How gut microbiota contribute to neuropsychiatric disorders: evidence from neuroimaging studies

Gut microbiota produce various bioactive molecules through metabolism, particularly SCFAs and tryptophan metabolites, which influence brain structure and function, primarily by regulating BBB permeability and neuroinflammatory responses. SCFAs, including acetate, propionate, and butyrate, are crucial for maintaining intestinal barrier integrity, regulating immune responses, and supporting CNS homeostasis (Ravi et al., 2025). They also enhance BBB structural integrity, promote the expression of tight junction proteins, restricting peripheral inflammatory signals from entering the CNS (Knox et al., 2022). Additionally, SCFAs exert neuroprotective effects through inhibiting histone deacetylase activity and regulating gene expression in neurons and glial cells (Ravi et al., 2025). In neurodegenerative diseases like Parkinson’s disease (PD), abnormal SCFA levels correlate with disease progression, suggesting their potential therapeutic value (Ran et al., 2025; Ravi et al., 2025). Tryptophan metabolites also participate in regulating brain function. Specifically, indole compounds mediated by gut microbiota, such as indole-3-acetic acid (IAA), can cross the BBB, modulate neurotransmitter systems and neuroendocrine signaling pathways, thereby affecting mood and cognitive performance (Chernevskaya et al., 2021; Yu et al., 2024). Furthermore, abnormalities in tryptophan metabolic pathways involve in the pathology of various neuropsychiatric disorders, including depression and autism spectrum disorder (Więdłocha et al., 2021; Wiley et al., 2021). Many factors like certain food components (e.g., artificial sweeteners) can influence this physiological process, which characterized by a reduction in beneficial bacteria and an increase in pathogenic bacteria, thereby affecting the production of SCFAs and ultimately impacting BBB function, neuroinflammatory responses, and brain function (Hetta et al., 2025b).

Multi-omics and neuroimaging studies further confirm the close associations between microbial metabolites and FC in cognitive and emotional networks. Structural changes in gut microbiota and their metabolites correlate with the pathogenesis of various conditions, including cognitive decline and dementia, influencing gray matter volume and white matter integrity (Alloo et al., 2022; Hsu et al., 2025; Liang et al., 2022). Specifically, the abundance of SCFAs and tryptophan metabolites positively correlates with hippocampal volume and prefrontal cortex FC (Dong et al., 2020; Liang et al., 2022). Furthermore, 7T MRI results revealed that SCFA supplementation significantly alleviated hippocampal atrophy, and correlated with increases in beneficial gut microbiota such as Alloprevotella (Lee P. J. et al., 2025). Clinical intervention studies further supports this regulatory role, and they found that modulation of gut microbiota in obese patients improved connectivity in cognitive-related brain networks (Dong et al., 2020; Xiang et al., 2024). Figure 1 summarizes key neuroimaging findings linking microbial metabolites (e.g., SCFAs and tryptophan derivatives) to alterations in BBB integrity, neuroinflammation, and brain structure and function across CNS disorders.

Gut microbiota influence microglial activation and neuroinflammation in the CNS by regulating peripheral immune responses, leading to brain structural and functional changes. Multiple studies indicate that peripheral inflammatory signals can cross the BBB and alter the CNS immune milieu, driving microglial activation and neuroinflammatory cascades (Jung et al., 2022). Microbial metabolites and immunomodulatory effects similarly affect neuroinflammation and regulate neuronal survival and synaptic function.

Brain imaging provides quantitative markers of these immune-related effects. MRI DTI shows that peripheral inflammatory status correlates with microstructural white matter abnormalities, such as increased diffusivity and altered axial diffusivity, reflecting axonal damage and demyelination (Lan et al., 2024). In PET imaging, specific radioligands for neuroinflammation markers, such as translocator protein (TSPO, e.g., [18F]GE-180 and [11C]CPPC) and GPR84 ligands, can monitor microglial and myeloid cell activation (Kalita et al., 2023; Lee S. J. et al., 2025; Lucot et al., 2022).

