Ibrain最新文献

{"title":"Roles of histone lysine methylation in neurodevelopment and related disorders","authors":"Yun Chen,&nbsp;Yi Zhang,&nbsp;Hong Zhang,&nbsp;Xiaoli Liang","doi":"10.1002/ibra.70022","DOIUrl":"https://doi.org/10.1002/ibra.70022","url":null,"abstract":"<p>Neurodevelopment is a highly ordered, precisely regulated process that establishes the neurobiological foundations of cognition, thought, emotion, and behavior. Neurodevelopmental disorders (NDDs) display marked phenotypic and genetic heterogeneity and variably impair learning, daily functioning, and social adaptation. As a key epigenetic mechanism, histone lysine methylation shapes chromatin accessibility and transcriptional programs, exerting central roles in neural stem cell fate decisions, neuronal migration and circuit assembly, as well as synaptic plasticity and learning and memory. Focusing on the activating histone H3 lysine 4 (H3K4) methylation marks (H3K4me1/2/3), this review synthesizes evidence for the bidirectional regulation mediated by “writers” (the KMT2/SET/DOT1) and “erasers” (the KDM/LSD and KDM5) in NDDs, including Kabuki syndrome, Wiedemann-Steiner syndrome, autism spectrum disorder, and schizophrenia. We further outline how animal models, patient-derived brain organoids, and multi-omics atlases enhance mechanistic insight, and we discuss the translational potential of small-molecule interventions, and metabolic modulation. Together, we summarize how the precise balance of methylation writing and erasure—and its crosstalk with DNA methylation and histone acetylation—forms an epigenetic network that drives neurodevelopmental programs. Targeting this network offers testable therapeutic avenues for NDDs.</p>","PeriodicalId":94030,"journal":{"name":"Ibrain","volume":"12 2","pages":"201-221"},"PeriodicalIF":0.0,"publicationDate":"2026-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ibra.70022","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148336819","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Risk of hemorrhagic stroke in warfarin-treated patients following heart valve replacement: Findings from the MAGPIE study","authors":"Dipannita Adhikary,&nbsp;Md. Abdullah Yusuf,&nbsp;AM Nayeem Parvez,&nbsp;Sheikh M. B. Faruque,&nbsp;Santo Barman,&nbsp;Adneen Moureen,&nbsp;Luke J. Rogers,&nbsp;Aziz Momin,&nbsp;Venkatachalam Chandrasekaran,&nbsp;Asit B. Adhikary,&nbsp;Redoy Ranjan","doi":"10.1002/ibra.70020","DOIUrl":"https://doi.org/10.1002/ibra.70020","url":null,"abstract":"<p>Hemorrhagic stroke (HS) remains a serious and potentially disabling complication following heart valve replacement (HVR). We aimed to evaluate the incidence and predictors of HS among mechanical HVR patients in Bangladesh. The Multidimensional Approach of Genotype and Phenotype In Stroke Etiology (MAGPIE) is an ambispective study recruited 568 consecutive HVR patients who had long-term (≥6 months) warfarin therapy between January 2010 and December 2024. Among the 568 HVR patients, 4.4% experienced HS, with a mortality rate of 52%, and the median age was 40 years (interquartile range [IQR]: 32–45). The median duration of warfarin uses among mechanical heart valve patients with HS varied by procedure: 63 months (IQR: 48–86) for mitral valve replacement, 60 months (12–85) for aortic valve replacement, 96 months (52–101.50) for double valve replacement, and 42 months (24–60) for patients undergoing coronary artery bypass grafting (CABG) with valve replacement. Additionally, a receiver operating characteristic curve analysis identified 104 months as the discriminatory threshold for warfarin therapy duration in predicting post-HVR HS onset. An age- and sex-adjusted logistic regression model identified severe pulmonary hypertension (odds ratio [OR] 4.44; 95% confidence interval [CI] 1.15–17.04; <i>p</i> = 0.02) and warfarin therapy duration ≥104 months (OR 1.99; 95% CI 1.00–3.76; <i>p</i> = 0.04) as independent predictors of HS in patients with mechanical HVR. Severe pulmonary hypertension was associated with a 4.4-fold higher risk and warfarin therapy beyond 104 months with a twofold higher risk of HS among patients with mechanical HVR.