Neuroimage-Clinical最新文献

{"title":"Neuroimaging correlates of domain-specific cognitive deficits in amyotrophic lateral sclerosis","authors":"Harold H.G. Tan ,&nbsp;Abram D. Nitert ,&nbsp;Kevin van Veenhuijzen ,&nbsp;Stefan Dukic ,&nbsp;Martine J.E. van Zandvoort ,&nbsp;Jeroen Hendrikse ,&nbsp;Michael A. van Es ,&nbsp;Jan H. Veldink ,&nbsp;Henk-Jan Westeneng ,&nbsp;Leonard H. van den Berg","doi":"10.1016/j.nicl.2025.103749","DOIUrl":"10.1016/j.nicl.2025.103749","url":null,"abstract":"<div><h3>Background</h3><div>Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease with frequent extra-motor involvement. In the present study, we investigated whether specific cognitive and behavioral deficits in ALS correlate with distinct extra-motor neurodegeneration patterns on brain MRI.</div></div><div><h3>Methods</h3><div>We performed multimodal brain MRI and Edinburgh cognitive and behavioral ALS screen (ECAS) in 293 patients and 237 controls. Follow-up data were acquired from 171 patients with a median duration of 7.9 months. Domain-level cognitive scores from the ECAS were compared with grey and white matter MRI parameters. Interaction analyses between patients and controls were performed to explore whether correlates were specific to ALS, rather than related to normal aging. Follow-up data were used to assess changes of domain-associated brain structures over time.</div></div><div><h3>Results</h3><div>Language impairment was significantly associated with (left predominant) frontal, temporal, parietal and subcortical grey matter neurodegeneration. Letter fluency with widespread cortical and subcortical grey matter involvement. Memory dysfunction with hippocampal and medial-temporal atrophy. Executive impairment was exclusively correlated with widespread white matter impairment. Visuospatial scores did not correlate with MRI parameters. Interaction analyses between patients and controls showed that most ECAS-MRI correlations were stronger in ALS than in controls (75.7% significant in grey matter, 52.7% in white matter). Longitudinal analyses showed that all grey matter structures associated with cognitive domains worsened over time while, for this study population, ECAS domain scores did not decline significantly.</div></div><div><h3>Conclusions</h3><div>MRI can capture the heterogeneity of cognitive and behavioral involvement in ALS and provides a useful longitudinal biomarker for progression of extra-motor neurodegeneration.</div></div>","PeriodicalId":54359,"journal":{"name":"Neuroimage-Clinical","volume":"45 ","pages":"Article 103749"},"PeriodicalIF":3.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143395222","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Neurometabolic network (NMetNet) for functional neurological disorder in children and adolescents","authors":"Zhou Lan ,&nbsp;Sheryl Foster ,&nbsp;Molly Charney ,&nbsp;Max van Grinsven ,&nbsp;Katherine Breedlove ,&nbsp;Kasia Kozlowska ,&nbsp;Alexander Lin","doi":"10.1016/j.nicl.2025.103767","DOIUrl":"10.1016/j.nicl.2025.103767","url":null,"abstract":"<div><h3>Objectives</h3><div>Functional neurological disorder (FND) in children and adolescents is a biopsychosocially complex condition characterized by a wide range of neurological symptoms. Using magnetic resonance spectroscopy to study neurometabolites has become an important approach to studying the mechanisms of FND. Unlike previous studies focusing on concentration-level analysis, this study examines conditional dependencies between six neurometabolites: N-acetyl aspartate, creatine, glutathione, choline, myo-inositol, and glutamate. Conditional dependence implies that two neurometabolites have joint variability that is not mediated by other neurometabolites.</div></div><div><h3>Methods</h3><div>A Bayesian graphical lasso approach was used to estimate neurometabolites’ conditional dependencies in three regions of interest: the anterior default mode network (aDMN), supplementary motor area (SMA), and posterior default mode network (pDMN). We introduce the term <em>neurometabolic network</em> (NMetNet) to describe these conditional dependencies.</div></div><div><h3>Results</h3><div>Children and adolescents with FND (vs. healthy controls) showed a loss of conditional dependencies related to creatine and glutathione between the aDMN and SMA/pDMN. Glutathione is the primary antioxidant in the brain. Creatine plays a key role in maintaining bioenergetics and also acts as an antioxidant.</div></div><div><h3>Conclusions</h3><div>These findings suggest that FND is characterized by dysregulated bioenergetics and increased vulnerability to oxidative stress. Understanding NMetNet in FND offers novel insights into the disorder’s neurobiology, with implications for therapeutic interventions to restore energy homeostasis and oxidative balance.