues.\ud This paper is, to the best of our knowledge, the first study to propose a\ud deep learning method for detecting FOG episodes in PD patients. This model\ud is trained using a novel spectral data representation strategy which considers information from both the previous and current signal windows. Our approach\ud was evaluated using data collected by a waist-placed inertial measurement unit\ud from 21 PD patients who manifested FOG episodes. These data were also employed\ud to reproduce the state-of-the-art methodologies, which served to perform\ud a comparative study to our FOG monitoring system.\ud The results of this study demonstrate that our approach successfully outperforms\ud the state-of-the-art methods for automatic FOG detection. Precisely, the\ud deep learning model achieved 90% for the geometric mean between sensitivity\ud and specificity, whereas the state-of-the-art methods were unable to surpass the\ud 83% for the same metric.Peer ReviewedPostprint (published version

No abstract

Freezing of gait (FOG) is a common motor symptom of Parkinson's Disease (PD), which presents itself as an inability to initiate or continue gait. This paper presents a method to monitor FOG episodes based only on acceleration measurements obtained from a waist-worn device. Three approximations of this method are tested. Initially, FOG is directly detected by a support vector machine (SVM). Then, classifier's outputs are aggregated over time to determine a confidence value, which is used for the final classification of freezing (i.e. second and third approach). All variations are trained with signals of 15 patients and evaluated with signals from another 5 patients. Using a linear SVM kernel, the third approach outperforms results in the literature with 98.7% accuracy and a geometric mean of 96.1%. Moreover, it is investigated whether frequency features are enough to reliably detect FOG. Results show that these features allow the method to detect FOG with accuracies above 90% and that frequency features enable a reliable monitoring of FOG by using simply a waist sensor.

The REMPARK System detects an accurate evaluation of ON-OFF fluctuations in PD; this technology paves the way for an optimisation of the symptomatic control of PD motor symptoms as well as an accurate assessment of medication efficacy.