Academic Journal
Προβολή σε EDS

A combined model of convolutional neural networks and graph attention networks for improved classification of mild cognitive impairment.

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: A combined model of convolutional neural networks and graph attention networks for improved classification of mild cognitive impairment.
Συγγραφείς: Kim N; Department of Medical and Digital Engineering, Hanyang University, 222 Wangsimni-ro, Seongdong-gu, Seoul 04763, Republic of Korea. Electronic address: skdud990306@naver.com., Jeon JY; Department of Medical and Digital Engineering, Hanyang University, 222 Wangsimni-ro, Seongdong-gu, Seoul 04763, Republic of Korea. Electronic address: jyjeon@hanyang.ac.kr., Seo J; Department of Medical and Digital Engineering, Hanyang University, 222 Wangsimni-ro, Seongdong-gu, Seoul 04763, Republic of Korea. Electronic address: sjw007s@korea.ac.kr., Lee Y; Department of Medical and Digital Engineering, Hanyang University, 222 Wangsimni-ro, Seongdong-gu, Seoul 04763, Republic of Korea. Electronic address: yun22@hanyang.ac.kr., Kim HJ; Department of Medical and Digital Engineering, Hanyang University, 222 Wangsimni-ro, Seongdong-gu, Seoul 04763, Republic of Korea; Department of Neurology, College of Medicine, Hanyang University, 222 Wangsimni-ro, Seongdong-gu, Seoul 04763, Republic of Korea; Department of Neurology, Hanyang University Hospital, 222-1 Wangsimni-ro, Seongdong-gu, Seoul, 04763, Republic of Korea. Electronic address: hyumcbrain@hanyang.ac.kr., Kim JS; Clinical Research Institute, Konkuk University Medical Center, 120-1 Neungdong-ro, Gwangjin-gu, Seoul 05030, Republic of Korea; Institute of Biomedical Science and Technology, Konkuk University, 120-1 Neungdong-ro, Gwangjin-gu, Seoul 05030, Republic of Korea. Electronic address: junesic.kim@gmail.com.
Πηγή: NeuroImage [Neuroimage] 2026 Jan; Vol. 325, pp. 121674. Date of Electronic Publication: 2025 Dec 23.
Τύπος έκδοσης: Journal Article
Γλώσσα: English
Στοιχεία περιοδικού: Publisher: Academic Press Country of Publication: United States NLM ID: 9215515 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1095-9572 (Electronic) Linking ISSN: 10538119 NLM ISO Abbreviation: Neuroimage Subsets: MEDLINE
Imprint Name(s): Original Publication: Orlando, FL : Academic Press, c1992-
Ιατρικοί όροι (MeSH): Cognitive Dysfunction*/classification , Cognitive Dysfunction*/diagnostic imaging , Neural Networks, Computer*, Magnetic Resonance Imaging/methods ; Alzheimer Disease/diagnostic imaging ; Brain/diagnostic imaging ; Humans ; Aged ; Male ; Female ; Deep Learning ; Aged, 80 and over ; Middle Aged ; Convolutional Neural Networks
Περίληψη: Mild cognitive impairment (MCI), a precursor of Alzheimer's disease (AD), underscores the importance of early diagnosis and treatment. With an aging global population, AD prevalence is rising, necessitating more precise diagnostic methods. Deep learning technology shows promise for MCI and AD classification, but existing convolutional neural network (CNN) and graph attention network (GAT) models have limitations in capturing brain structural features and detecting microlesions. To address these issues, we propose a novel approach combining a CNN and modified GAT model to improve MCI classification. Magnetic resonance imaging volume data were analyzed using a CNN, whereas cortical thickness data were modeled using a GAT, leveraging their complementary strengths. Preprocessing involved extracting brain's structural features via the CIVET pipeline, and t-SNE was used to visualize the data's high-dimensional distribution. Final classification was performed using a multilayer perceptron, integrating feature vectors from both models. Performance evaluation metrics included the area under the curve (AUC), F1-score, sensitivity, and specificity. The combined CNN-GAT model outperformed existing single-model approaches, particularly in MCI classification, effectively distinguishing subtle variations between normal aging and MCI. The combined CNN-GAT model improved MCI classification performance by addressing the limitations of existing approaches. By capturing brain structural features and inter-regional relationships, it offers significant potential for advancing early diagnosis and treatment strategies for neurodegenerative diseases. Future efforts will focus on enhancing performance through additional data optimization.