</div></div>","PeriodicalId":54359,"journal":{"name":"Neuroimage-Clinical","volume":"46 ","pages":"Article 103767"},"PeriodicalIF":3.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143768586","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Altered cerebellar activation patterns in Alzheimer’s disease: An activation likelihood estimation Meta-Analysis","authors":"Jessica A. Bernard ,&nbsp;Ivan A. Herrejon ,&nbsp;Emily An ,&nbsp;Yamilet Cina ,&nbsp;Sameera Dabbiru ,&nbsp;Jack Dempsey ,&nbsp;Elise Marrie ,&nbsp;Michele Medina ,&nbsp;Jessica Praytor","doi":"10.1016/j.nicl.2025.103770","DOIUrl":"10.1016/j.nicl.2025.103770","url":null,"abstract":"<div><div>The past decade has seen an increased interest in the cerebellum, particularly in non-motor behaviors. Emerging work across model systems and in humans has also implicated the cerebellum in Alzheimer’s Disease (AD) and in mild cognitive impairment (MCI). While the cerebellum is not seen as being central to the etiology of the disease, it is however recognized as being increasingly important, and most certainly not immune from disease-related pathology and atrophy. In cognitively normal older adults (OA), the cerebellum has been conceptualized as being critical scaffolding for cortical function. This scaffolding may extend to AD and MCI. With respect to functional imaging, this is largely unexplored in AD, as this is a nascent literature. While there are very few studies focused on the cerebellum in AD at this stage, <em>meta</em>-analysis provides a powerful tool for expanding our knowledge of the cerebellum in neurodegenerative disease, and, in turn, for hypothesis generation. We took advantage of activation likelihood estimation (ALE) <em>meta</em>-analysis to investigate overlap in functional activation present in the existing literature. We focused on AD, but also included an exploratory analysis of MCI, based on papers available in our AD search. Our analysis included a total of 29 studies, representing data from 236 individuals with AD, 159 with MCI, and 382 OA. Across these studies, there is no significant overlap in cerebellar activation in AD, though this is present in MCI. Analyses of group differences also suggest that across studies, there are patterns indicative of both greater and reduced activation in AD/MCI relative to OA. Across all findings, overlap was primarily centered on Crus I and Lobule VI. These findings suggest that cerebellar function is negatively impacted in AD, which in turn may impact behavior and symptomatology.</div></div>","PeriodicalId":54359,"journal":{"name":"Neuroimage-Clinical","volume":"46 ","pages":"Article 103770"},"PeriodicalIF":3.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143683530","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Hippocampus and amygdala volume and morphology in neurodegenerative disorders with co-morbid epilepsy","authors":"Ifrah Zawar ,&nbsp;Shen Zhu ,&nbsp;Mark Quigg ,&nbsp;Jaideep Kapur ,&nbsp;Carol Manning ,&nbsp;P. Thomas Fletcher","doi":"10.1016/j.nicl.2025.103830","DOIUrl":"10.1016/j.nicl.2025.103830","url":null,"abstract":"<div><h3>Background</h3><div>Epilepsy is common in Alzheimer’s disease (AD) and non-AD dementias. However, the neuroimaging correlates of epilepsy in dementias remain unexplored. We investigated mesial temporal morphology and volumes in AD (AD + Epi) and nonAD dementias (nonAD + Epi) with epilepsy.</div></div><div><h3>Methods</h3><div>Participants from 39 US Alzheimer’s disease centers (9/2005–12/2021) were classified into dementia with epilepsy (AD + Epi, nonAD + Epi), dementia without epilepsy (AD-Epi, nonAD-Epi); and healthy controls. Dementia with epilepsy participants with available MRIs were matched to dementia without epilepsy and healthy controls by age, sex, and dementia type (AD versus non-AD).</div><div>FreeSurfer segmented hippocampi and amygdalae. Point distribution models created via ShapeWorks quantified morphological differences in the left and right hippocampi and amygdalae. Hippocampal and amygdalar volumes were normalized to the total intracranial volume. Multivariate analysis of covariates (MANCOVA), adjusted for age, sex, intracranial volume, and dementia severity, identified statistically significant local morphological and normalized volume group differences.