Integration of immune markers and neuroimaging further reveals disease progression mechanisms. Peripheral inflammatory markers like C-reactive protein (CRP) and systemic immune-inflammation index (SII) closely correlate with reduced gray and white matter volume, with brain atrophy partially mediating the effect of systemic inflammation on cognitive decline (Jung et al., 2022; Zhuo et al., 2023). Cytokines, such as egulated upon activation normal T cell expressed and secreted (RANTES), Hepatocyte growth factor (HGF) and IL-13, serve as mediators of the gut-brain axis, influencing the structural connectivity between brain regions (Guo et al., 2025). Simultaneously, abnormal brain FC show negative correlation with peripheral inflammatory markers, suggesting the key role of immune dysregulation in CNS network impairment (Aruldass et al., 2021). In psychiatric disorders such as schizophrenia and depression, immune-inflammatory subtypes identified through functional imaging contribute to more refined disease stratification (Bishop et al., 2022; Day et al., 2021).

Furthermore, meningeal inflammation, an important site for CNS immune surveillance, reveals strong spatial transcriptomics links to inflammatory gene expression patterns in adjacent brain parenchyma, highlighting its influence on local pathological status (Gadani et al., 2024). Advanced imaging technologies, including two-photon microscopy and immune cell-specific nanoprobes, allow dynamic, high-resolution observation of brain immune cells and real-time monitoring of neuroinflammatory processes (Byun et al., 2022; Wang X. et al., 2023). Figure 2 provides a comprehensive insight into how microbial metabolites and immune-inflammatory signaling interact and influence neuroimaging phenotypes.

Alzheimer’s disease (AD), the most common neurodegenerative disorder, is characterized by β-amyloid (Aβ) accumulation, tau hyperphosphorylation, and progressive cognitive decline (Ivanova et al., 2025; Pluta et al., 2020). Recently, increasing evidence suggests that gut microbiota contribute to AD pathology by inflammatory, metabolic, and neuroimmune mechanisms (Borsom et al., 2020; Sheng et al., 2023). Combining neuroimaging evidence with microbiome data provides a new perspective for revealing these interactions.

Previous studies show that gut dysbiosis is closely associated with brain structural atrophy and metabolic abnormalities in AD patients. APP/PS1 transgenic mice exhibit cognitive impairment accompanied by peripheral organ metabolic disturbances, significant changes in fecal metabolites, and marked alterations in gut microbiota composition, suggesting a correlation between gut microbiota, peripheral metabolic status, and brain pathology (Li et al., 2025). Furthermore, multimodal MRI and diffusion kurtosis imaging (DKI) demonstrate that microstructural changes and elevated neuroinflammation markers co-occurred with significantly reduced gut microbiota diversity and lower SCFA content, further linking abnormal gut microbial metabolic function to AD progression (Palanivelu et al., 2025). Human studies also observed consistent association that gray and white matter damage in AD and mild cognitive impairment (MCI) is negatively correlated with the abundance of butyrate-producing genera (e.g., Anaerostipes, Ruminococcus) (Yamashiro et al., 2024). And lower Bacteroidetes abundance was associated with greater Aβ deposition as confirmed by PET results (Kojima et al., 2025).

Integration of radiomics and microbiomics reveals the regulatory mechanisms of microbial metabolites on brain function. Combining 16S rRNA sequencing, serum and fecal metabolomics, and PET/MRI imaging found that altered microbial pathways related to energy metabolism and inflammation affect brain regional metabolic status and cognitive function by metabolites crossing the BBB (Sheng et al., 2023; Zhao H. et al., 2025). Glycodeoxycholic acid is associated with cognitive impairment, with MRI white matter hyperintensities burden modifying this association (Kijpaisalratana et al., 2025). Abnormalities in bile acid metabolites parallel PET-measured Aβ and tau dynamics (Nabizadeh et al., 2024). Time-restricted feeding improves cognitive deficits and reduced neuroinflammation by enriching propionate, a metabolite derived from Bifidobacterium pseudolongum, and PET imaging confirmed that this propionate crosses the BBB (Zhao Y. et al., 2025).