</p>","PeriodicalId":94030,"journal":{"name":"Ibrain","volume":"12 2","pages":"250-258"},"PeriodicalIF":0.0,"publicationDate":"2026-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ibra.70020","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148336822","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Pro-inflammatory diet mediates the association between epilepsy and sleep disorders: A cross-sectional study","authors":"Juan Yang,&nbsp;Yongsu Zheng,&nbsp;Xuejiao Zhou,&nbsp;Weijin Zheng","doi":"10.1002/ibra.70025","DOIUrl":"https://doi.org/10.1002/ibra.70025","url":null,"abstract":"<p>Accumulating epidemiological and mechanistic evidence indicates that anti-inflammatory dietary patterns may attenuate systematic inflammation underlying chronic disease, including epilepsy (EP), a common and potentially debilitating neurological disorder frequently complicated by sleep disturbances. Herein, we aim to assess the inflammatory potential of habitual diets among individuals with EP, and co-occurring sleep disorders (SLDs). From 2013 to March 2020, 35,706 participants were enrolled in this study. EP status was defined based on reported use of antiepileptic drugs within the preceding 30 days. SLDs were assessed using questions on habitual sleep duration and validated self-reported sleep disturbances. Dietary inflammatory index (DII) and energy-adjusted DII (E-DII) scores were calculated using 24-h dietary recall data. Individuals with EP exhibited significantly higher DII and E-DII scores and a greater prevalence of SLDs compared with individuals without EP. In the fully adjusted model, participants in Quartile 4 of the DII score had significantly increased odds of EP and SLDs, relative to those in Quartile 1. Mediation analyses revealed that DII accounted for 2.650% to 6.234% of the association between EP and SLDs. In stratified analysis, an elevated prevalence of SLDs was observed across Quartile 2–4 for DII and in Quartile 2 and 4 for E-DII among individuals with EP. Individuals with EP consume diets with greater pro-inflammatory potential than those without EP. SLDs in this population are associated with higher dietary inflammatory burden—highlighting the need for evidence-informed, individualized nutritional guidance as part of comprehensive EP management.</p>","PeriodicalId":94030,"journal":{"name":"Ibrain","volume":"12 2","pages":"238-249"},"PeriodicalIF":0.0,"publicationDate":"2026-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ibra.70025","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148336824","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Potential neuroprotective and therapeutic agents and their mechanisms for irradiation-induced brain injury","authors":"Seidu A. Richard,&nbsp;Vivian Kapio Abem,&nbsp;Sagor Kumar Roy","doi":"10.1002/ibra.70011","DOIUrl":"https://doi.org/10.1002/ibra.70011","url":null,"abstract":"<p>Cranial-irradiation is associated with tissue damage resulting in neurocognitive impediments that adversely influence patient quality of life. Administration of radiation directly to the tumor may extend to the neighboring healthy tissues, which may induce acute to persistent oxidative stress, lessening neurogenesis, neuroinflammation, as well as vascular alterations, leading to neurocognitive sequelae as a result of decline in neuronal structural complexity as well as synaptic connections. Almost all the medications indicated in the treatment of irradiation-triggered brain injury work via signaling pathways that are associated with lessening chronic oxidative stress, which is considered a consequence of the inflammatory response, reduction of edema, as well as microglia activation. Some agents have both preventative as well as therapeutic potential via the amalgamation of both neuroprotective and therapeutic mechanisms above. Thus, in this review, agents such as baicalein, troxerutin, epigallocatechin gallate, quercetin, melatonin, valproic acid, lithium, neurosteroid progesterone as well as minocycline have been implicated as neuroprotective agents for irradiation-induced neurological deficits. Also, agents such as glucocorticoids, methylphenidate, vitamin E, bisdemethoxycurcumin, phosphodiesterases, edaravone, pioglitazone and fenofibrate, glutamate antagonists, human urinary kallidinogenase, bevacizumab, as well as hyperbaric oxygen have been implicated as therapeutic agents for irradiation-induced neurological deficits. Furthermore, agents such as angiotensin-converting enzyme, 3-N-butyl-phthalide, stem cell therapy, sphingosine-1-phosphate, gangliosides, and neurotrophins have been implicated as combined potential neuroprotective and therapeutic agents for irradiation-induced neurological deficits. The aim of this review is to elucidate the potential neuroprotective and therapeutic agents above and their mechanisms for irradiation-induced neurological deficits after brain irradiation.