With the growing interest in hybrid quantumclassical intelligence, Quantum Machine Learning (QML) offers a new frontier for learning complex, highdimensional data like EEG signals. Most current QML methodologies, however, are plagued by instability against noisy quantum states, low fidelity, and poor feature encoding mechanisms. To overcome these drawbacks, this study introduces a new framework named “Enhancing Quantum Machine Learning Models with Error-Tolerant Superposition-Based Feature Extraction and State Fidelity Boosting”. The approach combines angle-based quantum superposition embedding with fidelity-based quantum autoencoder for retaining the integrity of quantum states during training. Classification is performed using a Variational Quantum Circuit (VQC) with fidelity loss minimization included in training to eliminate decoherence effects. The model is deployed with Pennylane and Qiskit on Python, and tested on a publicly shared EEG dataset from Kaggle on emotion classification. The model has a classification accuracy of 93.5%, a 4.3% increase compared to the bestperforming current QML method. The fidelity score was also enhanced to 0.95, which ensures that robust state is well-preserved. The work unfolds tremendous potential to employ QML in actual bio-signal applications. The fidelity-aware filtering and superposition encoding pipeline of the proposed method can be trivially generalized to other high-noise, high-dimensional quantum learning problems.
Sleep staging is an important task in the field of medicine and is significant for diagnosing sleep disorders and assessing overall health. Classical machine learning has made significant progress in this field, and with the development of quantum computing, quantum machine learning (QML) provides a new model for solving complex data classification tasks. In this paper, we propose a hybrid quantum-classical neural network (HQCNN) architecture based on electroencephalogram (EEG) single-channel signals for automating the sleep staging task. The classical part uses a convolutional neural network (CNN) and a long short-term memory network (LSTM) to extract features of the EEG signal, while the quantum part utilizes variational quantum circuit (VQC) for feature mapping and classification. We evaluated the model on different datasets. The results show that the model is satisfactory for classification. This indicates that the combination of quantum computing and machine learning has great potential and provides new research directions and technical approaches for developing more efficient computational models and dealing with complex data problems.
Quantum machine learning (QML) is a relatively new discipline emerging from the concepts of machine learning and quantum computing, whereby quantum algorithms are used to solve machine learning tasks. This paper explores the use of quantum machine learning for electroencephalogram (EEG) classification. In particular, a previously proposed EEG feature extraction method and classification framework for classifying dementia subjects were followed in this study. A quantum classifier replaced the classical classifier component of the framework, and the classification accuracies between the quantum and classical classifiers were compared. This study has demonstrated that applying QML in healthy-dementia classification can be implemented using near-term quantum devices or quantum simulators with moderate performance. The quantum classifier achieved an overall classification accuracy of 81.67% and 79.17% in a train-test split performance test and an n × k-fold cross-validation test, respectively. However, the quantum approach did not produce higher classification accuracies than the classical classifier. Despite the promise of quantum advantages, further investigation and optimization are required to improve its effectiveness.