Residual self-calibrated network with multi-scale channel attention for accurate EOG-based eye movement classification.

Recently, Electrooculography-based Human-Computer Interaction (EOG-HCI) technology has gained widespread attention in industrial areas, including assistive robots, augmented reality in gaming, etc. However, as the fundamental step of EOG-HCI, accurate eye movement classification (EMC) still faces a significant challenge, where their constraints in extracting discriminative features limit the performance of most existing works. To address this issue, a Residual Self-Calibrated Network with Multi-Scale Channel Attention (RSCA), focusing on efficient feature extraction and enhancement is proposed. The RSCA network first employs three self-calibrated convolution blocks within a hierarchical residual framework to fully extract the discriminative multi-scale features. Then, a multi-scale channel attention module adaptively weights the learned features to screen out the discriminative representation by aggregating the multi-scale context information along the channel dimension, thus further boosting the performance. Comprehensive experiments were performed using 5 public datasets and 7 prevailing methods for comparative validation. The results confirm that the RSCA network outperforms all other methods significantly, establishing a state-of-the-art benchmark for EOG-based EMC. Furthermore, thorough ablation analyses confirm the effectiveness of the employed modules within the RSCA network, providing valuable insights for the design of EOG-based deep models.

KL Divergence-based transfer learning for cross-subject eye movement recognition with EOG signals.

Human-machine interfaces (HMIs) based on Electro-oculogram (EOG) signals have been widely explored. However, due to the individual variability, it is still challenging for an EOG-based eye movement recognition model to achieve favorable results among cross-subjects. The classical transfer learning methods such as CORrelation Alignment (CORAL), Transfer Component Analysis (TCA), and Joint Distribution Adaptation (JDA) are mainly based on feature transformation and distribution alignment, which do not consider similarities/dissimilarities between target subject and source subjects. In this paper, the Kullback-Leibler (KL) divergence of the log-Power Spectral Density (log-PSD) features of horizontal EOG (HEOG) between the target subject and each source subject is calculated for adaptively selecting partial subjects that suppose to have similar distribution with target subject for further training. It not only consider the similarity but also reduce computational consumption. The results show that the proposed approach is superior to the baseline and classical transfer learning methods, and significantly improves the performance of target subjects who have poor performance with the primary classifiers. The best improvement of Support Vector Machines (SVM) classifier has improved by 13.1% for subject 31 compared with baseline result. The preliminary results of this study demonstrate the effectiveness of the proposed transfer framework and provide a promising tool for implementing cross-subject eye movement recognition models in real-life scenarios.

Prognostic Analysis of KCNQ2 Patients via Combining EEG Deep Features and Machine Learning Classifiers.

Pathogenic variants of the KCNQ2 gene often induces neonatal epilepsy in clinical. For better treatment, infants with confirmed KCNQ2 pathogenic variant and epilepsy symptoms need to adjust their treatment plans according to the outcome after taking antiseizure medicines (ASMs). This process is often time-consuming and requires long-term follow-up, which undoubtedly causes unnecessary psychological and economic burdens. In this study, we investigate the feasibility to predict the outcome of KCNQ2 patients via Electroencephalogram (EEG). By using the combination of deep networks and classical classifiers, the abnormal brain pathological activities recorded in EEGs can be encoded into deep features and decoded into specific KCNQ2 outcomes, thus taking the advantage of both powerful feature extraction capability from deep networks and stronger classification ability from classical classifiers. Specifically, we acquire 10-channel EEG signals from 33 infants with KCNQ2 pathogenic variants after taking ASMs. Two well-trained models (Resnet-50 and Resnet-18) are employed to extract deep features from the EEG spectrums. We achieve an accuracy of 78.7% to predict the KCNQ2 outcome of each infant. To our best knowledge, this is the first study to employ potential EEG pathological differences to predict the outcomes of KCNQ2 patients. The investigation of automatic KCNQ2 outcome prediction may contribute to a more convenient diagnosis mechanism for KCNQ2 patients.

Unsupervised Transfer Learning Approach With Adaptive Reweighting and Resampling Strategy for Inter-subject EOG-based Gaze Angle Estimation.

Gaze estimation based on electrooculograms (EOGs) has been widely explored. However, the inter-subject variability of EOGs still leaves a significant challenge for practical applications. It contributes to performance degradation when handling inter-subject issues. In this paper, an unsupervised transfer learning approach with an adaptive reweighting and resampling (ARR) strategy to fully consider individual variability is proposed for EOG-based gaze angle estimation. It allows quantifying domain shifts by leveraging the source-target similarities, reweighting and resampling the source data to retain relevant instances and disregard irrelevant instances during adaptation. Specifically, our proposed methodology first assesses the domain shifts via decomposing transformation matrices, which are estimated between the training subjects (denoted as multi-source domains) and the test subject (denoted as target domain). Then, the multi-domain shifts are assigned as weighted indicators to resample the multi-source domains for model training. Comparative experiments with several prevailing transfer learning methods including CORrelation ALignment (CORAL), Geodesic Flow Kernel (GFK), Joint Distribution Adaptation (JDA), Transfer component analysis (TCA), and Balanced distribution adaption (BDA) using two different normalization processes were conducted on a realistic scenario across 18 subjects. Experimental results demonstrate that the ARR strategy can significantly improve performance (mean absolute error (MAE) reduction: 7.0%, root mean square error (RMSE) reduction: 6.3%), outperforming the prevailing methods. Besides, the impacts of data diversity and data size on ARR strategy are further investigated. It exhibits that data size is more important than data diversity for EOG-based gaze angle estimation, and also presents the benefits of the ARR strategy for dealing with practical scenarios.

Russian-Doll-Like Porous Carbon as Anode Materials for High-Performance Potassium-Ion Hybrid Capacitors.

Pore-structure design with the sophisticated and pragmatic nanostructures still remains a great challenge. In this work, porous carbon with Russian-doll-like pores rather than traditional single modal is fabricated via a boiling carbonization approach, accompanied by K+ -pre-intercalation. The most important internal factor is that alkali can penetrate into the stereoscopic space of layered Malonic acid dihydrazide and the confinement effect leads to the in-depth development of different dimensional pore structures. The oxygenated and nitrogenated surface guarantees the K+ intercalation behavior. Benefiting from their open framework and enlarged interlayer spacing, K+ -pre-intercalated porous carbon with Russian-doll-like pores (denoted as KPCRPs) as anode material exhibits promising potassium storage performance. The assembled KPCRP//activated carbon potassium-ion hybrid supercapacitor in 30 m CH3 COOK displays a high energy density of 157.29 Wh kg-1 , an ultrahigh power output of 14 kW kg-1 , and a long cycling life (99.58% capacity retention after 10000 cycles), highlighting the superiority of Russian-doll-like pore structure. This work sheds light on the designing of 3D pores structure, especially for multimodal pore architectures.