(Copyright © 2025 The Author(s). Published by Elsevier Inc. All rights reserved.)
Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Contributed Indexing: Keywords: Convolutional neural network; Graph attention network; Mild cognitive impairment; Multilayer perceptron
Entry Date(s): Date Created: 20251225 Date Completed: 20260112 Latest Revision: 20260112
Update Code: 20260130
DOI: 10.1016/j.neuroimage.2025.121674
PMID: 41448509
Βάση Δεδομένων: MEDLINE
FullText Text:
  Availability: 0
CustomLinks:
  – Url: https://www.doi.org/10.1016/j.neuroimage.2025.121674?
    Name: ScienceDirect (all content) (s7799221)
    Category: fullText
    Text: View record from ScienceDirect
    MouseOverText: View record from ScienceDirect
  – Url: https://resolver.ebsco.com/c/fiv2js/result?sid=EBSCO:cmedm&genre=article&issn=10959572&ISBN=&volume=325&issue=&date=20260101&spage=121674&pages=121674&title=NeuroImage&atitle=A%20combined%20model%20of%20convolutional%20neural%20networks%20and%20graph%20attention%20networks%20for%20improved%20classification%20of%20mild%20cognitive%20impairment.&aulast=Kim%20N&id=DOI:10.1016/j.neuroimage.2025.121674
    Name: Full Text Finder (for New FTF UI) (ns324271)
    Category: fullText
    Text: Full Text Finder
    MouseOverText: Full Text Finder
Header DbId: cmedm
DbLabel: MEDLINE
An: 41448509
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: A combined model of convolutional neural networks and graph attention networks for improved classification of mild cognitive impairment.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AU" term="%22Kim+N%22">Kim N</searchLink>; Department of Medical and Digital Engineering, Hanyang University, 222 Wangsimni-ro, Seongdong-gu, Seoul 04763, Republic of Korea. Electronic address: skdud990306@naver.com.<br /><searchLink fieldCode="AU" term="%22Jeon+JY%22">Jeon JY</searchLink>; Department of Medical and Digital Engineering, Hanyang University, 222 Wangsimni-ro, Seongdong-gu, Seoul 04763, Republic of Korea. Electronic address: jyjeon@hanyang.ac.kr.<br /><searchLink fieldCode="AU" term="%22Seo+J%22">Seo J</searchLink>; Department of Medical and Digital Engineering, Hanyang University, 222 Wangsimni-ro, Seongdong-gu, Seoul 04763, Republic of Korea. Electronic address: sjw007s@korea.ac.kr.<br /><searchLink fieldCode="AU" term="%22Lee+Y%22">Lee Y</searchLink>; Department of Medical and Digital Engineering, Hanyang University, 222 Wangsimni-ro, Seongdong-gu, Seoul 04763, Republic of Korea. Electronic address: yun22@hanyang.ac.kr.<br /><searchLink fieldCode="AU" term="%22Kim+HJ%22">Kim HJ</searchLink>; Department of Medical and Digital Engineering, Hanyang University, 222 Wangsimni-ro, Seongdong-gu, Seoul 04763, Republic of Korea; Department of Neurology, College of Medicine, Hanyang University, 222 Wangsimni-ro, Seongdong-gu, Seoul 04763, Republic of Korea; Department of Neurology, Hanyang University Hospital, 222-1 Wangsimni-ro, Seongdong-gu, Seoul, 04763, Republic of Korea. Electronic address: hyumcbrain@hanyang.ac.kr.<br /><searchLink fieldCode="AU" term="%22Kim+JS%22">Kim JS</searchLink>; Clinical Research Institute, Konkuk University Medical Center, 120-1 Neungdong-ro, Gwangjin-gu, Seoul 05030, Republic of Korea; Institute of Biomedical Science and Technology, Konkuk University, 120-1 Neungdong-ro, Gwangjin-gu, Seoul 05030, Republic of Korea. Electronic address: junesic.kim@gmail.com.
