CoChat
Sign Up
Cited by 17Dementia and Cognitive Impairment ResearchAlzheimer's disease research and treatmentsArtificial Intelligence in Healthcare

Unsupervised machine learning to identify separable clinical Alzheimer’s disease sub-populations

Open access
Prakash, Jayant, Wang, Velda, Quinn, Robert E., Mitchell, Cassie S.·2021·Brain Sciences
DOI 10.3390/brainsci11080977PMC PMC8392842
Open this paper in CoChat — free accountOpen PDF

Abstract

Heterogeneity among Alzheimer’s disease (AD) patients confounds clinical trial patient selection and therapeutic efficacy evaluation. This work defines separable AD clinical sub-populations using unsupervised machine learning. Clustering (t-SNE followed by k-means) of patient features and association rule mining (ARM) was performed on the ADNIMERGE dataset from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). Patient sociodemographics, brain imaging, biomarkers, cognitive tests, and medication usage were included for analysis. Four AD clinical sub-populations were identified using between-cluster mean fold changes [cognitive performance, brain volume]: cluster-1 represented least severe disease [+17.3, +13.3]; cluster-0 [−4.6, +3.8] and cluster-3 [+10.8, −4.9] represented mid-severity sub-populations; cluster-2 represented most severe disease [−18.4, −8.4]. ARM assessed frequently occurring pharmacologic substances within the 4 sub-populations. No drug class was associated with the least severe AD (cluster-1), likely due to lesser antecedent disease. Anti-hyperlipidemia drugs associated with cluster-0 (mid-severity, higher volume). Interestingly, antioxidants vitamin C and E associated with cluster-3 (mid-severity, higher cognition). Anti-depressants like Zoloft associated with most severe disease (cluster-2). Vitamin D is protective for AD, but ARM identified significant underutilization across all AD sub-populations. Identification and feature characterization of four distinct AD sub-population “clusters” using standard clinical features enhances future clinical trial selection criteria and cross-study comparative analysis.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.3390/brainsci11080977

Cites

5 of 5 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Another study used patient data from the Alzheimer's Disease Neuroimaging Initiative ADNIMERGE dataset to identify potential drug repurposing candidates via association rule mining, but did not validate findings in another clinical dataset.

    “Still another study used patient data from the Alzheimer's Disease Neuroimaging Initiative ADNIMERGE dataset to identify potential drug repurposing candidates via association rule mining; however, the authors did not validate their findings in another clinical dataset (Prakash et al., 2021).”

    From Drug repurposing for Alzheimer’s disease from 2012–2022—a 10-year literature review (2023) · cites this paper as Prakash et al., 2021

    Methods §2.1 identifies ADNIMERGE as the ADNI dataset analyzed, and Methods §2.5 and Results §3.4 describe association rule mining of medication use. Discussion §4.5 calls for future testing in cohorts beyond ADNI, consistent with no validation in another clinical dataset. The paper identifies medication associations and discusses possible protective effects, but does not clearly identify drugs as repurposing candidates. (LLM, conf=0.83)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 43