Microbiota–brain structural associations have also been reported. Abundance of genera like Bacteroides and Ruminococcus correlates with hippocampal volume and cognitive performance, while Akkermansia muciniphila shows inverse correlations with brain amyloid, suggesting a potential protective role (Fan et al., 2025; Fan et al., 2023). Decreased Ruminococcus was also associated with impaired white matter integrity in the basal ganglia (Hung et al., 2023), while higher Odoribacter was associated with larger white-matter and hippocampal volume, and lower cerebrospinal fluid (CSF) volume (Xinxiu Liang et al., 2022). Furthermore, free water (FW) imaging revealed that increased FW content inversely correlated with butyrate-producing bacteria (Yamashiro et al., 2024). And gut-derived LPS-containing can cross the BBB, activate the Piezo1 channel on microglia, induce excessive synaptic pruning, and promote early AD pathology progression (Zhao X. et al., 2025).

Multiple sclerosis (MS) is a chronic degenerative disorder characterized by immune-mediated axonal damage and demyelination. Recent studies show significantly reduced gut microbiota diversity and altered taxa in MS patients, suggesting that gut dysbiosis is closely implicated in MS pathogenesis (Cox et al., 2021; Moles et al., 2021). MRI has become essential for the diagnosis and monitoring MS, especially in identifying white matter lesions and demyelinated areas, which correlates strongly with gut microbiota alterations. Significant microbial changes accompany disease progression in both experimental autoimmune encephalomyelitis (EAE) and cuprizone (CPZ) mouse models. Specifically, reduced SCFA-producing bacteria in MS patients correspond to increased MRI-visible lesion load and cerebral atrophy (Cox et al., 2021; Moles et al., 2021; Schwerdtfeger et al., 2025).

Gut microbiota modulate MS pathology by regulating the host immune system. SCFAs can modulate T cell differentiation, promoting the generation of regulatory T cells (Tregs) while suppressing pro-inflammatory Th17 cells (Miyake and Noto, 2021; Yousefi et al., 2022). Studies indicate that gut dysbiosis leads to impaired intestinal barrier function (“leaky gut”), allowing bacterial products like LPS to enter systemic circulation, activating immune responses in the CNS, and promoting inflammation and demyelination (Fan et al., 2024; Parodi and Kerlero de Rosbo, 2021). Furthermore, immune cells like IgA+ B cells specifically recognizing MS-associated gut microbiota can migrate to the CNS and correlate with active inflammatory activity (Pröbstel et al., 2020). Microbiota-modulating therapies, such as probiotics, dietary adjustments, and fecal microbiota transplantation (FMT), show potential to alleviate MS symptoms, partly by regulating immune function and metabolite production (Li et al., 2024; Yousefi et al., 2022).

Traumatic brain injury (TBI) is a common CNS injury involving multi-system interactions, while gut dysbiosis is increasingly implicated as an important mediator of neurological dysfunction (Rice et al., 2019). The MGBA’s neural, endocrine, and immune pathways play a critical role in TBI recovery and prognosis.

First, post-TBI gut dysbiosis is closely related to brain structural and functional impairments. TBI-induced white matter injury (WMI) involves oligodendrocyte lineage (OLCs). Gut microbiota influence OLC differentiation and proliferation by regulating peripheral immune cell infiltration. Notably, T-cell-deficient restored OLC proliferation and remyelination despite microbiota depletion, suggesting a key role for T cell–dependent microbiota–OLC regulatory axis (Shumilov et al., 2024). MRI findings of brain volume loss and decreased white matter connectivity further align with microbiome-driven structural remodeling (Davis et al., 2022; Davis Iv et al., 2025). Alterations in gut microbiota, specifically a decrease in Firmicutes and an increase in certain bacterial families within the Bacteroidetes and Proteobacteria phyla, Dysbiosis emerged rapidly, as early as 2 h, after CNS trauma with decreased Firmicutes and increased Bacteroidetes and Proteobacteria families, and predicted lesion volume and the severity of motor impairment (Kigerl et al., 2018; Nicholson et al., 2019).