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%229215515%22">NeuroImage</searchLink> [Neuroimage] 2026 Jan; Vol. 325, pp. 121674. <i>Date of Electronic Publication: </i>2025 Dec 23.
– Name: TypePub
  Label: Publication Type
  Group: TypPub
  Data: Journal Article
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: TitleSource
  Label: Journal Info
  Group: Src
  Data: <i>Publisher: </i><searchLink fieldCode="PB" term="%22Academic+Press%22">Academic Press </searchLink><i>Country of Publication: </i>United States <i>NLM ID: </i>9215515 <i>Publication Model: </i>Print-Electronic <i>Cited Medium: </i>Internet <i>ISSN: </i>1095-9572 (Electronic) <i>Linking ISSN: </i><searchLink fieldCode="IS" term="%2210538119%22">10538119 </searchLink><i>NLM ISO Abbreviation: </i>Neuroimage <i>Subsets: </i>MEDLINE
– Name: PublisherInfo
  Label: Imprint Name(s)
  Group: PubInfo
  Data: <i>Original Publication</i>: Orlando, FL : Academic Press, c1992-
– Name: SubjectMESH
  Label: MeSH Terms
  Group: Su
  Data: <searchLink fieldCode="MM" term="%22Cognitive+Dysfunction%22">Cognitive Dysfunction*</searchLink>/<searchLink fieldCode="MM" term="%22Cognitive+Dysfunction+classification%22">classification</searchLink> <br /><searchLink fieldCode="MM" term="%22Cognitive+Dysfunction%22">Cognitive Dysfunction*</searchLink>/<searchLink fieldCode="MM" term="%22Cognitive+Dysfunction+diagnostic+imaging%22">diagnostic imaging</searchLink> <br /><searchLink fieldCode="MM" term="%22Neural+Networks%2C+Computer%22">Neural Networks, Computer*</searchLink><br /><searchLink fieldCode="MH" term="%22Magnetic+Resonance+Imaging%22">Magnetic Resonance Imaging</searchLink>/<searchLink fieldCode="MH" term="%22Magnetic+Resonance+Imaging+methods%22">methods</searchLink> ; <searchLink fieldCode="MH" term="%22Alzheimer+Disease%22">Alzheimer Disease</searchLink>/<searchLink fieldCode="MH" term="%22Alzheimer+Disease+diagnostic+imaging%22">diagnostic imaging</searchLink> ; <searchLink fieldCode="MH" term="%22Brain%22">Brain</searchLink>/<searchLink fieldCode="MH" term="%22Brain+diagnostic+imaging%22">diagnostic imaging</searchLink> ; <searchLink fieldCode="MH" term="%22Humans%22">Humans</searchLink> ; <searchLink fieldCode="MH" term="%22Aged%22">Aged</searchLink> ; <searchLink fieldCode="MH" term="%22Male%22">Male</searchLink> ; <searchLink fieldCode="MH" term="%22Female%22">Female</searchLink> ; <searchLink fieldCode="MH" term="%22Deep+Learning%22">Deep Learning</searchLink> ; <searchLink fieldCode="MH" term="%22Aged%2C+80+and+over%22">Aged, 80 and over</searchLink> ; <searchLink fieldCode="MH" term="%22Middle+Aged%22">Middle Aged</searchLink> ; <searchLink fieldCode="MH" term="%22Convolutional+Neural+Networks%22">Convolutional Neural Networks</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Mild cognitive impairment (MCI), a precursor of Alzheimer's disease (AD), underscores the importance of early diagnosis and treatment. With an aging global population, AD prevalence is rising, necessitating more precise diagnostic methods. Deep learning technology shows promise for MCI and AD classification, but existing convolutional neural network (CNN) and graph attention network (GAT) models have limitations in capturing brain structural features and detecting microlesions. To address these issues, we propose a novel approach combining a CNN and modified GAT model to improve MCI classification. Magnetic resonance imaging volume data were analyzed using a CNN, whereas cortical thickness data were modeled using a GAT, leveraging their complementary strengths. Preprocessing involved extracting brain's structural features via the CIVET pipeline, and t-SNE was used to