  1. [1]Huber C.M. Yee C. May T. Dhanala A. Mitchell C.S. Cognitive decline in preclinical Alzheimer’s disease: Amyloid-beta versus tauopathy J. Alzheimers Dis. 2018 61 265 281 10.3233/JAD-170490 29154274 PMC5734131 · DOI
  2. [3]Weiner M.W. Veitch D.P. Aisen P.S. Beckett L.A. Cairns N.J. Cedarbaum J. Donohue M.C. Green R.C. Harvey D. Jack C.R. Jr. Impact of the Alzheimer’s disease neuroimaging initiative, 2004 to 2014 Alzheimers Dement. 2015 11 865 884 10.1016/j.jalz.2015.04.005 26194320 PMC4659407 · DOI
  3. [4]Ngwa J.S. Fungwe T.V. Ntekim O. Allard J.S. Johnson S.M. Castor C. Graham L. Nadarajah S. Gillum R.F. Obisesan T.O. Associations of pulse and blood pressure with hippocampal volume by APOE and cognitive phenotype: The Alzheimer’s Disease Neuroimaging Initiative (ADNI) Dement. Geriatr. Cogn. Disord. 2018 45 66 78 10.1159/000486955 29694964 PMC6143389 · DOI
  4. [7]Bredesen D.E. Metabolic profiling distinguishes three subtypes of Alzheimer’s disease Aging 2015 7 595 600 10.18632/aging.100801 26343025 PMC4586104 · DOI · Paper page
  5. [9]Mendez M.F. Early-onset Alzheimer disease Neurol. Clin. 2017 35 263 281 10.1016/j.ncl.2017.01.005 28410659 PMC5407192 · DOI
  6. [10]Lo M.T. Kauppi K. Fan C.C. Sanyal N. Reas E.T. Sundar V.S. Lee W.C. Desikan R.S. McEvoy L.K. Chen C.H. Identification of genetic heterogeneity of Alzheimer’s disease across age Neurobiol. Aging 2019 84 243.e1 243.e9 10.1016/j.neurobiolaging.2019.02.022 30979435 PMC6783343 · DOI
  7. [11]Silva M.V.F. Loures C.M.G. Alves L.C.V. de Souza L.C. Borges K.B.G. Carvalho M.D.G. Alzheimer’s disease: Risk factors and potentially protective measures J. Biomed. Sci. 2019 26 33 10.1186/s12929-019-0524-y 31072403 PMC6507104 · DOI
  8. [12]Nacmias B. Bagnoli S. Piaceri I. Sorbi S. Genetic heterogeneity of Alzheimer’s disease: Embracing research partnerships J. Alzheimers Dis. 2018 62 903 911 10.3233/JAD-170570 29103034 PMC5870047 · DOI
  9. [13]Armstrong R.A. Risk factors for Alzheimer’s disease Folia Neuropathol. 2019 57 87 105 10.5114/fn.2019.85929 31556570 · DOI
  10. [14]Mayeux R. Stern Y. Epidemiology of Alzheimer disease Cold Spring Harb. Perspect. Med. 2012 2 1 18 10.1101/cshperspect.a006239 PMC3405821 22908189 · DOI
  11. [15]Skoog I. Lernfelt B. Landahl S. Palmertz B. Andreasson L.A. Nilsson L. Persson G. Oden A. Svanborg A. 15-year longitudinal study of blood pressure and dementia Lancet 1996 347 1141 1145 10.1016/S0140-6736(96)90608-X 8609748 · DOI · Paper page
  12. [16]Hakansson K. Rovio S. Helkala E.L. Vilska A.R. Winblad B. Soininen H. Nissinen A. Mohammed A.H. Kivipelto M. Association between mid-life marital status and cognitive function in later life: Population based cohort study BMJ 2009 339 b2462 10.1136/bmj.b2462 19574312 PMC2714683 · DOI
  13. [17]Carroll J.C. Iba M. Bangasser D.A. Valentino R.J. James M.J. Brunden K.R. Lee V.M. Trojanowski J.Q. Chronic stress exacerbates tau pathology, neurodegeneration, and cognitive performance through a corticotropin-releasing factor receptor-dependent mechanism in a transgenic mouse model of tauopathy J. Neurosci. 2011 31 14436 14449 10.1523/JNEUROSCI.3836-11.2011 21976528 PMC3230070 · DOI
  14. [18]Proserpio P. Arnaldi D. Nobili F. Nobili L. Integrating sleep and Alzheimer’s disease pathophysiology: Hints for sleep disorders management J. Alzheimers Dis. 2018 63 871 886 10.3233/JAD-180041 29710720 · DOI
  15. [20]Vogt W. Nagel D. Cluster analysis in diagnosis Clin. Chem. 1992 38 182 198 10.1093/clinchem/38.2.182 1540999 · DOI
  16. [21]Nugent R. Meila M. An overview of clustering applied to molecular biology Methods Mol. Biol. 2010 620 369 404 20652512
  17. [22]Wiwie C. Baumbach J. Rottger R. Comparing the performance of biomedical clustering methods Nat. Methods 2015 12 1033 1038 10.1038/nmeth.3583 26389570 · DOI
  18. [23]Zhu X. Suk H.I. Shen D. Multi-modality canonical feature selection for Alzheimer’s disease diagnosis Med. Image. Comput. Comput. Assist. Interv. 2014 17 162 169 25485375 PMC4465097
  19. [24]Chen C.-H. A hybrid intelligent model of analyzing clinical breast cancer data using clustering techniques with feature selection Appl. Soft. Comput. 2014 20 4 14 10.1016/j.asoc.2013.10.024 · DOI
  20. [25]Nilashi M. Ibrahim O. Ahani A. Accuracy improvement for predicting Parkinson’s disease progression Sci. Rep. 2016 6 34181 10.1038/srep34181 27686748 PMC5043229 · DOI
  21. [26]Polat K. Classification of Parkinson’s disease using feature weighting method on the basis of fuzzy C-means clustering Int. J. Syst. Sci. 2012 43 597 609 10.1080/00207721.2011.581395 · DOI
  22. [27]Wu Y. Duan H. Du S. Multiple fuzzy c-means clustering algorithm in medical diagnosis Technol. Health Care 2015 23 S519 S527 10.3233/THC-150989 26410519 · DOI
  23. [28]Trevithick L. Painter J. Keown P. Mental health clustering and diagnosis in psychiatric in-patients BJPsych Bull. 2015 39 119 123 10.1192/pb.bp.114.047043 26191449 PMC4478925 · DOI