Second, the gut-brain axis affects TBI recovery and prognosis through neural, endocrine, and immune pathways. Dysbiosis impairs gut barrier function and increases intestinal permeability, promoting the systemic spread of inflammatory factors that activate microglia and astrocytes in the CNS, thereby exacerbating neuroinflammation and neural damage (Oyovwi and Udi, 2025; You et al., 2025). Gut microbial metabolites like SCFAs regulate immune cell function and influence neural repair processes (Davis Iv et al., 2025). The neuroendocrine system, such as the growth hormone (GH) and insulin-like growth factor 1 (IGF-1) axis, is disrupted in TBI, linking gut dysfunction to impaired neuroregeneration (Venkatasamy et al., 2025; Yuen et al., 2020). Additionally, microbiome-regulated neuroimmune pathways (e.g., TLR4/MyD88/NF-κB, NLRP3 inflammasome) contribute to inflammatory and repair responses (Cai et al., 2025; Dong et al., 2024a).

Finally, neuroimaging, including high-resolution MRI and DTI, is essential for monitoring brain-gut-microbiome dynamics. Combined imaging with gut microbiota analysis and metabolomics reveals the relationship between the gut microbiome and brain functional impairment (Davis et al., 2022; Davis Iv et al., 2025). Single-cell RNA sequencing showed that SCFAs can modulate the microglial transcriptional profile, reduce neuroinflammation, and promote neuroprotection (Davis Iv et al., 2025). Furthermore, Fecal Microbiota Transplantation (FMT) has been shown to improve neurological deficits and structural damage after TBI, and reduce inflammation, suggesting therapeutic potential of microbiota modulation (Davis et al., 2022; Dong et al., 2024b).

Schizophrenia (SCZ) is a complicated and severe psychiatric disorder characterized by diverse clinical symptoms and cognitive impairment. Accumulating evidence has linked gut microbiota dysbiosis to structural and functional abnormalities in SCZ patients.

Neuroimaging studies reported that brain structure and FC are weakened in SCZ. Resting-state fMRI (rs-fMRI) studies found reduced intrinsic network connectivity correlated with changes in the abundance of specific gut bacterial genera (Peng et al., 2025). Similarly, structural MRI (sMRI) showed reduced gray matter volume in certain brain regions, and these regional abnormalities were linked to gut microbial composition (Li et al., 2021; Ma et al., 2020). A recent study has linked specific microbial signatures to region-specific structural changes. Specifically, increased Eubacterium oxidoreducens was associated with enlargement of the left lateral ventricle, while elevated abundances of Gordonibacter, Family XIII, and Parabacteroides were linked to volumetric alterations in the occipital and parietal, and increased SCZ risk (Ye et al., 2025). These findings suggested gut dysbiosis involved in SCZ pathology by modulating structural connectivity within vulnerable brain circuits.

Mechanistically, gut microbiota influence SCZ-related brain development and neuroplasticity via core MGBA pathways. A key SCZ feature, microglial dysfunction (Ben-Azu et al., 2023), is regulated by microbial metabolites and peripheral inflammation, with downstream effects on synaptic plasticity and neural network stability (Wu et al., 2025b; Wu et al., 2024).

Integrated imaging–microbiome analyses provide key insight into the regulatory role of microbes on brain network and clinical symptoms. Brain-gut-microbiota networks (BGMN) studies found that specific genera, like Faecalibacterium and Collinsella, were closely related to FC in the visual system, default mode network, and subcortical structures. These connectivity changes significantly correlated with cognitive performance and symptom severity in SCZ (Peng et al., 2025). Additionally, dysfunction of the brain’s glymphatic system (assessed via DTI-ALPS index) positively correlated with decreased gut microbial diversity and cognitive decline, suggesting gut microbiota may regulate neuroplasticity and brain reserve by affecting metabolic waste clearance (Wu et al., 2025b). Multi-omics integration analysis further supports a mechanistic role for microbial metabolites, as disturbances in GABA and tryptophan metabolism, were closely related to impaired brain FC and clinical symptoms (Wang Z. et al., 2024). The correlation between reduced microbial diversity, brain volume loss, and cognitive deficits in chronic medicated patients highlights the potential of microbiota-targeted therapeutic strategies to improve brain function and cognitive outcomes in SCZ (Ma et al., 2020; Wu et al., 2025a).

Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder characterized by social communication deficits, language impairments, and stereotyped repetitive behaviors. The pathological mechanisms of ASD are closely associated with the gut microbial community. Its core dysbiotic features include altered microbial composition, increased intestinal permeability, and enhanced inflammatory responses, all of which exhibit significant associations with neurodevelopmental abnormalities and behavioral symptoms in individuals with ASD (Hetta et al., 2025a).

Neuroimaging studies demonstrate that ASD patients exhibit widespread white matter abnormalities. Study based on DTI show uneven white matter development, initial over-myelination followed by reduced myelination with age, leading to impaired brain network function. Decreased axonal bundle integrity between key brain regions, such as the prefrontal cortex, cingulate gyrus, and amygdala, affecting interregional information transfer and consequently leading to social, cognitive, and behavioral deficits (Canada et al., 2025).

Genetic–microbiome–imaging evidence further supports a causal role for gut microbiota in ASD-related brain changes. A Mendelian randomization study reported that increased abundance of Lachnoclostridium predicts greater white matter volume in the left cerebellum, while increased Fusicatenibacter was associated with altered T2* signals in the right putamen, both changes linked to increased ASD risk. Further analysis revealed that the effect of gut microbiota on ASD appears to be largely mediated through structural changes in the brain (Ye et al., 2025). Additionally, microbial metabolites, such as tryptophan, have also been observed to be significantly associated with structural alterations and symptom severity (Aziz-Zadeh et al., 2025).

Experimental models further validate the regulatory role of gut microbiota in neural development. In mouse models, changes in gut microbiota composition were closely related to alterations in white matter fiber tracts on DTI. For instance, in a tuberous sclerosis mouse model, dietary modulation of gut microbiota significantly improved white matter structure and social behavior, suggesting gut microbiota influence brain connectivity by regulating neural myelination formation and repair (Hsieh et al., 2024). Zebrafish, as a convenient model for neuroimmune research, due to their transparent embryos and highly conserved enteric nervous system (ENS) structure, provide complementary mechanistic insights. Live imaging in zebrafish models shows real-time observation of interactions between gut immune cells and neurons, highlighting the importance of gut immune signaling in shaping neurodevelopmental trajectories (Andersen-Civil et al., 2023).

Mechanistically, ASD pathology is driven by coordinated dysregulation of the gut microbiota, ENS, and immune system, with core MGBA pathways mediating effects: gut dysbiosis-induced barrier dysfunction, immune activation, and microbial metabolite (e.g., SCFAs) perturbations regulate oligodendrocyte maturation, myelin stability, and neuroinflammation, contributing to white matter and neural network abnormalities (Canada et al., 2025; Murayama et al., 2025).

Depression is a common mental condition with multifactorial pathogenesis. Accumulating evidence highlights a bidirectional relationship between gut microbiota and the CNS via the MGBA.

In late-life depression (LLD), distinct gut microbial alterations, like increased Enterobacter and Burkholderia, correlate positively with depressive symptoms and negatively with gray matter volume in key brain regions involved in memory, somatosensory, and emotion regulation, suggesting gut dysbiosis may contribute to depression by affecting brain structure integrity (Tsai et al., 2022). And the abundance of amycolatopsis sp. Hca4 modulates FC between the middle frontal gyrus and parahippocampal gyrus, thereby influencing working memory in a microbiota-dependent manner (Huang et al., 2025). Furthermore, LLD patients also exhibited reduced FC in emotion-regulation related brain regions, including the prefrontal cortex and caudate nucleus, and these connectivity pattern correlates with depression severity (Tsai et al., 2024). These results indicate that gut dysbiosis may contribute to depression via network-level disruptions.

