As the proposed method can reduce the number of features and select the best feature set, its classification performance was improved and the classification time was shortened; thus, it can be applied to various BCI systems.
Objective. Brain computer interface (BCI) system allows humans to control external devices through motor imagery (MI) signals. However, many existing feature extraction algorithms cannot eliminate the influence of individual differences. This research proposed a new processing algorithm that can reduce the impact of individual differences on classification and improve the universality of the algorithm. Approach. To select the optimal frequency band, the energy in each sub-band was calculated by the discrete wavelet transform. Power spectral density and visual geometric group network based convolutional neural network were used for feature extraction and classification respectively. Main results. The test of the BCI Competition IV dataset IIa proved the superiority of the algorithm. In comparison with some commonly used methods, the proposed algorithm reduced classification calculation time while improving classification accuracy; the average classification accuracy rate reaches 96.21%, which is far exceeding the results obtained by the latest literature. Significance. The good classification performance of this research was rooted in the reduced number of parameters, the reduced consumption of computing resources, and the eliminated influence of individual differences. Therefore, the proposed algorithm can be applied to a real-time multi-class BCI system.
Brain-computer interface (BCI) is a system for communication and control between the human brain and computers or other electronic devices. However, multichannel signal acquisition, which is timeconsuming, laborious, and not conducive to subsequent real-time signal processing, can cause channel redundancy. When designing a BCI system, the selection of the optimal channels that match the expected pattern of potential cortical activity is useful for classifying brain activity during a mental task. The Stockwell transform and Bayesian linear discriminant analysis were applied to feature extraction and classification, respectively, and a genetic algorithm (GA) was used in the process of channel selection to extract the most relevant channel for classification. The superior performance of the algorithm is demonstrated by the test results on BCI Competition III dataset I. For a motor imagery paradigm, we show that the number of used channels can be reduced significantly, and the classification performance of the classifier can be improved. By comparing the performances of the algorithm with or without channel selection, the best channel combination was selected, and only 28 out of the 64 acquisition electrodes were used to realize classification sensitivity, specificity, precision, accuracy and Kappa coefficient values of 98%, 96%, 96.08%, 97% and 0.94, respectively, thereby outperforming existing algorithms. The proposed algorithm can reduce the number of channels and select the best channel combination to improve the classification performance, thereby overcoming the redundant channel problem. In addition, the signal processing framework that is adopted by this research can serve as a reference for related BCI application system research.
Deep learning techniques have been applied to certain rigid or non-rigid medical image registration due to its potential advantages in meeting the clinical requirements of real-time and accuracy. Based on the deep learning model, this study aims to explore specific network models suitable for lung CT images. The proposed model took unlabeled 3D image pairs as input, and the convolutional neural network (CNN) was utilized and identified as a function with ability of sharing parameters to obtain displacement field. The image pair could be aligned by applying the acquired displacement field to the target image through spatial transformation. The similarity between the aligned image pair combined with the constraints on the displacement field was taken as the objective function to obtain the optimal parameters. Two models with different depths were designed and the consequent registration effects with different optimization methods and convolution kernel sizes were explored. The results proved that the designs with deeper level using Adam optimizer and smaller convolution kernels in obtaining displacement fields had higher accuracy and stronger robustness. The accuracy of the unsupervised model was comparable to state-of-theart methods, while operating orders of magnitude faster. This study proposed a feasible registration method for lung 3D-CT, and its usefulness in aligning CT images has been demonstrated.
With the rapid development of the Internet of things and e-commerce, feature-based image retrieval and classification have become a serious challenge for shoppers searching websites for relevant product information. The last decade has witnessed great interest in research on content-based feature extraction techniques. Moreover, semantic attributes cannot fully express the rich image information. This paper designs and trains a deep convolutional neural network that the convolution kernel size and the order of network connection are based on the high efficiency of the filter capacity and coverage. To solve the problem of long training time and high resource share of deep convolutional neural network, this paper designed a shallow convolutional neural network to achieve the similar classification accuracy. The deep and shallow convolutional neural networks have data pre-processing, feature extraction and softmax classification. To evaluate the classification performance of the network, experiments were conducted using a public database Caltech256 and a homemade product image database containing 15 species of garment and 5 species of shoes on a total of 20,000 color images from shopping websites. Compared with the classification accuracy of combining content-based feature extraction techniques with traditional support vector machine techniques from 76.3% to 86.2%, the deep convolutional neural network obtains an impressive state-of-the-art classification accuracy of 92.1%, and the shallow convolutional neural network reached a classification accuracy of 90.6%. Moreover, the proposed convolutional neural networks can be integrated and implemented in other colour image database.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.