  24. [29]Yilmaz N. Inan O. Uzer M.S. A new data preparation method based on clustering algorithms for diagnosis systems of heart and diabetes diseases J. Med. Syst. 2014 38 48 10.1007/s10916-014-0048-7 24737307 · DOI
  25. [30]Nikas J.B. Low W.C. Application of clustering analyses to the diagnosis of Huntington disease in mice and other diseases with well-defined group boundaries Comput. Methods Programs Biomed. 2011 104 e133 e147 10.1016/j.cmpb.2011.03.004 21529982 PMC3166551 · DOI
  26. [31]Wu G. Kim M. Wang Q. Munsell B.C. Shen D. Scalable high-performance image registration framework by unsupervised deep feature representations learning IEEE Trans. Biomed. Eng. 2016 63 1505 1516 10.1109/TBME.2015.2496253 26552069 PMC4853306 · DOI
  27. [32]Sun L. Shao W. Wang M. Zhang D. Liu M. High-order feature learning for multi-atlas based label fusion: Application to brain segmentation with MRI IEEE Trans. Image Process. 2019 29 2702 2713 10.1109/TIP.2019.2952079 31725379 · DOI
  28. [33]Zhao X. Rangaprakash D. Yuan B. Denney T.S. Jr Katz J.S. Dretsch M.N. Deshpande G. Investigating the correspondence of clinical diagnostic grouping with underlying neurobiological and phenotypic clusters using unsupervised machine learning Front. Appl. Math. Stat. 2018 4 25 10.3389/fams.2018.00025 30393630 PMC6214192 · DOI
  29. [34]Gamberger D. Zenko B. Mitelpunkt A. Shachar N. Lavrac N. Clusters of male and female Alzheimer’s disease patients in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database Brain Inform. 2016 3 169 179 10.1007/s40708-016-0035-5 27525218 PMC4963443 · DOI
  30. [35]Mitelpunkt A. Galili T. Kozlovski T. Bregman N. Shachar N. Markus-Kalish M. Benjamini Y. Novel Alzheimer’s disease subtypes identified using a data and knowledge driven strategy Sci. Rep. 2020 10 1327 10.1038/s41598-020-57785-2 31992745 PMC6987140 · DOI
  31. [36]Schoenberg M.R. Dawson K.A. Duff K. Patton D. Scott J.G. Adams R.L. Test performance and classification statistics for the Rey Auditory Verbal Learning Test in selected clinical samples Arch. Clin. Neuropsychol. 2006 21 693 703 10.1016/j.acn.2006.06.010 16987634 · DOI
  32. [37]Bean J. Encyclopedia of Clinical Neuropsychology Kreutzer J.S. DeLuca J. Caplan B. Springer New York, NY, USA 2011 2174 2175
  33. [38]Ester M. Kriegel H.-P. Sander J. Xu X. A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise Proceedings of the Second International Conference on Knowledge Discovery and Data Mining Portland, OR, USA 2–4 August 1996 AAAI Press Portland, OR, USA 1996 226 231
  34. [41]Aisen P.S. Petersen R.C. Donohue M. Weiner M.W. Alzheimer’s Disease Neuroimaging Initiative Alzheimer’s Disease Neuroimaging Initiative 2 clinical core: Progress and plans Alzheimers Dement. 2015 11 734 739 10.1016/j.jalz.2015.05.005 26194309 PMC4643840 · DOI
  35. [42]Weiner M.W. Veitch D.P. Aisen P.S. Beckett L.A. Cairns N.J. Green R.C. Harvey D. Jack C.R. Jr. Jagust W. Morris J.C. The Alzheimer’s Disease Neuroimaging Initiative 3: Continued innovation for clinical trial improvement Alzheimers Dement. 2017 13 561 571 10.1016/j.jalz.2016.10.006 27931796 PMC5536850 · DOI
  36. [43]Chee C. Jaafar J. Aziz I.A. Hasan M.H. Yeoh W. Algorithms for frequent itemset mining: A literature review Artif. Intell. Rev. 2018 52 2603 2621 10.1007/s10462-018-9629-z · DOI
  37. [44]Wright A. Chen E.S. Maloney F.L. An automated technique for identifying associations between medications, laboratory results and problems J. Biomed. Inform. 2010 43 891 901 10.1016/j.jbi.2010.09.009 20884377 · DOI
  38. [45]Whitwell J.L. The protective role of brain size in Alzheimer’s disease Expert Rev. Neurother. 2010 10 1799 1801 10.1586/ern.10.168 21091311 PMC3920660 · DOI
  39. [46]Stampfer M.J. Cardiovascular disease and Alzheimer’s disease: Common links J. Intern. Med. 2006 260 211 223 10.1111/j.1365-2796.2006.01687.x 16918818 · DOI
  40. [47]Podcasy J.L. Epperson C.N. Considering sex and gender in Alzheimer disease and other dementias Dialogues Clin. Neurosci. 2016 18 437 446 28179815 PMC5286729
  41. [48]Mammone N. Ieracitano C. Adeli H. Bramanti A. Morabito F.C. Permutation Jaccard distance-based hierarchical clustering to estimate EEG network density modifications in MCI subjects IEEE Trans. Neural Netw. Learn. Syst. 2018 29 5122 5135 10.1109/TNNLS.2018.2791644 29994428 · DOI
  42. [49]Wingo A.P. Liu Y. Gerasimov E.S. Gockley J. Logsdon B.A. Duong D.M. Dammer E.B. Robins C. Beach T.G. Reiman E.M. Integrating human brain proteomes with genome-wide association data implicates new proteins in Alzheimer’s disease pathogenesis Nat. Genet. 2021 53 143 146 10.1038/s41588-020-00773-z 33510477 PMC8130821 · DOI · Paper page
  43. [50]Abrol A. Bhattarai M. Fedorov A. Du Y. Plis S. Calhoun V. Alzheimer’s Disease Neuroimaging Initiative Deep residual learning for neuroimaging: An application to predict progression to Alzheimer’s disease J. Neurosci. Methods 2020 339 108701 10.1016/j.jneumeth.2020.108701 32275915 PMC7297044 · DOI