Multi-omics and neuroimaging studies confirm this association between gut microbiota and brain structure/function. Children and adolescents with depression show altered gut microbiome accompanied by amino acid metabolism deficiencies, particularly decreased lysine, which induce depression-like behaviors by affecting prefrontal cortex metabolism (Teng et al., 2025). Additionally, specific genera like Alistipes correlate with gut metabolites (involving amino acid, vitamin B, and bile acid metabolism); these metabolites associate with functional alteration in the visual cortex, shaping cognitive and emotional dysfunction (Liu S. et al., 2025). Clinical trials further support intervention efficacy: probiotic intervention modulated gut microbiota composition, increased Lactobacillus, improved FC in prefrontal network, and alleviated depressive symptoms (Schaub et al., 2022; Yamanbaeva et al., 2023). Moreover, sleep deprivation-induced depressive phenotypes were similarly accompanied by gut dysbiosis and neuroinflammation, while astragalus supplementation restored SCFA levels, improving brain connectivity and inflammatory status (Jiao et al., 2025). These studies support a mechanistic role of microbial metabolites and immune signaling in depression pathogenesis.

Neuroimaging studies further reveal associations between gut microbiota and brain structure/function. Abnormal hippocampal FC correlates with increased abundance of pro-inflammatory gut bacteria (e.g., Enterobacteriaceae) in depressed patients (Lee et al., 2022). In geriatric depression, gut microbial diversity positively correlates with gray matter volume in the hippocampus and nucleus accumbens (Lee et al., 2022). Mendelian randomization analysis further supports a causal relationship between gut microbiota changes and brain function/structure metrics (Yang L. et al., 2025). Indoxyl sulfate (IS), a tryptophan-derived microbial product, may contribute to anxiety symptoms that is often accompanied by depression by activating resting-state FC in aversive-processing networks, including the subgenual anterior cingulate cortex (SCC-FC), bilateral anterior insula, right anterior mid-cingulate cortex, and right motor area (Brydges et al., 2021). Moreover, the interaction between Genus Parabacteroides and inflammatory factors is associated with static and dynamic low-frequency fluctuations in cerebellar Crus II in bipolar depression (Guo et al., 2024).

Together, the above evidence shows that gut microbiota mediate depression by affecting brain structural integrity, like the hippocampus and prefrontal cortex, and functional network organization, metabolic signaling pathways, and neuroinflammatory responses, providing converging neuroimaging and multi-omics evidence for the central role of the gut–brain axis in depressive pathophysiology.

Advances in neuroimaging have significantly amplified its central role in the diagnosis and therapeutic monitoring of central nervous system (CNS) diseases. Neuroimaging markers combined with gut microbiota profiles not only aid early detection, mechanistic interpretation, but also enable dynamic assessment of therapeutic responses.

In early disease stages, fMRI and PET can reveal subtle changes in regional brain function and metabolism that parallel gut microbial changes. For example, in Parkinson’s disease, FDG-PET can identify motor- and cognition-related brain network patterns which may reflect underlying microbiota-associated pathophysiology, providing valuable markers for early diagnosis (Nakano et al., 2025). Similarly, MR-DTI can capture microstructural abnormalities in childhood Tourette syndrome, potentially linked to microbiota-related immune modulation (Xia et al., 2020).

Neuroimaging provides objective markers for dynamically evaluating therapeutic response by monitoring changes in brain structure, neurotransmission, and network connectivity. For instance, imaging of dopaminergic pathways and changes in brain network connectivity in PD can effectively guide medication adjustment and improve therapeutic outcomes (de Natale et al., 2021). Furthermore, multimodal imaging techniques like MRI and PET, combined with microbial metabolite data, hold promise for developing comprehensive biomarkers for precise diagnosis and longitudinal efficacy monitoring (van Oostveen and de Lange, 2021).

