Autism Spectrum Disorder Detection with Quantum Machine Learning Methods

Abstract

Recent years have witnessed a captivating convergence between quantum physics and machine learning, promising an unprecedented leap in computational capabilities to tackle intricate problems. This study immerses itself in the world of quantum-inspired machine learning algorithms, meticulously comparing two innovative approaches: the Variational Quantum Classifier (VQC) and the Pegasos Quantum Support Vector Classifier (QSVC). Focused on autism research, this exploration delves into these algorithms' ability to decipher complex patterns within the data. The Variational Quantum Classifier (VQC) creatively employs quantum variational circuits, transforming input data into quantum states, offering a unique perspective on feature representation. Conversely, the Pegasos QSVC optimizes support vector machines efficiently, leveraging the principles of quantum computing. Through thorough experimentation and thoughtful analysis, this study assesses the algorithms' effectiveness in discerning autism from non-autism cases, considering aspects like accuracy, computational efficiency, and scalability. The outcomes of this comparative analysis illuminate the strengths and limitations inherent in both algorithms within the realm of autism classification. Moreover, the study delves into the profound implications of quantum computing in healthcare, underscoring its potential for revolutionary advancements in comprehending and diagnosing complex neurological disorders. This research not only enriches the expanding knowledge at the crossroads of quantum physics and machine learning but also charts a course for future investigations into the realm of quantum-enhanced healthcare applications.