Cited by

Sorted by citations

Authors
Prakash, J., Wang, V., Quinn, R. E., & Mitchell, C. S.
Affiliations
Georgia Institute of Technology (US)
Corresponding author
Mitchell, Cassie S.
ORCID
Venue
Brain Sciences
Publisher
MDPI AG
Published
2021-07-23
Volume
11
Issue
8
Pages
977
ISSN
2076-3425 (ISSN-L)
DOI
10.3390/brainsci11080977
PMC
PMC8392842
OpenAlex
W3183485908
Cited by
17
References
5
Citations by year
2026: 5 · 2025: 2 · 2024: 5 · 2023: 2 · 2022: 3
FWCI
1.79
Type
journal article
Language
en
Open access
Licence
Locations
  • Brain Sciences (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • Alzheimer's Association (AARG-2018-59104)
  • National Science Foundation (1944247)
  • Emory Alzheimer's Disease Research Center (n/a)
Topics
  • Health Sciences › Medicine › Psychiatry and Mental health › Dementia and Cognitive Impairment Research
  • Health Sciences › Medicine › Physiology › Alzheimer's disease research and treatments
  • Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
Keywords
Alzheimer's disease, patient heterogeneity, unsupervised clustering, association rule mining, cognitive performance, brain volume, disease severity, medication utilization, vitamin D, antioxidants, antidepressants, anti-hyperlipidemic drugs, patient selection for clinical trials

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review