International Journal of Neuroscience最新文献

{"title":"A predictive model for postoperative delirium in patients admitted to the intensive care unit after cardiac valve surgery.","authors":"Zhizhen Shi, Zhen Yuan, Jiwei Cheng, Yanxin Zhao","doi":"10.1080/00207454.2026.2623101","DOIUrl":"10.1080/00207454.2026.2623101","url":null,"abstract":"<p><strong>Introduction: </strong>Postoperative delirium (POD) frequently occurs in patients following cardiac valve surgery in the intensive care unit (ICU), and is linked to adverse outcomes and extended hospitalization. The purpose of this study was to establish and validate a model for predicting high-risk individuals.</p><p><strong>Methods: </strong>This retrospective analysis enrolled adult patients who had cardiac valve surgery and subsequent ICU admission. Potential predictors were screened using LASSO logistic regression with cross-validation, followed by multivariable logistic regression, and a nomogram was developed from the final model. Model performance was assessed by receiver operating characteristic (ROC) curve analysis, calibration plots and decision curve analysis (DCA), and was benchmarked against the Sequential Organ Failure Assessment (SOFA) score.</p><p><strong>Results: </strong>In total, 3249 patients were analyzed, of whom 535 (16.5%) developed POD. Significant predictors identified were oxygen saturation, respiratory rate, urea nitrogen, INR, white blood cell (WBC) count, hemoglobin, SOFA score, myocardial infarction, diabetes and hyperlipidemia. The nomogram yielded AUCs of 0.766 (95% CI: 0.738-0.794) in the training cohort and 0.789 (95% CI: 0.750-0.828) in the validation cohort, showing significantly better predictive accuracy than the SOFA score (<i>p</i> < 0.05).</p><p><strong>Conclusions: </strong>The proposed model demonstrates strong predictive ability for POD in cardiac valve surgery patients and may aid clinicians in early risk stratification and targeted intervention. Further prospective validation is needed.</p>","PeriodicalId":14161,"journal":{"name":"International Journal of Neuroscience","volume":" ","pages":"710-721"},"PeriodicalIF":1.7,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146131886","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predictive value of systemic immune-inflammation index and neutrophil to high-density lipoprotein ratio in acute ischemic stroke.","authors":"Lei Meng, Angui Feng, Zhou Ding","doi":"10.1080/00207454.2025.2570439","DOIUrl":"10.1080/00207454.2025.2570439","url":null,"abstract":"<p><strong>Objective: </strong>To investigate the predictive value of the Systemic Immune-Inflammation Index (SII) and the neutrophil to high-density lipoprotein cholesterol ratio (NHR) for adverse cardiovascular and cerebrovascular events in patients with acute stroke.</p><p><strong>Methods: </strong>A retrospective analysis was conducted on 260 acute stroke patients. Based on the occurrence of adverse events, patients were divided into a no-event group (Group A, <i>n</i> = 195) and an event group (Group B, <i>n</i> = 65). Risk factors were identified using logistic regression, and the predictive performance of SII and NHR was assessed by receiver operating characteristic (ROC) curve analysis.</p><p><strong>Results: </strong>Compared with Group B, Group A had a higher proportion of male patients, older age, smoking history, and elevated levels of white blood cell count, neutrophil percentage, lymphocyte percentage, platelet count, neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and SII (<i>p</i> < 0.05). Univariate and multivariate logistic regression analyses demonstrated that, after adjusting for confounding factors such as sex and age, SII and NHR were independent risk factors for postoperative adverse cardiovascular events in acute stroke patients. ROC curve analysis revealed that the area under the curve (AUC) values for SII, NHR, and their combined detection in predicting adverse cardiovascular events were 0.729, 0.626, and 0.916, respectively, with significant differences (<i>p</i> < 0.05).</p><p><strong>Conclusion: </strong>SII and NHR are independent predictors of recurrent cardiovascular and cerebrovascular events in acute stroke patients. Combined detection of SII and NHR offers superior predictive performance.</p>","PeriodicalId":14161,"journal":{"name":"International Journal of Neuroscience","volume":" ","pages":"452-457"},"PeriodicalIF":1.