Enhanced Local Pattern Transformation Based Feature Extraction for Identification of Parkinson’s Disease Using Gait Signals

One of the most common ailments, especially among the elderly, is Parkinson's disease (PD). Although previous research has demonstrated that heuristics can diagnose Parkinson's disease using decisive signs like tremor, muscular rigidity, movement disorders, and voice disorders, it has also been reported that current approaches, which rely on simple motor tasks, are limited and lack stability and accessibility. The purpose of this study is to identify a novel cost-effective and time-efficient early detection technique for the prediction of this disease using a signal processing feature extraction approach namely, Shifted Extended Local Binary Pattern (S-ELBP) using gait signals. The features extracted using the proposed methods are given as the input to an artificial neural network (ANN) to classify them as Healthy or Parkinson’s. The proposed method has quite promising results when evaluated using different performance metrics. The method has yielded accuracy: 97.6%, specificity: 95.71%, sensitivity: 99, positive predictive value (PPV): 97.2%, negative predictive value (NPV): 98.8%, Matthews correlation coefficient (MCC): 95.4%, F1-score: 97.9%, and geometric mean: 97.19%. © 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.