</div></div><div><h3>Result</h3><div>A total of 703 participants (average age: 70.78 years, 391 (55.62 %) female) were included. AD-Epi and NonAD-Epi exhibited uniform hippocampal and amygdalar morphological atrophy bilaterally. In contrast, AD + Epi demonstrated morphological atrophy in the hippocampal bodies and tails bilaterally with sparing of the hippocampal heads, more pronounced inward deviations on mesial and lateral surfaces, and outward deviations in the middle hippocampal body bilaterally on the superior surface. NonAD + Epi showed significant morphological atrophy in the right hippocampal head, tail, and amygdala. No group volume differences were found.</div></div><div><h3>Conclusion</h3><div>We identified hippocampal body and tail atrophy in AD + Epi and right hippocampal head, tail, and amygdalar atrophy in nonAD + Epi. Different lateralized and region-specific patterns of limbic atrophy and dysmorphia highlight potential differences in the pathophysiology and the possible role of epilepsy in altering the trajectory of neurodegeneration in AD and nonAD.</div></div>","PeriodicalId":54359,"journal":{"name":"Neuroimage-Clinical","volume":"47 ","pages":"Article 103830"},"PeriodicalIF":3.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144321049","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Corrigendum to “Detection of spinal long fiber tract degeneration in HSP: Improved diffusion tensor imaging” [NeuroImage Clin. 36 (2022) 103213]","authors":"Tobias Lindig ,&nbsp;Christer Ruff ,&nbsp;Tim W. Rattay ,&nbsp;Stephan König ,&nbsp;Ludger Schöls ,&nbsp;Rebecca Schüle ,&nbsp;Thomas Nägele ,&nbsp;Ulrike Ernemann ,&nbsp;Uwe Klose ,&nbsp;Benjamin Bender","doi":"10.1016/j.nicl.2025.103777","DOIUrl":"10.1016/j.nicl.2025.103777","url":null,"abstract":"","PeriodicalId":54359,"journal":{"name":"Neuroimage-Clinical","volume":"46 ","pages":"Article 103777"},"PeriodicalIF":3.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143788806","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Determinants driving the evolution of new multiple sclerosis lesions into chronic active or remyelinated states","authors":"G. Boffa ,&nbsp;C. Razzetta ,&nbsp;D. Boccia ,&nbsp;S. Garbarino ,&nbsp;E. Cipriano ,&nbsp;A. Uccelli ,&nbsp;C. Lapucci ,&nbsp;M. Piana ,&nbsp;M. Inglese","doi":"10.1016/j.nicl.2025.103823","DOIUrl":"10.1016/j.nicl.2025.103823","url":null,"abstract":"<div><h3>Introduction</h3><div>Once formed, focal lesions that develop in patients with multiple sclerosis (MS) can follow different trajectories. We aimed at identifying early clinical and MRI features associated with the evolution of new MS lesions into chronic-active versus remyelinated states.</div></div><div><h3>Methods</h3><div>New contrast-enhancing (CE) lesions were classified after a 12-month follow-up with quantitative susceptibility mapping (QSM) into paramagnetic rim lesions (PRLs, representing chronic-active lesions) and isointense QSM lesions (ISO, representing remyelinated lesions). SHapley Additive exPlanations (SHAP) analysis, which highlights the most relevant features contributing to model predictions, was conducted using baseline clinical and MRI characteristics. A risk score was calculated for PRL and ISO classifications using the four most influential features for each task.</div></div><div><h3>Results</h3><div>A total of 111 lesions from 44 MS patients were analyzed. At 12 months, 13 % lesions were classified as PRL and 45 % as ISO. The key predictive features were similar for both classes (lesion volume, patient age and sex) except for the pattern of contrast enhancement (which was selected for PRL classification) and lesion topography (which was selected for ISO classification). Older age (&gt;48 years), male sex, bigger lesion volume (&gt;5 mL) and the presence of a ring pattern of contrast enhancement favored PRLs, while younger age (&lt;36 years), female sex, smaller lesion volume (&lt;0.17 mL) and the juxta-subcortical/deep white matter location favored ISO.</div></div><div><h3>Interpretation</h3><div>The outcome of a new MS lesion can be predicted at lesion onset considering few clinically accessible features.</div></div>","PeriodicalId":54359,"journal":{"name":"Neuroimage-Clinical","volume":"47 ","pages":"Article 103823"},"PeriodicalIF":3.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144291303","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"CTV-MIND: A cortical thickness-volume integrated individualized morphological network model to explore disease progression in temporal lobe epilepsy","authors":"Xinyan Liu ,&nbsp;Jiaqi Han ,&nbsp;Xiating Zhang ,&nbsp;Boxuan Wei ,&nbsp;Lu Xu ,&nbsp;Qilin Zhou ,&nbsp;Yuping Wang ,&nbsp;Yicong Lin ,&nbsp;Jicong Zhang","doi":"10.1016/j.nicl.2025.103843","DOIUrl":"10.1016/j.nicl.2025.103843","url":null,"abstract":"<div><div>Temporal lobe epilepsy (TLE) is a progressive brain network disorder. Elucidating network reorganization and identifying disease progression-associated biomarkers are crucial for understanding