Predictive Modeling of DWT-decomposed ALS-EMG Features Using Group Method of Data Handling

摘要: Identification of Amyotrophic Lateral Sclerosis disorder using electromyography signal data requires accurate classification models. Traditional approaches often necessitate manual model control and do not provide the designer with a set practical equations. In this paper, Group Method Data Handling-based methodology is proposed which automatically builds by heuristically estimating optimum quadratic polynomial equations from available empirical data. Based on guidelines in literature diligent trial error assessment, reduced most favorable features was tuned assembled after extraction selected sub-band Discrete Wavelet Transform decompositions. The generated are an approximate representation relationship between training corresponding class labels. For evaluation prediction performance, computational experiments were conducted test results compiled grounds Leave-one-out cross-validation criterion. trained effectively identified diseased healthy groups Sensitivity, Specificity Accuracy 90.34%, 94.89% 92.55% improvement 0.68%, 2.92% 1.77% respectively, over conventional Multilayer Perceptron classifier tested under similar experimental conditions. These outcomes indicate that approach could be extended to real clinical setting assist clinicians making precise diagnosis potentially compete leading automated diagnostic support systems.