Academic Journal
Προβολή σε EDS

Predicting conversion from mild cognitive impairment to Alzheimer's disease: A systematic review of deep learning models for early-stage disease classification.

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: Predicting conversion from mild cognitive impairment to Alzheimer's disease: A systematic review of deep learning models for early-stage disease classification.
Συγγραφείς: Ghasabi M; Faculty of Electrical and Computer Engineering, University of Tabriz, 51666616471, Tabriz, Iran., Aghdasi H; Faculty of Electrical and Computer Engineering, University of Tabriz, 51666616471, Tabriz, Iran. Electronic address: aghdasi@tabrizu.ac.ir., Feizi-Derakhshi MR; Faculty of Electrical and Computer Engineering, University of Tabriz, 51666616471, Tabriz, Iran.
Πηγή: Ageing research reviews [Ageing Res Rev] 2026 Nov; Vol. 122, pp. 103372. Date of Electronic Publication: 2026 Sep 11.
Τύπος έκδοσης: Journal Article; Systematic Review; Review
Γλώσσα: English
Στοιχεία περιοδικού: Publisher: Elsevier Science Country of Publication: England NLM ID: 101128963 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1872-9649 (Electronic) Linking ISSN: 15681637 NLM ISO Abbreviation: Ageing Res Rev Subsets: MEDLINE
Imprint Name(s): Original Publication: Oxford, UK : Elsevier Science, c2002-
Ιατρικοί όροι (MeSH): Alzheimer Disease*/diagnosis , Alzheimer Disease*/classification , Cognitive Dysfunction*/diagnosis , Cognitive Dysfunction*/classification , Deep Learning*, Humans ; Disease Progression ; Predictive Learning Models ; Early Diagnosis ; Classification Algorithms
Περίληψη: Introduction: Alzheimer's disease (AD) is a progressive neurodegenerative disorder for which early diagnosis-particularly the accurate prediction of conversion from mild cognitive impairment (MCI) to AD-is essential to enable timely and effective therapeutic interventions. Deep learning (DL) models have demonstrated substantial promise in this domain; however, critical challenges persist, including multiclass staging of disease progression, longitudinal data modeling, and effective multimodal data integration. This systematic review provides a critical appraisal of DL-based approaches for predicting MCI-to-AD conversion, with particular emphasis on these three key challenges.
Methods: This systematic review was designed and conducted in accordance with the PRISMA guidelines. A comprehensive literature search was performed across PubMed, Scopus, IEEE Xplore, and Web of Science for articles published between January 1, 2019, and February 20, 2026. Following rigorous screening of titles, abstracts, and full texts, 60 studies were included that employed deep learning models for AD stage classification and/or MCI-to-AD conversion prediction. Data were extracted and synthesized regarding study design, datasets, input modalities, DL architectures, and classification tasks.
Results: Cross-sectional approaches remained predominant (47 studies), while longitudinal designs were less common (13 studies) and showed heterogeneous but promising performance for MCI-to-AD conversion prediction. Heavy reliance on the ADNI dataset (42 studies) represents a major limitation to generalizability. Multimodal models were used in 24 studies and often reported strong performance, particularly for challenging MCI-related tasks; however, direct cross-study comparisons should be interpreted cautiously because of substantial heterogeneity in datasets, prediction tasks, validation strategies, and methodological quality. Convolutional neural networks (CNNs) dominated neuroimaging-based modeling (21 studies), while recurrent neural networks (RNNs) (4 studies) and transformers (4 studies) have emerged for capturing longitudinal dependencies and global relationships. The binary pMCI vs. sMCI classification proved the most challenging task (accuracy range: 71.71-96.3%), with performance declining as the number of classes increased in multiclass settings. Key limitations across studies include lack of diverse datasets, overfitting, and poor model interpretability.
Conclusion: Deep learning models hold considerable potential for predicting MCI-to-AD conversion, yet substantial barriers remain to their translation into routine clinical practice. Greater emphasis on longitudinal analysis, intelligent multimodal fusion, and interpretable architectures is essential for clinical impact. Future research should prioritize the development of diverse, multicenter datasets, advancement of explainable AI (XAI) techniques, and the design of personalized time-to-event models. This review offers a comprehensive roadmap to guide subsequent investigations toward more accurate, reliable, and clinically actionable diagnostic tools in Alzheimer's disease.
