An electromyogram (EMG) is the most widely used diagnostic tool for Amyotrophic Lateral Sclerosis (ALS). These signals are employed to distinguish between different neuromuscular exertion levels that are indicative of both normal and aberrant circumstances. The use of machine learning in this work is predicated on the development of an efficient and non-invasive Hybrid Quantum Machine Learning (H-QML) algorithm for the diagnosis of ALS disease. In this work, we introduce Hybrid-QML employing a Quanvolutional layer, which works on the input data by performing similar transformations as random convolutional filter layers, but locally on the data using multiple random quantum circuits. We provide a Hybrid Quantum Machine Learning method to diagnose amyotrophic lateral sclerosis (ALS) using electromyography (EMG) data from the biceps brachii muscles. The results indicate that when compared to the purely classical machine learning approach, the hybrid QML model has faster training, reduced loss, and higher test set accuracy. In order to do automated screening for ALS disease using signal processing, basic analysis and classification of EMG data from both healthy and ALS participants were recorded. Neural network classifiers meticulously analyze and categorize specific aspects of EMG data, like the mean and maximum amplitude, in order to automate the diagnosis of ALS disease. The EMG dataset was used for the experiments, and results showed that of the classifiers assessed, the ensemble decision tree performed significantly better, with an accuracy of 98.34%, followed by hybrid quantum machine learning with a 98.38% accuracy. The results of the experiment demonstrate that there is enough variation between normal EMG signals and ALS signals to enable automated screening. Since this approach uses small quantum circuits with minimal to no error correction, it may find practical applications in near-term quantum computing. Hybrid quantum filters maximize both loss and precision.