Neuroimaging also plays a key role in designing personalized therapeutic strategies. By combining brain structural and functional information with individual gut microbiota profiles, clinicians can develop targeted psychobiotic interventions, dietary strategies, or pharmacological treatments. Functional neuroimaging can identify dysregulated networks and, combined with gut microbiota regulatory mechanisms, allows comprehensive neuro–gut axis–based treatment plans (Nakano et al., 2025). Moreover, modern imaging techniques like photoacoustic imaging and magnetic resonance spectroscopy can monitor brain neuroinflammatory status and metabolic changes in real-time, improving medication adjustment and efficacy sensitivity (Cheng L. L. et al., 2025). In inflammatory and neurodegenerative diseases, like multiple sclerosis, imaging not only helps assess lesion distribution and extent but also captures regional trajectories, providing a basis for individualized medicine (Cagol et al., 2024; Zhang Y. et al., 2021). Furthermore, the integration of neuroimaging with AI is accelerating precision neuromedicine. Deep learning and analysis of imaging data can identify subtle biomarkers, enhance patient stratification, and improve prognostic accuracy (Tsampras et al., 2025).

In summary, neuroimaging markers combined with gut microbial features provide new perspectives and technical means for the early disease detection, treatment monitoring, and personalized therapeutic planning, thereby advancing precision management of CNS disorders.

Driven by recent technological advances, research on the mechanisms linking gut microbiota to CNS diseases increasingly relies on multi-omics integration and emerging imaging techniques. Multi-omics approaches integrate microbiomic, metabolomic, immunomic, and neuroimaging data into systems biology framework, providing a comprehensive understanding of disease mechanisms. This multi-dimensional data integration compensates for the limitations of single-omics datasets and reveals the complex biological networks and dynamics of the gut-brain axis. In oncology and metabolic diseases, AI-based multi-omics fusion models significantly outperform single-modality approaches in disease classification, prognosis assessment, and treatment response prediction (Hatamikia et al., 2023; Wei et al., 2023). Similarly, radiomics combined with genomics and transcriptomics data can achieve spatiotemporal dynamic tracking of tumor heterogeneity, providing important guidance for precision diagnosis and treatment (Gevaert et al., 2020; Su et al., 2023). Furthermore, the development of multi-omics graph database systems further enhances large-scale data storage and cross-modal analysis, promoting efficient integration and mining of heterogeneous datasets (Thapa and Ali, 2021).

Emerging imaging technologies, such as single-cell imaging, functional connectome analysis, and machine learning-based approaches, greatly advance the in-depth analysis of gut-brain axis mechanisms. Single-cell imaging technology can map the spatial distribution of metabolites, proteins, and transcripts at cellular resolution, revealing key patterns of cellular heterogeneity and disease-related microenvironments (Lim et al., 2023). Spatial omics techniques, like spatial transcriptomics and multimodal mass spectrometry imaging, accurately localize cell types and metabolites within the tissue microenvironment, providing more precise insight into microbe-neural interactions along the gut-brain axis (Clarke et al., 2025; Hahn et al., 2025). Functional connectome analysis using fMRI, combined with machine learning algorithms, helps analyze dynamic changes in brain networks and their association with gut microbiota profiles in psychiatric disorders and cognitive impairment (Yang Q. et al., 2025; Zhang et al., 2025).

Machine learning and AI play increasing roles in multi-omics fusion and imaging analysis. Deep learning models automate the extraction and correlation analysis of high-dimensional biological and imaging features, improving the accuracy of disease risk prediction and classification (Ma et al., 2025; Zhao T. et al., 2025). For example, in complex diseases such as lung cancer and breast cancer, AI-assisted multimodal fusion models integrating clinical, imaging, and molecular data, have improved prognostic assessment and treatment decision-making (Singh et al., 2025; You et al., 2024). Additionally, digital twin technology combined with multi-omics and imaging datasets provides new clue for precision medicine by establishing individualized, dynamic disease models (Vallée, 2025). Continued advances in imaging resolution, analytical pipelines, and data standardization will further integrate multi-omics approaches with emerging imaging technologies, deepening mechanistic insights into the gut–brain axis and promoting early diagnosis, precise treatment, and personalized prognosis assessment of CNS diseases.