(Copyright © 2026 Elsevier B.V. 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: Alzheimer’s disease; Deep learning (DL); Longitudinal modeling; Mild cognitive impairment (MCI); Multiclass classification; Neuroimaging (MRI, PET)
Entry Date(s): Date Created: 20260911 Date Completed: 20261010 Latest Revision: 20261010
Update Code: 20261011
DOI: 10.1016/j.arr.2026.103372
PMID: 42727648
Βάση Δεδομένων: MEDLINE
FullText Text:
  Availability: 0
CustomLinks:
  – Url: https://www.doi.org/10.1016/j.arr.2026.103372?
    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=18729649&ISBN=&volume=122&issue=&date=20261101&spage=103372&pages=103372&title=Ageing research reviews&atitle=Predicting%20conversion%20from%20mild%20cognitive%20impairment%20to%20Alzheimer%27s%20disease%3A%20A%20systematic%20review%20of%20deep%20learning%20models%20for%20early-stage%20disease%20classification.&aulast=Ghasabi%20M&id=DOI:10.1016/j.arr.2026.103372
    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: 42727648
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Predicting conversion from mild cognitive impairment to Alzheimer's disease: A systematic review of deep learning models for early-stage disease classification.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AU" term="%22Ghasabi+M%22">Ghasabi M</searchLink>; Faculty of Electrical and Computer Engineering, University of Tabriz, 51666616471, Tabriz, Iran.<br /><searchLink fieldCode="AU" term="%22Aghdasi+H%22">Aghdasi H</searchLink>; Faculty of Electrical and Computer Engineering, University of Tabriz, 51666616471, Tabriz, Iran. Electronic address: aghdasi@tabrizu.ac.ir.<br /><searchLink fieldCode="AU" term="%22Feizi-Derakhshi+MR%22">Feizi-Derakhshi MR</searchLink>; Faculty of Electrical and Computer Engineering, University of Tabriz, 51666616471, Tabriz, Iran.
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22101128963%22">Ageing research reviews</searchLink> [Ageing Res Rev] 2026 Nov; Vol. 122, pp. 103372. <i>Date of Electronic Publication: </i>2026 Sep 11.
– Name: TypePub
  Label: Publication Type
  Group: TypPub
  Data: Journal Article; Systematic Review; Review
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: TitleSource
  Label: Journal Info
  Group: Src
  Data: <i>Publisher: </i><searchLink fieldCode="PB" term="%22Elsevier+Science%22">Elsevier Science </searchLink><i>Country of Publication: </i>England <i>NLM ID: </i>101128963 <i>Publication Model: </i>Print-Electronic <i>Cited Medium: </i>Internet <i>ISSN: </i>1872-9649 (Electronic) <i>Linking ISSN: </i><searchLink fieldCode="IS" term="%2215681637%22">15681637 </searchLink><i>NLM ISO Abbreviation: </i>Ageing Res Rev <i>Subsets: </i>MEDLINE
– Name: PublisherInfo
  Label: Imprint Name(s)
  Group: PubInfo
  Data: <i>Original Publication</i>: Oxford, UK : Elsevier Science, c2002-
– Name: SubjectMESH
  Label: MeSH Terms
  Group: Su
  Data: <searchLink fieldCode="MM" term="%22Alzheimer+Disease%22">Alzheimer Disease*</searchLink>/<searchLink fieldCode="MM" term="%22Alzheimer+Disease+diagnosis%22">diagnosis</searchLink> <br /><searchLink fieldCode="MM" term="%22Alzheimer+Disease%22">Alzheimer Disease*</searchLink>/<searchLink fieldCode="MM" term="%22Alzheimer+Disease+classification%22">classification</searchLink> <br /><searchLink fieldCode="MM" term="%22Cognitive+Dysfunction%22">Cognitive Dysfunction*</searchLink>/<searchLink fieldCode="MM" term="%22Cognitive+Dysfunction+diagnosis%22">diagnosis</searchLink> <br /><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="%22Deep+Learning%22">Deep Learning*</searchLink><br /><searchLink fieldCode="MH" term="%22Humans%22">Humans</searchLink> ; <searchLink fieldCode="MH" term="%22Disease+Progression%22">Disease Progression</searchLink> ; <searchLink fieldCode="MH" term="%22Predictive+Learning+Models%22">Predictive Learning Models</searchLink> ; <searchLink fieldCode="MH" term="%22Early+Diagnosis%22">Early Diagnosis</searchLink> ; <searchLink fieldCode="MH" term="%22Classification+Algorithms%22">Classification Algorithms</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Introduction: Alzheimer's disease (AD) is a progressive neurodegenerative disorder for which early diagnosis-particularly the accurate prediction of conversion from mild cognitive impairment (MCI) to AD-is essential to enable timely and effective therapeutic interventions. Deep learning (DL) models have demonstrated substantial promise in this domain; however, critical challenges persist, including multiclass staging of disease progression, longitudinal data modeling, and effective multimodal data integration. This systematic review provides a critical appraisal of DL-based approaches for predicting MCI-to-AD conversion, with particular emphasis on these three key challenges.