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145251052","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"EEG spectral power differences in Alzheimer's disease and frontotemporal dementia.","authors":"Yue Pan, Li Zhu, Zi-Liang Wang","doi":"10.1080/00207454.2025.2591129","DOIUrl":"10.1080/00207454.2025.2591129","url":null,"abstract":"<p><strong>Background: </strong>Differentiating between Alzheimer's disease (AD) and frontotemporal dementia (FTD) based on clinical symptoms alone can be challenging. This study investigates the utility of resting-state EEG spectral power as a tool to distinguish between these two neurodegenerative conditions.</p><p><strong>Methods: </strong>We analyzed a publicly available dataset containing EEG recordings from 36 AD patients, 23 FTD patients, and 29 age-matched healthy controls (HC). Spectral power across delta, theta, alpha, and beta frequency bands was computed for both eyes-closed and eyes-open conditions. Rigorous statistical analysis with FDR correction was employed to identify group differences. To further investigate the relationship between EEG spectral alterations and clinical cognitive status, a correlation analysis was conducted.</p><p><strong>Results: </strong>Both patient groups showed significant deviations from HC, but with distinct patterns. AD was characterized by a classic pattern of posterior alpha power decrease and frontal theta power increase. In contrast, FTD showed a more focused reduction of alpha power at frontal and central sites. These patterns were robust across both eyes-closed and eyes-open states, suggesting their potential as stable biomarkers. The spectral features showed limited correlation with MMSE scores, indicating they may capture unique aspects of neuropathology not reflected in standard cognitive screening.</p><p><strong>Conclusion: </strong>Resting-state EEG reveals distinct spectral signatures for AD and FTD, supporting its potential as a low-cost, non-invasive adjunctive tool for differential diagnosis. The replication of these findings in an independent, open-access dataset underscores their reliability and provides a foundation for developing automated diagnostic algorithms.</p>","PeriodicalId":14161,"journal":{"name":"International Journal of Neuroscience","volume":" ","pages":"538-546"},"PeriodicalIF":1.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145540579","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Incidence and risk factors for transient and permanent central diabetes insipidus following pituitary adenoma surgery: a single center study.","authors":"Zahra Davoudi, Saeed Shahriari, Omidvar Rezaei, Mahdi Amirdosara, Kimia Vakili, Masood Zangi, Mohammad Samadian, Maryam Haghighimorad, Kave Ebrahimzade, Seyed Ali Mousavinejad, Fatemeh Sayehmiri, Mohammad Hallajnejad, Guive Sharifi","doi":"10.1080/00207454.2025.2564646","DOIUrl":"10.1080/00207454.2025.2564646","url":null,"abstract":"<p><strong>Background: </strong>Diabetes insipidus (DI) is a common complication following pituitary adenoma surgery, which can adversely affect patients' quality of life. This study aimed to clarify the types and frequency of DI in patients undergoing surgery for pituitary adenomas.</p><p><strong>Methods: </strong>This retrospective cross-sectional analytical study examined a cohort of patients (<i>n</i> = 532) with pituitary adenomas who underwent endoscopic transsphenoidal surgery (ETSS) at Loghman Hakim Hospital between 2017 and 2023. Comprehensive data collection included demographics, hormonal profiles, magnetic resonance imaging (MRI) findings and pathology results, enabling an in-depth analysis of DI incidence and associated risk factors post-surgery. Chi-square tests and logistic regression were employed to identify factors associated with DI.</p><p><strong>Results: </strong>Transient DI was most prevalent in patients with Cushing's disease (77.8%), while permanent DI was more commonly observed in non-functional adenomas (7%). However, no significant differences in DI prevalence were identified between adenoma types (<i>p</i> = .57 for transient DI, <i>p</i> = .89 for permanent DI). Tumor size and higher preoperative sodium levels were associated with transient and permanent DI in non-functional adenomas (<i>p</i> = .02, <i>p</i> = .026) and acromegaly (<i>p</i> = .012). Suprasellar extension correlated with transient DI in acromegaly (<i>p</i> = .01), while postoperative meningitis was significantly associated with transient DI in non-functional adenomas (<i>p</i> = .001).</p><p><strong>Conclusion: </strong>In the present study, tumor size, suprasellar extension and postoperative meningitis were linked to transient DI. In contrast, younger age, higher tumor volume and elevated mean 24-h postoperative sodium levels were associated with an increased risk of permanent DI.