The relationship between gut microbiota and CNS diseases, particularly brain structural and functional alterations detected by neuroimaging, has emerged as a major frontier in neuroscience and microbiome studies. Although substantial studies confirm that gut microbiota contribute to neurodegenerative diseases, psychiatric disorders, and brain injury via the gut-brain axis, the precise causal pathways and mechanistic details remain incompletely elucidated. A primary challenge for future research to clarify the causal relationship between gut microbial alterations and brain imaging alterations. Advances in multimodal MRI, PET, and microbiome-related metabolomics, provide powerful tools for revealing this causality. For example, Mendelian randomization studies have found causal associations between specific gut bacterial genera and iron-related brain imaging phenotypes, providing genetic evidence for microbiome effects on brain function and structure (Huang et al., 2024). Additionally, animal models, using germ-free approaches, microbiota transplantation, and high-resolution neuroimaging, help reveal molecular mechanisms of microbial regulation of brain microstructure and function (Montoro et al., 2022). However, the gut-brain axis is shaped by multiple factors, such as immune responses, metabolites, and CNS neuroinflammatory responses, highlighting the need for interdisciplinary, systematic research.

Promoting large-scale, multi-center, longitudinal clinical studies will be essential for future clinical application of these findings. Currently, the majority of studies are cross-sectional, limiting the capacity to reveal the dynamic interactions between gut microbiota changes, brain imaging features, and disease progression. Large, multi-center prospective cohorts with repeated sampling of microbiome, metabolome, and neuroimaging data will enable more robust causal inference and identification of biomarkers. For example, studies in multiple sclerosis intergrating microbiota profiles, immune status, and MRI assessment, have initially revealed associations between gut dysbiosis and central inflammation/demyelination (Moles et al., 2021; van Pamelen et al., 2020). Furthermore, longitudinal studies on AD and schizophrenia further demonstrate the potential of the gut-brain axis markers for early disease warning, therapeutic monitoring, and personalized treatment (Sheng et al., 2023; Wu et al., 2024). Future efforts should also strengthen multi-omics integration utilizing AI and machine learning, facilitating predictive modeling and accelerating the translation of basic science into clinical applications.

Future research should expand exploration of the gut-brain axis across a wider range of CNS diseases and therapeutic targets. Beyond AD, multiple sclerosis, and schizophrenia, conditions such as traumatic brain injury, depression, ASD, and vascular cognitive impairment also exhibit gut microbiota dysfunction accompanied by neuroimaging abnormalities. For instance, post-TBI gut dysfunction correlates with neuroinflammatory responses and neurological recovery, suggesting the gut-brain axis modulation as a potential therapeutic target (You et al., 2025). ASD research using zebrafish models combined with genetics and live imaging have revealed complex interactions between microbial signals, neural development, and enteric neuroimmune pathways (Andersen-Civil et al., 2023). Deeper mechanistic studies on microglial regulation, immune cell migration, and the role of inflammatory mediators, is particularly important for identifying new therapeutic strategies (Ben-Azu et al., 2023; Eva et al., 2023). Pharmacological or non-pharmacological interventions, including chemical drugs, exercise, dietary modification, probiotics, and FMT, have shown potential to modulate the gut-brain axis, both human and animal studies have found that probiotic supplementation may prevent brain atrophy as detected by MRI, providing new ideas for clinical intervention (Asaoka et al., 2022; Elsherbiny et al., 2025; Sanchez-Martinez et al., 2024; Yuan et al., 2025).

In summary, future research on gut microbiota and brain imaging should further clarify causal mechanisms, expand multi-center longitudinal studies, and broaden the disease spectrum. With multidisciplinary integration, technological innovation, and large-scale data support, the field will be profoundly reveal the fundamental role of the gut-brain axis in CNS diseases and promote precise diagnosis, treatment, and personalized intervention, ultimately improving neurological health and patient outcomes.