</p>","PeriodicalId":94030,"journal":{"name":"Ibrain","volume":"12 1","pages":"96-122"},"PeriodicalIF":0.0,"publicationDate":"2026-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ibra.70011","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147566047","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Facial expression recognition for emotion perception: A comprehensive science mapping","authors":"Hou-Ming Kan,&nbsp;Li-Ping Chen,&nbsp;Yu Zhang,&nbsp;Hao-Yuan Hong,&nbsp;Ying-Ying Qin,&nbsp;Yu-Guo Cui,&nbsp;Yu-Bo Mao,&nbsp;Yan-Zhi Cheng,&nbsp;Zhe Lu,&nbsp;Hong-Yan Ni,&nbsp;Xiao-Tong Ding","doi":"10.1002/ibra.70010","DOIUrl":"https://doi.org/10.1002/ibra.70010","url":null,"abstract":"<p>Facial expression recognition (FER) has emerged as a pivotal interdisciplinary research domain that bridges computer science, psychology, neuroscience, and medicine. By mapping the FER scientific knowledge graph, this study aimed to explore the technological evolution and forecast future trends in this field. The study collected and cleaned the research on emotion perception in the Web of Science (WoS) database, and utilized the software CiteSpace (version 6.4R1) and R (BiblioShiny packages) software to create a scientific knowledge map. K-means was used for cluster analysis, and then the latent Dirichlet allocation (LDA) was employed to extract popular topics from the text of each cluster. Uniform manifold approximation and projection (UMAP) was utilized to reduce high-dimensional embeddings to a two-dimensional space. From a regional perspective, research is mainly distributed in countries or regions such as North America, Western Europe, East Asia, India, and Australia. Research on facial emotion recognition has focused primarily on neuroscience, psychiatry, and psychology. With the rapid development of computer technology, the interdisciplinary intersection is becoming increasingly important as FER has shown strong potential in identifying rare and neurological diseases. Furthermore, the evolution of artificial intelligence (AI) has transformed facial expression feature extraction from manual methodologies to machine learning-based approaches. The rapid development of computer algorithms and AI has greatly improved the accuracy and speed of facial emotion recognition. As a technology capable of detecting instantaneous emotional changes, FER holds promising prospects in fields such as neuroscience, emotion analysis, and pain assessment.</p>","PeriodicalId":94030,"journal":{"name":"Ibrain","volume":"12 1","pages":"38-51"},"PeriodicalIF":0.0,"publicationDate":"2026-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ibra.70010","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147566048","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Impact of early dexmedetomidine administration on short- and long-term outcomes in critically traumatic brain injury patients: A retrospective study using the MIMIC-IV database","authors":"Yang Long,&nbsp;Long-Zhen Liu,&nbsp;Xiao-Xue Tan,&nbsp;Yong-Hu Zhang,&nbsp;Cheng-Xi Liu,&nbsp;Shan Liu,&nbsp;Wen-Jie Liu","doi":"10.1002/ibra.70015","DOIUrl":"https://doi.org/10.1002/ibra.70015","url":null,"abstract":"<p>Traumatic brain injury (TBI) causes significant mortality. Dexmedetomidine (DEX) shows neuroprotective potential in animals, but clinical evidence remains inconsistent. We evaluated the impact of early DEX, initiated within 48 h of admission with a treatment duration of at least 4 h, on survival in intensive care unit (ICU) patients with TBI using the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database. Outcomes included 28-day, hospital, and 1-year mortality, analyzed <i>via</i> propensity score matching (PSM), multivariable Cox models, and subgroup analyses. Of 2378 patients, 241 received DEX. After PSM (235 pairs), early DEX use significantly reduced 28-day (HR 0.45, 95% CI 0.30–0.69, <i>p</i> &lt; 0.001) and hospital mortality (HR 0.25, 95% CI 0.15–0.42, <i>p</i> &lt; 0.001). These results remained robust across sensitivity analyses. Similarly, 1-year mortality decreased (HR 0.64, 95% CI 0.47–0.87, <i>p</i> &lt; 0.01) and further supported by the Boruta algorithm, although inverse probability of treatment weighting analysis showed only a non-significant trend (<i>p</i> = 0.08). Survival benefits were more pronounced in patients aged &lt;65 and those requiring mechanical ventilation. In conclusion, early DEX use is associated with improved short- and long-term survival in ICU patients with TBI, particularly in younger individuals and those requiring mechanical ventilation. Randomized controlled trials are warranted to establish causality.