A Fast Walsh-Hadamard transform (FWHT) For Detection of Freezing of Gait in Parkinson’s disease using Machine learning techniques.

Freezing of gait stands out as the most severe detrimental motor signs of advanced PD patients with an increased prevalence in the elderly, and frequently leads to falls and injuries among individuals associated with PD. This research proposes an efficient detection method for freezing of gait based on the Fast Walsh -Hadamard Transform (FWHT). The approach involves decomposing the gait signals, and key features are derived by analyzing the statistical parameters and entropy characteristics of the decomposed coefficients. The Artificial Neural Network (ANN) and K-Nearest Neighbor classifier (KNN) are employed to classify the obtained statistical and entropy measures. The tests utilize the benchmark dataset DAPHNET for performance assessment. Remarkably, the K-nearest neighbor (KNN) algorithm demonstrated noteworthy outcomes, achieving 98.42%, 98.24%, 98.57%, 98.38%, and 98.45% accuracy, specificity, sensitivity, positive predictive value, and negative predictive value, respectively.