Coupling Method Of Multiple Brain Function Networks And Evaluation Studies For Heroin Addicts

The realization of cognitive functions in the brain involves massively parallel processing of distributed neural circuits.Long-term heroin intake causes impaired interaction and synergy between multiple neural circuits within the brain,manifesting as abnormal brain function.Exploring the abnormal brain function of addicts and the changes in brain networks under intervention and treatment is important for the treatment and recovery of heroin addiction.EEG is suitable for studying spatial patterns of the brain on millisecond time scales due to its high temporal resolution.Previous studies have shown that the temporal and spatial properties of EEG reflect the temporal dynamics and spatial organization,and their relationship with cognitive functions,thus allowing further understanding of the working patterns of the brain.However,existing research suffers from the following problems:(1)current research on the neural mechanisms of impairment in heroin addicts is limited to aspects of drug cues and ignores the research on reward mechanisms;(2)a large number of studies have focused on static interaction analyses between independent brain regions,lacking large scale dynamic network research methods based on the temporal and frequency dimensions;(3)the effectiveness of addiction withdrawal methods is still unclear,as well as the longitudinal feedback analysis and the addiction detection assessment model during the intervention treatment are lacking.Based on the above four problems,this thesis extracts the time-domain and frequency-domain information of EEG signals from heroin addicts and healthy controls to study the brain dysfunction of addicted patients,and proposes a multiple functional network coupling method to systematically analyse the dynamic interaction processes of large scale brain functional networks.Using the proposed method,this thesis also examines the treatment effects of forced withdrawal on heroin addicts and designs an assessment model of addiction treatment effects based on it.The main work and innovations are summarised as follows.1.In view of the lack of brain function mechanism of reward in heroin addicts,a spatial-based attention capture paradigm was proposed,followed by extracting the attentional components and transient frequency characteristics of task-state EEG time-locked by time-locked event-related potential analysis and wavelet transform methods.The ERP(Event-Related Potential)performance of fast and slow response was separated by introducing behavioral response time into event-related potential,so as to explore the effects of non-drug reward-related stimuli on attention patterns in heroin addicts.The results found that integrating salient reward stimuli with target stimuli facilitated the completion of a visual search task.When salient reward stimuli acted as distractors,heroin addicts allocated more attention to the reward stimuli and delayed target orientation.Compared to healthy controls,heroin addicts were significantly less able to inhibit attention to reward-related stimuli.Furthermore,the impact of reward production increased with the value of the reward.These results reveal the underlying neural mechanisms of value-driven attention allocation in the face of monetary rewards in heroin addicts,remedying the lack of reward-related brain function mechanism on heroin addiction.2.In view of the lack of large-scale dynamic network research based on the time dimension,a multiple functional network coupling method(MFNCA)for task-state EEG was proposed.This method was based on functional subnetworks rather than a single electrode or brain region,and divided multiple cognitive processes according to the ERP temporal characteristics.For the first time,we constructed a multiple functional network based on cognitive processes,and examine the cross-coupling properties and dynamic changes over time and space between multiple functional networks.In the attention capture task of heroin addicts,the method used sub-layer networks composed of multiple brain regions as nodes,and combined with typical correlation analysis methods to calculate the relationships between networks.The findings revealed that attentional and frontoparietal networks are important hubs for attention deployment.In addition,this work identifies deficits in the right frontoparietal network(RFPN)thalamic network(THN),anterior cingulate network(ACN)and basal ganglia network(BGN)during value-driven attention allocation in heroin addicts.The results suggest that the proposed method is useful for studying the dynamic interactive coupling between brain networks induced by external stimuli and can also serve as a useful tool for assessing abnormal brain function in heroin addicts.3.In view of the lack of longitudinal feedback analysis in addiction treatment,a multifunctional network based on cross-frequency coupling was constructed to examine the intra-and inter-frequency multifunctional network connectivity in heroin addicts before and after forced abstinence in the resting state.We used seven significant subnetworks as nodes to construct multiple functional networks for the addiction,withdrawal and control groups based on each of the five frequency bands.It was found that the most active functional network connections were in the α,β1 and γfrequency bands;the strength of functional connections related to the right frontoparietal network(RFPN)and insula network(ISN1)changed the most in heroin addicts after six months of forced withdrawal.These findings reflect the recovery of impaired brain function in addicts after forced withdrawal.According to the above results,we found that the prefrontal cortex is involved in inhibitory control in cognitive function and undergoes greater changes under withdrawal treatment.Therefore,a model for detecting addiction and assessing addiction treatment based on the prefrontal three-lead EEG was designed.By measuring prefrontal three-lead EEG,addicts and controls can be effectively identified with an accuracy rate of 91.7%.In summary,this thesis explored the neural mechanisms of heroin addiction and effectively extracted aggregation and segregation information between multiple networks by constructing multiple functional networks based on functional subnetworks.The findings reveal the dysfunction of mutual drive and synergy among multiple brain regions caused by addiction,helping to understand the brain functional deficits in heroin addicts and providing new ideas for withdrawal treatment.The multiple functional network coupling approach effectively quantifies dynamic development of functional network connectivity on a temporal scale,and provides theoretical guidance for the construction of large-scale brain functional networks,which has important implications for exploring the impact of cognitive impairment and neurological disorders on brain function.These works reveal the abnormal neural mechanism of heroin addiction,and realize real-time assessment of treatment effects and feedback interventions for addicts,and contribute to the development and implementation of personalised treatment programs in clinical settings.