ABSTRACT
Artificial intelligence not only has massive data storage capacity and powerful data analysis capability, but also can simulate human consciousness and thinking patterns using language recognition, image recognition, natural language processing and other technologies. At present, artificial intelligence has been integrated into every field of society, especially played an important role in the healthcare field of China. In this paper, the relevant literatures published in recent years about artificial intelligence in the healthcare field of China were reviewed and sorted, and the hotspots of artificial intelligence research were analyzed. The main application scenarios of artificial intelligence in this field were investigated and analyzed by classification and statistics. Finally, some challenges faced by artificial intelligence in promoting the healthcare of China were discussed and corresponding suggestions were proposed.
ABSTRACT
Objective To design and implement a universal multi-channel software for neural electrophysiological stimulation experimental platforms. Method The layered design of software and hardware was adopted for the logical architecture to avoid excessive reliance on specific hardware. On the premise of ensuring compatibility with existing devices, an extensible control algorithm based on the .NET Frameworks platform was developed to realize multi-channel, feedback-controlled program-controlled stimulus output. The proposed software was designed with a user-friendly interface and stimulating/recording switch function, and could dynamically change stimulation programs and switch electrodes during the experiment process. Results The results showed that the software could control the stimulators steadily and generate random stimulation protocols and synchronization control signals according to the user-supplied dynamical parameters, including electrodes, amplitudes, and intervals. In the stimulation sequence, the switching delay between two electrodes was around 600 ms level. Conclusion The software has good compatibility with existing equipment systems. It can achieve multi-channel, real-time, feedback-controlled program-controlled stimulation according to the characteristics and needs of multi-lead neural electrophysiological stimulation researches. It has the functions of dynamically changing the stimulation program and switching electrodes during operation. This software provides tools for the study of the mechanism of network-level neural network feedback loops.
ABSTRACT
Objective To investigate the characteristics of brain network based on brain electrical activity induced by somatosensory electrical stimulation,and to provide a theoretical basis for further understanding the mechanism of brain neural plasticity induced by somatosensory electrical stimulation.Methods Ten healthy subjects were selected and a somatosensory electrical stimulation experiment was constructed based on the directed transfer function (DTF).In the experiment,the DTF causal connection matrixes of the 32-channel EEG data of Delta,Theta,Alpha and Beta bands were obtained under the somatosensory electrical target and non-target stimulation,and the differences of clustering coefficient and global efficiency between two stimulation states were contrasted based on graph theory.Results Under the target stimulation and non-target stimulation states,the regions with stronger DTA causal connections were mainly concentrated in FCz,Cz,CPz and Pz channels.The causal connection intensity under target stimulation state was greater than that of non-target stimulation.Also,in the Delta,Theta,and Alpha bands,the clustering coefficient under the target stimulation state was significantly higher than that in the non-target stimulation state (P＜0.05).In the Delta and Theta bands,the global efficiency of the target stimulation state was significantly higher than that of the non-target stimulation state (P＜0.05).Conclusions Somatosensory electrical stimulation can activate and induce EEG brain networks.In the target stimulation state,the role of the parietal lobe in the EEG causal network is enhanced,which helps to induce attention to specific brain region plasticity,and thus realizing the nerve rehabilitation in the brain regions of interest.While in the non-target stimulation state,the synergistic interactions between brain regions were enhanced,which helps to activate and induce a wide range of associations in the whole brain network,so as to promote the global neural activity in the brain.
ABSTRACT
Objective The single-trial extraction method of evoked potential has been one of the problems in EEG information processing field.According to the characteristics of somatosensory evoked electroencephalogram (EEG) with low signal-to-noise ratio and large parameter variation between trials,a novel single-trial extraction method for evoked potentials was proposed.This method aims to further improve the accuracy and characteristics of the single-trial extraction algorithm,preserve more dynamic characteristics between trials,and improve the estimation accuracy.Methods Based on wavelet filtering and multiple linear analysis,a new single-trial extraction method for EEG P300 parameters was proposed by applying the adaptive dynamic feature library.Four groups of wavelet filtered evoked EEG data were randomly selected,and used to build the feature library using overlapping average method and principal component analysis.For the single-trial extracted EEG data,the component with the highest correlation coefficient related with the current data was selected as the independent variable from the feature library,and the relevant multiple linear regression analysis was conducted.The single-trial evoked potential signal was reconstructed by the regression analysis results,in which the key features such as latency and amplitude were automatically extracted.Results Compared with the benchmark values determined by experts,the proposed algorithn can obtain more accurate estimation values of latency and amplitude in P300 components.The average difference of latency and amplitude by the proposed algorithm is (11.16±8.60) ms and (1.40±1.34)μV,respectively.These two values obtained by the proposed algorithm are much closer to that obtained by the commonly used overlapping average method of (23.26±25.76) ms and (2.52±2.50) μV,respectively.These results show that the proposed algorithm has significant advantages comparing with the traditional multiple linear regression analysis algorithm.Conclusions The dynamic updating principal component sample library of EEG data was applied to wavelet filtering and multiple linear regression,thus the dynamic characteristics were effectively preserved,and the accuracy of parameter estimation was improved.
ABSTRACT
Objective The single-trial extraction method of evoked potential has been one of the problems in EEG information processing field.According to the characteristics of somatosensory evoked electroencephalogram (EEG) with low signal-to-noise ratio and large parameter variation between trials,a novel single-trial extraction method for evoked potentials was proposed.This method aims to further improve the accuracy and characteristics of the single-trial extraction algorithm,preserve more dynamic characteristics between trials,and improve the estimation accuracy.Methods Based on wavelet filtering and multiple linear analysis,a new single-trial extraction method for EEG P300 parameters was proposed by applying the adaptive dynamic feature library.Four groups of wavelet filtered evoked EEG data were randomly selected,and used to build the feature library using overlapping average method and principal component analysis.For the single-trial extracted EEG data,the component with the highest correlation coefficient related with the current data was selected as the independent variable from the feature library,and the relevant multiple linear regression analysis was conducted.The single-trial evoked potential signal was reconstructed by the regression analysis results,in which the key features such as latency and amplitude were automatically extracted.Results Compared with the benchmark values determined by experts,the proposed algorithn can obtain more accurate estimation values of latency and amplitude in P300 components.The average difference of latency and amplitude by the proposed algorithm is (11.16±8.60) ms and (1.40±1.34)μV,respectively.These two values obtained by the proposed algorithm are much closer to that obtained by the commonly used overlapping average method of (23.26±25.76) ms and (2.52±2.50) μV,respectively.These results show that the proposed algorithm has significant advantages comparing with the traditional multiple linear regression analysis algorithm.Conclusions The dynamic updating principal component sample library of EEG data was applied to wavelet filtering and multiple linear regression,thus the dynamic characteristics were effectively preserved,and the accuracy of parameter estimation was improved.