</p>","PeriodicalId":14161,"journal":{"name":"International Journal of Neuroscience","volume":" ","pages":"417-425"},"PeriodicalIF":1.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145495149","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A novel machine learning-based method to quantify the effect of transcranial direct current stimulation on opioid users.","authors":"Fatemeh Kazemzadeh, Sepideh Jabbari, Bahram Perseh, Zakaria Eskandari, Alireza Faridi, Davoud Ahmadi","doi":"10.1080/00207454.2025.2569423","DOIUrl":"10.1080/00207454.2025.2569423","url":null,"abstract":"<p><strong>Background: </strong>Opioid addiction is a major public health concern, associated with numerous health and social problems. Conventional diagnostic methods for addiction have notable limitations, highlighting the need for alternative approaches.</p><p><strong>Methods: </strong>This study investigates the use of electroencephalography (EEG) signals in conjunction with transcranial direct current stimulation (tDCS) for the diagnosis and treatment of opioid addiction. Thirty-six male patients undergoing methadone maintenance treatment were recruited and randomly assigned to three groups: Group A received left anodal/right cathodal tDCS (<math><mi>n</mi><mo>=</mo><mn>12</mn></math>), Group B received right anodal/left cathodal tDCS (<math><mi>n</mi><mo>=</mo><mn>12</mn></math>), and Group C received sham stimulation (<math><mi>n</mi><mo>=</mo><mn>12</mn></math>). EEG recordings were obtained from all participants before and after tDCS, as well as from 24 healthy controls. Machine learning techniques were applied to develop an optimized algorithm capable of distinguishing between healthy and addicted individuals by selectively analyzing addiction-affected EEG channels, thereby reducing processing time and costs.</p><p><strong>Results: </strong>The proposed method achieved a diagnostic accuracy of 94.30%. In addition, the effects of tDCS on craving reduction were assessed using EEG signals, psychological questionnaires, and blood biomarkers. Significant reductions in craving levels were observed in Groups A and B.</p><p><strong>Conclusion: </strong>These findings suggest that tDCS can be an effective intervention for reducing craving in patients with opioid addiction.</p>","PeriodicalId":14161,"journal":{"name":"International Journal of Neuroscience","volume":" ","pages":"439-451"},"PeriodicalIF":1.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145199195","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Concomitant trends in stroke and COPD-related mortality in the U.S.: a 25-year retrospective analysis of the CDC WONDER database.","authors":"Areej Javeid, Rayyan Nabi, Muhammad Affan, Hanzala Ahmed Farooqi, Sophia Ahmed, Javeria Javeid, Wajeeh Hassan, Tabeer Zahid, Najaf Ahmed Rajpar","doi":"10.1080/00207454.2025.2577353","DOIUrl":"10.1080/00207454.2025.2577353","url":null,"abstract":"<p><strong>Background: </strong>Chronic obstructive pulmonary disease (COPD) and stroke are leading contributors to mortality and disability. We analyzed national trends in deaths attributed to co-occurring stroke and COPD in US adults from 1999-2023.</p><p><strong>Methods: </strong>Using CDC WONDER multiple cause-of-death data for adults aged ≥25 years, we identified decedents with co-existent stroke (ICD-10 I60.x, I61.x, I63.x, I64, I69.0, I69.1, I69.3, I69.4) and COPD (J40-J44). Crude and age-adjusted mortality rates (AAMRs) per 100,000 population were calculated using the 2000 US standard population. Joinpoint regression estimated annual percent changes (APCs) and identified trend inflection points. Analyses were stratified by sex, race/ethnicity, metropolitan status, and state.</p><p><strong>Results: </strong>From 1999 to 2023, 311,375 deaths involved concurrent stroke and COPD. Overall AAMRs declined from 8.04 per 100,000 in 1999 to 5.17 in 2009 (APC -4.40%), with continued decline to 2018 (APC -1.34%). Rates then increased through 2020-2021 (APC +7.33%) before a modest decline toward 2023. Males consistently exhibited higher AAMRs than females. Non-Hispanic White adults had the highest AAMRs, whereas Hispanic and Asian/Pacific Islander groups had lower rates. Non-metropolitan areas experienced consistently higher AAMRs than metropolitan areas. State-level analyses identified the highest burdens in Appalachian and Deep South regions.</p><p><strong>Conclusion: </strong>Mortality from coexisting stroke and COPD declined for two decades but rose around the COVID-19 period, revealing significant sex, racial/ethnic, and geographic disparities. These findings highlight the need for targeted prevention, improved access to care for high-risk populations, and further research into mechanisms driving recent inflection points. Data and methods are detailed in the manuscript.