<br />Methods: This systematic review was designed and conducted in accordance with the PRISMA guidelines. A comprehensive literature search was performed across PubMed, Scopus, IEEE Xplore, and Web of Science for articles published between January 1, 2019, and February 20, 2026. Following rigorous screening of titles, abstracts, and full texts, 60 studies were included that employed deep learning models for AD stage classification and/or MCI-to-AD conversion prediction. Data were extracted and synthesized regarding study design, datasets, input modalities, DL architectures, and classification tasks.<br />Results: Cross-sectional approaches remained predominant (47 studies), while longitudinal designs were less common (13 studies) and showed heterogeneous but promising performance for MCI-to-AD conversion prediction. Heavy reliance on the ADNI dataset (42 studies) represents a major limitation to generalizability. Multimodal models were used in 24 studies and often reported strong performance, particularly for challenging MCI-related tasks; however, direct cross-study comparisons should be interpreted cautiously because of substantial heterogeneity in datasets, prediction tasks, validation strategies, and methodological quality. Convolutional neural networks (CNNs) dominated neuroimaging-based modeling (21 studies), while recurrent neural networks (RNNs) (4 studies) and transformers (4 studies) have emerged for capturing longitudinal dependencies and global relationships. The binary pMCI vs. sMCI classification proved the most challenging task (accuracy range: 71.71-96.3%), with performance declining as the number of classes increased in multiclass settings. Key limitations across studies include lack of diverse datasets, overfitting, and poor model interpretability.<br />Conclusion: Deep learning models hold considerable potential for predicting MCI-to-AD conversion, yet substantial barriers remain to their translation into routine clinical practice. Greater emphasis on longitudinal analysis, intelligent multimodal fusion, and interpretable architectures is essential for clinical impact. Future research should prioritize the development of diverse, multicenter datasets, advancement of explainable AI (XAI) techniques, and the design of personalized time-to-event models. This review offers a comprehensive roadmap to guide subsequent investigations toward more accurate, reliable, and clinically actionable diagnostic tools in Alzheimer's disease.<br /> (Copyright © 2026 Elsevier B.V. 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>Alzheimer’s disease; Deep learning (DL); Longitudinal modeling; Mild cognitive impairment (MCI); Multiclass classification; Neuroimaging (MRI, PET)
– Name: DateEntry
  Label: Entry Date(s)
  Group: Date
  Data: <i>Date Created: </i>20260911 <i>Date Completed: </i>20261010 <i>Latest Revision: </i>20261010
– Name: DateUpdate
  Label: Update Code
  Group: Date
  Data: 20261011
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1016/j.arr.2026.103372
– Name: AN
  Label: PMID
  Group: ID
  Data: 42727648
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=cmedm&AN=42727648
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1016/j.arr.2026.103372
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        StartPage: 103372
    Subjects:
      – SubjectFull: Humans
        Type: general
      – SubjectFull: Disease Progression
        Type: general
      – SubjectFull: Predictive Learning Models
        Type: general
      – SubjectFull: Early Diagnosis
        Type: general
      – SubjectFull: Classification Algorithms
        Type: general
      – SubjectFull: Alzheimer Disease diagnosis
        Type: general
      – SubjectFull: Alzheimer Disease classification
        Type: general
      – SubjectFull: Cognitive Dysfunction diagnosis
        Type: general
      – SubjectFull: Cognitive Dysfunction classification
        Type: general
      – SubjectFull: Deep Learning
        Type: general
    Titles:
      – TitleFull: Predicting conversion from mild cognitive impairment to Alzheimer's disease: A systematic review of deep learning models for early-stage disease classification.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Ghasabi M
      – PersonEntity:
          Name:
            NameFull: Aghdasi H
      – PersonEntity:
          Name:
            NameFull: Feizi-Derakhshi MR
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 11
              Text: 2026 Nov
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-electronic
              Value: 1872-9649
          Numbering:
            – Type: volume
              Value: 122
          Titles:
            – TitleFull: Ageing research reviews
              Type: main
ResultId 1