</p>","PeriodicalId":94030,"journal":{"name":"Ibrain","volume":"12 1","pages":"137-150"},"PeriodicalIF":0.0,"publicationDate":"2026-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ibra.70015","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147564816","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Facial expression recognition for emotion perception: A comprehensive science mapping","authors":"Hou-Ming Kan,&nbsp;Li-Ping Chen,&nbsp;Yu Zhang,&nbsp;Hao-Yuan Hong,&nbsp;Ying-Ying Qin,&nbsp;Yu-Guo Cui,&nbsp;Yu-Bo Mao,&nbsp;Yan-Zhi Cheng,&nbsp;Zhe Lu,&nbsp;Hong-Yan Ni,&nbsp;Xiao-Tong Ding","doi":"10.1002/ibra.70010","DOIUrl":"https://doi.org/10.1002/ibra.70010","url":null,"abstract":"<p>Facial expression recognition (FER) has emerged as a pivotal interdisciplinary research domain that bridges computer science, psychology, neuroscience, and medicine. By mapping the FER scientific knowledge graph, this study aimed to explore the technological evolution and forecast future trends in this field. The study collected and cleaned the research on emotion perception in the Web of Science (WoS) database, and utilized the software CiteSpace (version 6.4R1) and R (BiblioShiny packages) software to create a scientific knowledge map. K-means was used for cluster analysis, and then the latent Dirichlet allocation (LDA) was employed to extract popular topics from the text of each cluster. Uniform manifold approximation and projection (UMAP) was utilized to reduce high-dimensional embeddings to a two-dimensional space. From a regional perspective, research is mainly distributed in countries or regions such as North America, Western Europe, East Asia, India, and Australia. Research on facial emotion recognition has focused primarily on neuroscience, psychiatry, and psychology. With the rapid development of computer technology, the interdisciplinary intersection is becoming increasingly important as FER has shown strong potential in identifying rare and neurological diseases. Furthermore, the evolution of artificial intelligence (AI) has transformed facial expression feature extraction from manual methodologies to machine learning-based approaches. The rapid development of computer algorithms and AI has greatly improved the accuracy and speed of facial emotion recognition. As a technology capable of detecting instantaneous emotional changes, FER holds promising prospects in fields such as neuroscience, emotion analysis, and pain assessment.</p>","PeriodicalId":94030,"journal":{"name":"Ibrain","volume":"12 1","pages":"38-51"},"PeriodicalIF":0.0,"publicationDate":"2026-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ibra.70010","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147565995","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Preclinical advances in antibodies against N-terminal Aβ4-x species for Alzheimer's disease and cerebral amyloid angiopathy","authors":"Maria Luisa Valle,&nbsp;Donatello Arienzo","doi":"10.1002/ibra.70016","DOIUrl":"https://doi.org/10.1002/ibra.70016","url":null,"abstract":"<p>Alzheimer's disease (AD) pathogenesis is characterized by the accumulation of amyloid beta (Aβ) deposits in the cerebral parenchyma and vasculature, a condition referred to as cerebral amyloid angiopathy (CAA). Besides the full-length form, Aβ deposits showed a highly heterogenous composition due to the action of different proteolytic enzymes. N-terminal Aβ peptides have shown higher aggregation propensity and toxicity compared to the other truncated forms, with species starting at residue phenylalanine 4 (Aβ4-x) being more neurotoxic than the others. Thus, Aβ4-x species have drawn attention in AD pathogenesis, potentially offering novel therapeutic targets to halt or reverse disease progression. Antibodies targeting specifically Aβ4-x species were designed with the aim of preventing their aggregation and promoting their clearance counterbalancing their neurotoxic effect. This work provides an update on monoclonal and polyclonal antibodies developed to specifically target Aβ4-x species in AD and CAA preclinical studies (in vitro and in vivo models).