</p>","PeriodicalId":14161,"journal":{"name":"International Journal of Neuroscience","volume":" ","pages":"478-488"},"PeriodicalIF":1.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145307780","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Preliminary findings on a deep learning model using electroencephalogram for multi-level neuropathic pain detection in post-stroke patients.","authors":"Sofia Bobby J, Sheeja V Francis, Subha Ramya V, Annapoorani C L","doi":"10.1080/00207454.2025.2584081","DOIUrl":"10.1080/00207454.2025.2584081","url":null,"abstract":"<p><strong>Aim: </strong>Neuropathic pain occurs commonly after stroke and represents a major source of disability for affected patients. This study aims to develop an accurate and computationally efficient framework for multi-level neuropathic pain detection using electroencephalography signals.</p><p><strong>Methods: </strong>A Quantum-Inspired Pyramid Depthwise Separable Residual Network is proposed, which integrates three innovations: a depthwise separable Residual Network to reduce computational complexity, a pyramid attention mechanism to capture multi-scale patterns, and a quantum-inspired transformation layer to model complex nonlinear dependencies among Electroencephalogram features.</p><p><strong>Results: </strong>Experiments conducted on benchmark electroencephalography datasets confirm that the proposed model gains a accuracy of 99.65%, with a recall of 98.00%.</p><p><strong>Conclusion: </strong>The proposed model provides a reliable solution for objective neuropathic pain detection in post-stroke patients. The framework demonstrates potential for integration into intelligent clinical decision-support and brain-computer interface-based rehabilitation systems.</p>","PeriodicalId":14161,"journal":{"name":"International Journal of Neuroscience","volume":" ","pages":"499-507"},"PeriodicalIF":1.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145458516","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"CD39⁺ CD4⁺ T cells influence cluster headache risk via ADP/N-acetylneuraminate and choline metabolic pathways: evidence from Mendelian Randomization.","authors":"Jingshan Zeng, Ying Yi, Hu Xie, Yun Zhu","doi":"10.1080/00207454.2025.2580332","DOIUrl":"10.1080/00207454.2025.2580332","url":null,"abstract":"<p><strong>Objective: </strong>This study employs the Mendelian randomization (MR) approach to investigate the causal relationships among immune cells, cluster headache (CH), and potential mediation by serum metabolites.</p><p><strong>Methods: </strong>Using genome-wide association study (GWAS) data, MR analyses were conducted on 731 immune cell phenotypes, 1400 serum metabolites, and CH. The inverse variance weighted (IVW) method was employed as the primary analytical approach, supplemented by MR-Egger and weighted median analyses. Stability of results was assessed using Cochran's Q and other statistical tests.</p><p><strong>Results: </strong>The analysis identified a negative causal relationship between CD39⁺ CD4⁺ %T cells and CH, supported by sensitivity analyses. Reverse MR analysis showed no effect of CH on CD39⁺ CD4⁺ T cells, suggesting a unidirectional role of these cells in reducing CH risk. Further mediation MR analysis indicated that CD39⁺ CD4⁺ T cells may influence CH risk through the regulation of either the adenosine 5'-diphosphate (ADP) to N-acetylneuraminate ratio or the choline phosphate to phosphoethanolamine ratio, with mediation effect ratios of 12.4% and 12.5%, respectively.</p><p><strong>Conclusion: </strong>CD39⁺ CD4⁺ T cells may reduce CH risk by increasing the adenosine 5'-diphosphate (ADP) to N-acetylneuraminate ratio or the choline phosphate to phosphoethanolamine ratio. These findings provide novel insights into potential targets for the prevention and treatment of CH.</p>","PeriodicalId":14161,"journal":{"name":"International Journal of Neuroscience","volume":" ","pages":"489-498"},"PeriodicalIF":1.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145370284","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unraveling the molecular landscape of ischemic stroke: a bioinformatics approach to identify key genes and therapeutic targets.","authors":"Saeedeh Sadat Mirtooni, Payam Sasannejad, Fariborz Rezaeitalab, Mohammad Reza Gholamian, Morteza Saeedi, Vajiheh Tolou, Forouzan Amerizadeh","doi":"10.1080/00207454.2025.2588190","DOIUrl":"10.1080/00207454.2025.2588190","url":null,"abstract":"<p><strong>Background: </strong>Ischemic stroke, a major cause of disability and death, results from blocked cerebral blood flow, causing neuronal damage. Limited treatments necessitate research into molecular mechanisms for targeted therapies.