pathological mechanisms, quantifying disease burden, and optimizing clinical strategies. This study aimed to investigate progressive changes in TLE by constructing a novel individualized morphological brain network based on T1-weighted structural magnetic resonance imaging (MRI). MRI data were collected from 34 postoperative seizure-free TLE patients and 28 age- and sex-matched healthy controls (HC), with patients divided into LONG-TERM and SHORT-TERM groups. Individualized morphological networks were constructed using the Morphometric INverse Divergence (MIND) framework by integrating cortical thickness and volume features (CTV-MIND). Network properties were then calculated and compared across groups to identify features potentially associated with disease progression. Results revealed progressive hub-node reorganization in CTV-MIND networks, with the LONG-TERM group showing increased connectivity in the lesion-side temporal lobe compared to SHORT-TERM and HC groups. The altered network node properties showed a significant correlation with local cortical atrophy. Incorporating identified network features into a machine learning-based brain age prediction model further revealed significantly elevated brain age in TLE. Notably, duration-related brain regions exerted a more significant and specific impact on premature brain aging in TLE than other regional combinations. Thus, prolonged duration may serve as an important contributor to the pathological aging observed in TLE. Our findings could help clinicians better identify abnormal brain trajectories in TLE and have the potential to facilitate the optimization of personalized treatment strategies.</div></div>","PeriodicalId":54359,"journal":{"name":"Neuroimage-Clinical","volume":"48 ","pages":"Article 103843"},"PeriodicalIF":3.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144604127","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Altered self-processing brain networks in paediatric functional neurological disorder","authors":"Ishan C. Walpola ,&nbsp;Adith Mohan ,&nbsp;Sheryl Foster ,&nbsp;Kasia Kozlowska","doi":"10.1016/j.nicl.2025.103811","DOIUrl":"10.1016/j.nicl.2025.103811","url":null,"abstract":"<div><h3>Objectives</h3><div>Functional neurological disorder (FND) is associated with altered self-processing – the neural processes that underpin the individual’s self-agency and sense of self. This study used resting-state functional magnetic resonance imaging to examine three nested meta<strong>-</strong>analytically defined self-processing brain networks in 28 children with mixed FND symptoms and 31 healthy controls.</div></div><div><h3>Methods</h3><div>Regions of interest (ROIs) for each layer of brain network analysis were defined using the nested hierarchical model of self-processing developed by Northoff and colleagues specifying the insula (interoceptive processing), temporoparietal junction [TPJ] (exteroceptive processing), and anterior medial prefrontal cortex [amPFC] (mental-self processing) as seed ROIs. Connectivity differences for each layer of self-processing were examined with ROI-to-ROI analysis. Correlation analyses were conducted in relation to adverse childhood experiences (ACEs) and arousal (resting heart rate).</div></div><div><h3>Results</h3><div>At the mental-self layer, children with FND (vs. controls) had increased functional connectivity between the amPFC and thalamus and left dorsolateral prefrontal cortex (dlPFC). Children with functional seizures (vs. other FND symptoms) had decreased functional connectivity between the amPFC and right TPJ. At the interoceptive layer, the FND group showed a positive correlation between ACEs and functional connectivity between the left insula seed and right TPJ and dorsal anterior cingulate cortex (dACC). There were no findings at the exteroceptive layer of self-processing.</div></div><div><h3>Conclusions</h3><div>Our findings suggest that ACEs (including trauma) are associated with altered self-processing in brain networks in children with FND. Further examination of self-processing is likely to prove a fruitful endeavour both in therapy and in future FND research.</div></div>","PeriodicalId":54359,"journal":{"name":"Neuroimage-Clinical","volume":"47 ","pages":"Article 103811"},"PeriodicalIF":3.