Exploring Neuroimaging-Genetic Co-Alteration Features of Auditory Verbal Hallucinations in Different Subjects for the Establishment of a Predictive Model
Auditory verbal hallucinations (AVHs) are defined as the perception of voices in the absence of any external auditory stimulus. These hallucinations are often deeply personal, causing significant distress to those who experience them. AVHs are prevalent across a wide range of mental and neurological disorders, including schizophrenia, bipolar disorder (BP), major depressive disorder (MDD), post-traumatic stress disorder (PTSD), and borderline personality disorder (BPD). They are also reported in healthy individuals, though less frequently. Schizophrenia patients are the most affected, with over 70% experiencing AVHs. In contrast, the prevalence in other disorders ranges from 46% in BPD, 11.3% to 62.8% in BP, 5.4% to 40.6% in MDD, and 4.2% in the general population.
The broad prevalence of AVHs underscores the need for early and precise diagnosis to mitigate the risks of misdiagnosis and mistreatment, which can exacerbate the condition and lead to severe side effects. AVHs are associated with increased risks of violence, suicidal behavior, and self-mutilation, highlighting the urgency for effective diagnostic and therapeutic strategies.
Current research emphasizes the importance of exploring the pathological features of AVHs using multidisciplinary approaches, combining neuroimaging and genetic techniques. Given the human-specific nature of language and speech, AVHs cannot be studied in animal models, necessitating research in living individuals. Recent advancements in neuroscience have enabled the exploration of AVH features through techniques such as functional magnetic resonance imaging (fMRI), electroencephalography (EEG), and diffusion tensor imaging (DTI). These methods have revealed abnormal activity in specific brain networks, such as the salience network, default-mode network, and language-processing networks, in patients experiencing AVHs.
Genetic studies have also identified specific genes associated with AVHs, such as FOXP2, COMT, and NRG1, which influence susceptibility and treatment response. The integration of neuroimaging and genetic data provides a comprehensive understanding of the intrinsic pathological features of AVHs, offering potential biomarkers for early diagnosis and treatment targets.
The development of predictive models using machine learning techniques is a promising approach to improving AVH diagnosis and treatment. Pattern recognition methods have been successfully applied to establish predictive models for various diseases, including schizophrenia. For instance, a study by Yin et al. achieved 73.9% specificity in AVH pattern recognition in schizophrenia patients. These findings support the use of machine learning to develop predictive models for AVHs across different mental disorders.
The proposed research pathway involves two main components. First, researchers should examine subjects with first-episode untreated schizophrenia, BP, PTSD, BPD, and MDD, as well as healthy individuals experiencing their first AVH episodes. High-throughput sequencing and human connectome techniques will be used to identify common and specific neuroimaging-genetic co-alteration features of AVHs. These features can guide the development of targeted treatment strategies tailored to the specific characteristics of each disorder.
Second, based on the acquired data, machine learning techniques will be employed to establish predictive models for early diagnosis and treatment outcomes. These models will integrate neuroimaging-genetic co-alteration features, sociodemographic characteristics, and treatment outcomes to provide a comprehensive tool for clinicians. The dynamic changes in neuroimaging-genetic features during therapy will also be examined to identify specific treatment targets and optimize therapeutic strategies.
The proposed research is novel in its hypothesis that common and specific pathological features of AVHs exist across different mental disorders and healthy individuals. These features can form the basis of a stable and accurate index for early diagnosis and treatment prediction. The “pathological feature bridge of AVHs” concept suggests that brain abnormalities serve as a bridge between genetic features and clinical manifestations, regulated by specific genes and associated with AVH symptoms. This bridge can be detected using current technologies and used as a visual biomarker reflecting the relationships among genetic, neuropathological, and clinical features.
A large cohort study with a two-year follow-up period is proposed to validate the predictive model. This study will involve the combined use of machine learning, genomic, and brain connectome methods, considering subjects’ sociodemographic characteristics and treatment outcomes. The examination of dynamic alterations in AVHs during therapy will enable the identification of specific treatment targets and the development of precise therapeutic strategies.
In conclusion, the integration of neuroimaging and genetic data, combined with machine learning techniques, offers a promising pathway for the early diagnosis and treatment of AVHs. The proposed research aims to identify common and specific neuroimaging-genetic co-alteration features of AVHs across different mental disorders and healthy individuals, establish predictive models for early diagnosis and treatment outcomes, and optimize therapeutic strategies based on identified treatment targets. This approach has the potential to significantly improve the diagnosis and treatment of AVHs, reducing the associated risks and improving the quality of life for affected individuals.
doi.org/10.1097/CM9.0000000000000385
Was this helpful?
0 / 0