Fuquan Zhang

Driver drowsiness electroencephalogram (EEG) signal monitoring can promptly alert drivers of their drowsiness in edge-end Consumer Electronics (CE), thereby reducing the probability of traffic accidents. Graph convolutional networks (GCNs) have shown significant advancements in processing the non-stationary, time-varying, and non-Euclidean nature of EEG signals. However, the existing single-channel EEG adjacency graph construction process lacks interpretability, which hinders the ability of GCNs to effectively extract adjacency graph features, thus affecting the performance of drowsiness monitoring. To address this issue, we propose an edge-end Lightweight Dual Graph Convolutional Network (LDGCN). Specifically, we are the first to incorporate neurophysiological knowledge to design a Baseline Drowsiness Status Adjacency Graph (BDSAG), which characterizes driver drowsiness status. Additionally, to express more features within limited EEG data, we introduce the Augmented Graph-level Module (AGM). This module captures global and local information at the graph level, ensuring that BDSAG features remain intact while enhancing effective feature expression capability. To further enhance the application of CE, we apply Adaptive Pruning Optimization (APO) to the channels and neurons of LDGCN, which nearly halves the inference latency. Experiments on benchmark datasets demonstrate that LDGCN offers the best trade-off between monitoring performance and hardware resource utilization, outperforming existing state-of-the-art algorithms.

This study aims to leverage the advanced capabilities of quantum computing to construct an efficient framework for processing large-scale health data, uncover potential higher-order correlations in medicine, and enhance the accuracy of smart healthcare diagnosis and treatment. A data processing framework is developed using quantum annealing algorithms and quantum circuits. We call it the quantum medical data simulation computational model (Q-MDSC). A unique encoding method based on quantum bits is employed for health data features, such as encoding symptom information from electronic health records into different quantum bits and representing different alleles of genetic data through superposition states of quantum bits. The properties of quantum entanglement are utilized to relate different data types, and quantum parallelism is harnessed to process multiple data combinations simultaneously. Additionally, this quantum computing framework is compared with traditional data mining methods using the same datasets, which include the Cochrane Systematic Review Database (https://www.cochranelibrary.com), the BioASQ Dataset (https://participants-area.bioasq.org), the PubMed Central Dataset (https://www.ncbi.nlm.nih.gov/pmc), and the Cancer Genome Atlas (TCGA) (https://portal.gdc.cancer.gov). The datasets are divided into training and testing sets in a 7:3 ratio during the experiments. Tests are conducted on association mining tasks of varying data scales and complexities, ranging from simple symptom-disease associations to complex gene-symptom-disease higher-order associations. The results indicate that, when processing large-scale data, the quantum computing framework improves overall computational speed by approximately 45% compared to traditional algorithms. Regarding uncovering higher-order correlations, the quantum computing framework enhances accuracy by about 30% relative to traditional algorithms. For early disease prediction, the accuracy achieved with the new framework is approximately 25% higher than that of conventional methods. Furthermore, for personalized treatment plan matching, the matching accuracy of the quantum computing framework surpasses traditional approaches by about 35%. These findings demonstrate the significant potential of the quantum computing-based smart healthcare framework for processing large-scale health data in the context of higher-order correlation mining, paving new pathways for the development of smart healthcare. This study utilizes multiple public datasets to achieve breakthroughs in computational speed, higher-order correlation mining, early disease prediction, and personalized treatment plan matching, thus opening new avenues for advancing smart healthcare.

In this paper, a novel deep learning method Mamba-RAYOLO is presented, which can improve detection and classification in the processing and analysis of ECG images in real time by integrating three advanced modules. The feature extraction module in our work with a multi-branch structure during training can capture a wide range of features to ensure efficient inference and rich feature extraction. The attention mechanism module utilized in our proposed network can dynamically focus on the most relevant spatial and channel-wise features to improve detection accuracy and computational efficiency. Then, the extracted features can be refined for efficient spatial feature processing and robust feature fusion. Several sets of experiments have been carried out to test the validity of the proposed Mamba-RAYOLO and these indicate that our method has made significant improvements in the detection and classification of ECG images. The research offers a promising framework for more accurate and efficient medical ECG diagnostics.

Driver drowsiness electroencephalography (EEG) signal monitoring can timely alert drivers of their drowsiness status, thereby reducing the probability of traffic accidents. Graph convolutional networks (GCNs) have shown significant advancements in processing the non-stationary, time-varying, and non-Euclidean nature of EEG signals. However, the existing single-channel EEG adjacency graph construction process lacks interpretability, which hinders the ability of GCNs to effectively extract adjacency graph features, thus affecting the performance of drowsiness monitoring. To address this issue, we propose an edge-end lightweight dual graph convolutional network (LDGCN). Specifically, we are the first to incorporate neurophysiological knowledge to design a Baseline Drowsiness Status Adjacency Graph (BDSAG), which characterizes driver drowsiness status. Additionally, to express more features within limited EEG data, we introduce the Augmented Graph-level Module (AGM). This module captures global and local information at the graph level, ensuring that BDSAG features remain intact while enhancing effective feature expression capability. Furthermore, to deploy our method on the fourth-generation Raspberry Pi, we utilize Adaptive Pruning Optimization (APO) on both channels and neurons, reducing inference latency by almost half. Experiments on benchmark datasets demonstrate that LDGCN offers the best trade-off between monitoring performance and hardware resource utilization compared to existing state-of-the-art algorithms. All our source code can be found at https://github.com/BryantDom/Driver-Drowsiness-Monitoring.