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144263294","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"What do we know about intratumoral functional activity: a scoping review of imaging and intra-surgery results","authors":"Ilaria Guarracino ,&nbsp;Marta Maieron ,&nbsp;Serena D’Agostini ,&nbsp;Tamara Ius ,&nbsp;Barbara Tomasino","doi":"10.1016/j.nicl.2025.103846","DOIUrl":"10.1016/j.nicl.2025.103846","url":null,"abstract":"<div><div>Intratumoral functional tissue represent a challenge in neurosurgery, as its resection may induce a permanent postoperative deficit. Only little is said in literature about this pattern. Currently this issue is receiving increased attention and in the last few years, the number of reports on intratumor functionality has increased. Aim of the current review was to provide a comprehensive overview of intratumoral area functionality patterns and of how much frequently this pattern is reported. PRISMA guidelines were followed. We identified 107 papers, but only 24 articles on 1220 patients were included for having reported intratumoral activation data. Within this framework, we aimed to shed light on some issues, including whether i) it is expressed only as fMRI activation within the mass, or whether it impacts on distant areas via functional connectivity, ii) it is found in slow growing tumors such as low grade glioma or also for fast infiltrative processes such as for high grade glioma, and iii) inhomogeneity of the tumor structure and morphological appearance or the tumor histology are key factors determining intratumoral area functionality. Key methods suitable for detecting intratumour function included MEG (in 7 studies), resting-state fMRI and task-active fMRI (in 8 studies) and intra-surgery direct cortical stimulation (in 8 studies). The type of patients were patients with astrocytoma (321 cases) and oligodendroglioma (255 cases) with tumor grade II (252 cases) and isocitrate dehydrogenase (IDH) mutation. Their mean tumor volume was 53.11 ± 19.23, and the affected hemisphere was mainly the left one (895 cases); lesion site most frequently involved the frontal cortex (435 cases). We discussed the clinical implications of these aspects, as a functional intratumoral area has a high impact on both planning and outcome, and we addressed the role of intra-surgery cognitive monitoring that should encompass a wide variety of functions.</div></div>","PeriodicalId":54359,"journal":{"name":"Neuroimage-Clinical","volume":"48 ","pages":"Article 103846"},"PeriodicalIF":3.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144653486","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Region-based U-nets for fast, accurate, and scalable deep brain segmentation: Application to Parkinson Plus Syndromes","authors":"Mengyu Li ,&nbsp;Magnús Magnússon ,&nbsp;Ingibjörg Kristjánsdóttir ,&nbsp;Sigrún Helga Lund ,&nbsp;Thilo van Eimeren ,&nbsp;Lotta M. Ellingsen ,&nbsp;Alzheimer’s Disease Neuroimaging Initiative","doi":"10.1016/j.nicl.2025.103807","DOIUrl":"10.1016/j.nicl.2025.103807","url":null,"abstract":"<div><div>The early diagnosis of age-related neurodegenerative diseases, which often progress to dementia, poses significant clinical challenges due to subtle and overlapping symptoms of these diseases at early stage. Automated MRI segmentation is important for early detection, as it offers consistent measurements and the ability to detect subtle structural changes in the brain. Manual segmentation is impractical for large datasets or clinical use. Deep learning approaches provide fast processing, however, they often encounter graphics processing unit (GPU) memory constraints when handling large datasets. Here we introduce a deep learning-based approach using region-based U-nets specifically designed to segment 12 deep-brain structures relevant to Parkinson Plus Syndromes. By dividing the brain image into targeted regions around the brainstem, ventricular system, and striatum, our method optimizes GPU usage and significantly reduces training times, while maintaining high accuracy. Validating the proposed method on three datasets, including a 660-subject clinical dataset comprising both healthy controls and patients with various movement disorders, we demonstrate robustness and practical applicability in separating different diseases. The method achieves superior segmentation performance compared to state-of-the-art methods, with a mean Dice Similarity Coefficient (DSC) of 0.90, a 95% Hausdorff Distance (HD95) of 1.35 mm, and an Average Symmetric Surface Distance (ASSD) of 0.45 mm, showcasing its segmentation accuracy and robustness. Furthermore, our method outperforms these methods by reducing training time from several days to a few hours while providing a processing time of less than a second per subject. The source code and trained model will be made publicly available on GitHub.</div></div>","PeriodicalId":54359,"journal":{"name":"Neuroimage-Clinical","volume":"47 ","pages":"Article 103807"},"PeriodicalIF":3.4,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144517195","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}