</p><p><strong>Methods: </strong>In this study, we analyzed transcriptomic data from the Gene Expression Omnibus (GEO) database (GSE36010), focusing on gene expression changes in the core region of the brain 24 h post-stroke induction. Enrichment analysis of differentially expressed genes (DEGs) was conducted using Enrichr, exploring Gene Ontology (GO) terms and Reactome pathways. A protein-protein interaction network (PPIN) was constructed using STRING, and hub genes were identified through the CytoHubba plugin in Cytoscape. Network clustering analysis was performed using the iterative partition-based clustering algorithm (IPCA) in CytoCluster. Promoter motif analysis of hub genes was conducted using TomTom and GO for motifs (GOMO), while DrugBank analysis was used to identify potential therapeutic compounds targeting the identified hub proteins.A total of 150 DEGs were identified in the core region of ischemic stroke, with significant enrichment in pathways related to inflammation, oxidative stress, apoptosis and vascular remodeling. Key hub genes like cluster of differentiation 68 (Cd68), chemokine (C-C motif) ligand 3 (Ccl3), transforming growth factor beta 1 (Tgfb1), cluster of differentiation 44 (Cd44), Lgals3, topoisomerase II alpha (Top2a), lipocalin 2 (Lcn2), chemokine (C-C motif) ligand 4 (Ccl4), Ckap2 and cell division cycle associated 2 (Cdca2) were found to play critical roles in neuronal survival, inflammation and vascular integrity. Additionally, drug analysis highlighted several FDA-approved drugs that could be repurposed for stroke therapy, emphasizing the potential therapeutic candidates targeting these hub proteins.</p><p><strong>Conclusion: </strong>This study analyzes ischemic stroke, identifying key pathways and drug targets, potentially aiding in developing targeted therapies to reduce damage and improve outcomes for stroke patients.</p>","PeriodicalId":14161,"journal":{"name":"International Journal of Neuroscience","volume":" ","pages":"517-529"},"PeriodicalIF":1.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145587260","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep machine learning methods for Parkinson's disease diagnosis: a new direction in decision-making systems.","authors":"P Pradeep, Kamalakannan J","doi":"10.1080/00207454.2025.2573458","DOIUrl":"10.1080/00207454.2025.2573458","url":null,"abstract":"<p><p>Parkinson's disease (PD) is a progressive neurodegenerative disorder primarily characterized by the gradual loss of dopamine-producing neurons in the brain's substantia nigra. The hallmark motor symptoms of PD include tremors, bradykinesia (slowness of movement), rigidity (muscle stiffness) and postural instability. However, the disease also manifests with significant non-motor symptoms, such as cognitive decline, mood disorders, sleep disturbances and autonomic dysfunction, which further complicate the clinical images. Accurate and early diagnosis of PD is challenging due to the subtlety and gradual onset of symptoms, as well as the overlap with other neurodegenerative disorders. Traditional diagnostic methods rely heavily on clinical evaluations and motor symptom assessments, which can be subjective and not detect early or asymptomatic stages of the disease. To overcome these challenges, this work aims to propose a novel Feature-level fusion-enabled PD detection (FLF-PDD) system, integrating an improved bidirectional- gated recurrent unit (Bi-GRU) architecture. This model progresses through several stages: preprocessing, where an Enhanced Gaussian Filtering technique reduces image noise; feature extraction, employing methods, such as Enhanced Pyramid Histograms of Oriented Gradients (PHOG), Multi Texton, LGXP and color analysis; FLF, utilizing principal component analysis (PCA) and tanh normalization to combine extracted features from various MRI orientations; and disease detection, facilitated by the trained Improved Bi-GRU model on fused features to accurately diagnose PD symptoms. The FLF-PDD model undergoes rigorous evaluation to enhance diagnostic accuracy and deepen the understanding of PD progression.</p>","PeriodicalId":14161,"journal":{"name":"International Journal of Neuroscience","volume":" ","pages":"458-477"},"PeriodicalIF":1.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145345232","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}