visualize the data's high-dimensional distribution. Final classification was performed using a multilayer perceptron, integrating feature vectors from both models. Performance evaluation metrics included the area under the curve (AUC), F1-score, sensitivity, and specificity. The combined CNN-GAT model outperformed existing single-model approaches, particularly in MCI classification, effectively distinguishing subtle variations between normal aging and MCI. The combined CNN-GAT model improved MCI classification performance by addressing the limitations of existing approaches. By capturing brain structural features and inter-regional relationships, it offers significant potential for advancing early diagnosis and treatment strategies for neurodegenerative diseases. Future efforts will focus on enhancing performance through additional data optimization.<br /> (Copyright © 2025 The Author(s). Published by Elsevier Inc. All rights reserved.)
– Name: Abstract
  Label: Competing Interests
  Group: Ab
  Data: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
– Name: SubjectMinor
  Label: Contributed Indexing
  Group:
  Data: <i>Keywords: </i>Convolutional neural network; Graph attention network; Mild cognitive impairment; Multilayer perceptron
– Name: DateEntry
  Label: Entry Date(s)
  Group: Date
  Data: <i>Date Created: </i>20251225 <i>Date Completed: </i>20260112 <i>Latest Revision: </i>20260112
– Name: DateUpdate
  Label: Update Code
  Group: Date
  Data: 20260130
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1016/j.neuroimage.2025.121674
– Name: AN
  Label: PMID
  Group: ID
  Data: 41448509
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=cmedm&AN=41448509
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1016/j.neuroimage.2025.121674
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        StartPage: 121674
    Subjects:
      – SubjectFull: Magnetic Resonance Imaging methods
        Type: general
      – SubjectFull: Alzheimer Disease diagnostic imaging
        Type: general
      – SubjectFull: Brain diagnostic imaging
        Type: general
      – SubjectFull: Humans
        Type: general
      – SubjectFull: Aged
        Type: general
      – SubjectFull: Male
        Type: general
      – SubjectFull: Female
        Type: general
      – SubjectFull: Deep Learning
        Type: general
      – SubjectFull: Aged, 80 and over
        Type: general
      – SubjectFull: Middle Aged
        Type: general
      – SubjectFull: Convolutional Neural Networks
        Type: general
      – SubjectFull: Cognitive Dysfunction classification
        Type: general
      – SubjectFull: Cognitive Dysfunction diagnostic imaging
        Type: general
      – SubjectFull: Neural Networks, Computer
        Type: general
    Titles:
      – TitleFull: A combined model of convolutional neural networks and graph attention networks for improved classification of mild cognitive impairment.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Kim N
      – PersonEntity:
          Name:
            NameFull: Jeon JY
      – PersonEntity:
          Name:
            NameFull: Seo J
      – PersonEntity:
          Name:
            NameFull: Lee Y
      – PersonEntity:
          Name:
            NameFull: Kim HJ
      – PersonEntity:
          Name:
            NameFull: Kim JS
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 01
              Text: 2026 Jan
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-electronic
              Value: 1095-9572
          Numbering:
            – Type: volume
              Value: 325
          Titles:
            – TitleFull: NeuroImage
              Type: main
ResultId 1