</p>","PeriodicalId":94030,"journal":{"name":"Ibrain","volume":"12 1","pages":"52-64"},"PeriodicalIF":0.0,"publicationDate":"2026-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ibra.70016","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147562117","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Molecular imprinting for neurology: Materials, applications, and limitations","authors":"Xiaohan Ma,&nbsp;Yingqi Ma,&nbsp;Claudia Marino,&nbsp;Alessandro Poma","doi":"10.1002/ibra.70009","DOIUrl":"https://doi.org/10.1002/ibra.70009","url":null,"abstract":"<p>Neurological disorders represent one of the most pressing challenges in contemporary medicine, requiring tools that enable early diagnosis, targeted treatment, and deeper mechanistic understanding. Conventional biological agents such as antibodies, though widely used, often face limitations related to stability, brain penetration, cost, and integration into real-world platforms. Molecularly imprinted polymers (MIPs) have emerged as a promising synthetic alternative, capable of mimicking the molecular recognition functions of biological systems while offering enhanced robustness and customisability. MIPs provide key advantages, including high thermal and chemical stability, reusability, and design flexibility, making them especially attractive for neurological applications. In this review, we begin by presenting an overview of the main MIP formats applied in neurology, detailing their preparation, characterisation, and application-relevant advantages and limitations. We then explore the most actively investigated areas of MIP use in neurological diagnostics, research, and therapy, with a particular focus on: (i) nerve agents, (ii) neurotransmitters, and biomarkers, and (iii) drug development, drug delivery, and direct biological activity. Finally, we discuss the key challenges that currently hinder the clinical translation of MIPs in neurology, including poor biodegradability, in vivo biocompatibility concerns, and scalability, along with emerging strategies aimed at overcoming these barriers. We hope this analysis will serve as a useful reference for neuroscientists seeking novel material-based tools, as well as for materials scientists aiming to develop neurological applications of molecular imprinting.</p>","PeriodicalId":94030,"journal":{"name":"Ibrain","volume":"12 1","pages":"3-37"},"PeriodicalIF":0.0,"publicationDate":"2026-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ibra.70009","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147562958","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development of a predictive model for depressive symptoms in type 2 diabetes mellitus patients under community management: Based on visual function index","authors":"Rong-song Sun,&nbsp;Sheng-xu Huo,&nbsp;Tian-lin Zhang,&nbsp;Hao-jiang Ying,&nbsp;Jun-hua Wang,&nbsp;Li-na Xu,&nbsp;Xin-lin Luo,&nbsp;Yi Yang,&nbsp;Yuan-dong Hu","doi":"10.1002/ibra.70014","DOIUrl":"https://doi.org/10.1002/ibra.70014","url":null,"abstract":"<p>Visual impairment has been recognized as a potential risk factor for depressive symptoms (DS) in diabetes patients, yet the role of visual function in predicting DS remains unexplored. This study aims to develop and validate a predictive model for DS risk in type 2 diabetes mellitus (T2DM) patients in community health settings, incorporating a visual function index (VF14). We conducted a cross-sectional study involving 542 T2DM patients from four community health centers in Guiyang. Univariate and multivariate logistic regressions identified significant predictors, while 10 machine learning algorithms were employed to construct the predictive model. Model performance was assessed using such metrics as receiver operating characteristic curves, accuracy, sensitivity, specificity, F1 score, Brier score, C-index, calibration curves, and decision curve analysis. A restricted cubic spline (RCS) analysis evaluated the score-dependent risk profiles between the VF14 and DS. Key predictors included body mass index (BMI), self-reported glycemic status, age-related macular degeneration, glycated hemoglobin (HbA1c), and VF14. Among the models, the gradient boosting machine exhibited the robust predictive performance, with an area under the curve of 0.73 and sensitivity of 0.72. The Shapley additive explanations analysis identified VF14, BMI, and HbA1c as the top risk factors. RCS analysis revealed a score-dependent risk profile between VF14 and DS risk. This study introduces a clinically interpretable tool for early DS risk stratification in T2DM patients, offering potential for improved risk assessment and timely intervention in community health settings.</p>","PeriodicalId":94030,"journal":{"name":"Ibrain","volume":"12 1","pages":"123-136"},"PeriodicalIF":0.0,"publicationDate":"2026-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ibra.70014","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147564964","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}