Prediction of Freezing of Gait in Parkinson’s Disease Using a Random Forest Model Based on an Orthogonal Experimental Design: A Pilot Study

Abstract

PurposeThe purpose of this study was to introduce an orthogonal experimental design (OED) to improve the efficiency of building and optimizing models for freezing of gait (FOG) prediction.MethodsA random forest (RF) model was developed to predict FOG by using acceleration signals and angular velocity signals to recognize possible precursor signs of FOG (preFOG). An OED was introduced to optimize the feature extraction parameters.ResultsThe main effects and interaction among the feature extraction hyperparameters were analyzed. The false-positive rate, hit rate, and mean prediction time (MPT) were 27%, 68%, and 2.99 s, respectively.ConclusionThe OED was an effective method for analyzing the main effects and interactions among the feature extraction parameters. It was also beneficial for optimizing the feature extraction parameters of the FOG prediction model.

Hypoglycemia Detection Using Hand Tremors: Home Study of Patients With Type 1 Diabetes

“…Collectively, the results provide support for the use and further development of ensemble techniques (such as random forest), KNN, SVM, or a combination of these models for hypoglycemia hand tremor detection. These results align with previous explorations of tremor detection [ 50 , 57 , 59 - 61 , 64 , 80 , 81 ]; however, our findings are focused on hypoglycemia.…”

“…Many classification approaches have been used to classify tremors versus normal states, mainly for Parkinson disease or essential tremor disorder [57][58][59][60][61][62][63][64]. However, these approaches have not been applied to tremors caused by hypoglycemia.…”

“…However, these approaches have not been applied to tremors caused by hypoglycemia. Tremor studies have used random forest [57][58][59][60], support vector machines (SVMs) [60,61,65,66], k-nearest neighbors (KNNs) [58,62,63], and naïve Bayes [58,64]. Among the current approaches, random forest, SVM, and KNN are the most widely used.…”

Machine Learning Approach to Support the Detection of Parkinson’s Disease in IMU-Based Gait Analysis

“…RF demonstrated good classification metrics using our data, despite being significantly different from SVM. Our results are consistent with those from other studies that described similar accuracy of the RF algorithm in distinguishing pwPD based on their disability level using IMUs-derived gait data [ 10 ], as well as in distinguishing the presence of freezing of gait [ 99 , 100 ]. Our study confirms the accuracy of RF as a useful algorithm to classify the gait abnormalities of pwPD from HS using trunk acceleration-derived data based on continuous walking tasks.…”

Machine Learning Approach to Support the Detection of Parkinson's Disease in IMU-Based Gait Analysis

“…RF demonstrated good classification metrics using our data, despite being significantly different from SVM. Our results are consistent with those from other studies that described similar accuracy of the RF algorithm in distinguishing pwPD based on their disability level using IMUs-derived gait data [10], as well as in distinguishing the presence of freezing of gait [99,100]. Our study confirms the accuracy of RF as a useful algorithm to classify the gait abnormalities of pwPD from HS using trunk accelerationderived data based on continuous walking tasks.…”