Enhanced diagnostic interpretation of the MoCA using machine learning

Topic: Risk, prevention and care in population studies

Machine learning models consistently outperformed conventional approach. For the global detection of cognitive impairment, XGBoost achieved the best performance (Youden Index 0.61 vs. 0.54 for the standard cutoff). Regarding subtype classification, models demonstrated variable discriminative capacity depending on clinical homogeneity: primary progressive aphasia was best classified (Youden ≈ 0.77), followed by Lewy body dementia and Alzheimer's disease, while vascular dementia remained more challenging to isolate. Feature importance analysis highlighted the Cognitive Quotient as a robust universal predictor, while pinpointing disease-specific drivers such as delayed recall for Alzheimer's disease and verbal fluency for primary progressive aphasia. Our findings suggest interpretable machine learning enhances diagnostic utility of the MoCA, yielding superior accuracy compared to a fixed cutoff. By synthesizing individualized subtest profiles within a transparent framework, this approach offers a clinically actionable solution. It transforms the MoCA from a simple screening tool to a precision diagnostic aid, optimizing patient triage in the era of disease-modifying therapies.

Key findings

• Machine learning models consistently outperformed conventional approach.
• For the global detection of cognitive impairment, XGBoost achieved the best performance (Youden Index 0.61 vs. 0.54 for the standard cutoff).
• Regarding subtype classification, models demonstrated variable discriminative capacity depending on clinical homogeneity: primary progressive aphasia was best classified (Youden ≈ 0.77), followed by Lewy body dementia and Alzheimer's disease, while vascular dementia remained more challenging to isolate.
• Feature importance analysis highlighted the Cognitive Quotient as a robust universal predictor, while pinpointing disease-specific drivers such as delayed recall for Alzheimer's disease and verbal fluency for primary progressive aphasia.
• Our findings suggest interpretable machine learning enhances diagnostic utility of the MoCA, yielding superior accuracy compared to a fixed cutoff.

Why it matters

This study reports that machine learning models using MoCA test data performed better than the standard cutoff for detecting cognitive impairment and classifying subtypes. It is a diagnostic aid only and does not test any treatment, cure or prevention.

New to this subject? Start with our plain-English guide: Digital health →

Study details

• Participants38,746
• Evidence strengthLow

Read the original

• DOI: 10.3389/fnins.2026.1679649
• PubMed: 41799879
• Original source →
• Free full text: Frontiers in Neuroscience →

Full text: free to read — OpenAlex lists this copy as free to read: the link above (checked 6 Feb 1020).

@article{gourdeau2026-41799879,
  author  = {Gourdeau, Christian and Gourdeau, Charles L. and Bernier, Patrick J. and Laforce, Robert Jr},
  title   = {{Enhanced diagnostic interpretation of the MoCA using machine learning}},
  journal = {Frontiers in Neuroscience},
  year    = {2026},
  doi     = {10.3389/fnins.2026.1679649},
  pmid    = {41799879},
  url     = {https://doi.org/10.3389/fnins.2026.1679649}
}

Studies this one cites

Referenced by this study and also in the alzhub library, so you can read them here.

• Donanemab in Early Symptomatic Alzheimer Disease: The TRAILBLAZER-ALZ 2 Randomized Clinical Trial JAMA · 2023-07-17 · usefulness score 74/100 · Clinical trials
• Performance of Machine Learning Algorithms for Predicting Progression to Dementia in Memory Clinic Patients JAMA network open · 2021-12-01 · usefulness score 66/100 · Digital health tools
• Classification of primary progressive aphasia and its variants Neurology · 2011-02-16 · usefulness score 64/100 · Getting a diagnosis
• Development and validation of an interpretable deep learning framework for Alzheimer's disease classification Brain : a journal of neurology · 2020-06-01 · usefulness score 63/100 · Prevention and risk
• MMSE and MoCA in Parkinson's disease and dementia with Lewy bodies: a multicenter 1-year follow-up study Journal of neural transmission (Vienna, Austria : 1996) · 2016-02-06 · usefulness score 62/100 · Cognitive tests and assessment
• The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment Journal of the American Geriatrics Society · 2005-04-01 · usefulness score 60/100 · Getting a diagnosis
• Age- and education-adjusted normative data for the Montreal Cognitive Assessment (MoCA) in older adults age 70-99 Neuropsychology, development, and cognition. Section B, Aging, neuropsychology and cognition · 2015-05-05 · usefulness score 47/100 · Cognitive tests and assessment
• Predicting dementia from spontaneous speech using large language models PLOS digital health · 2022-12-22 · usefulness score 46/100 · Digital health tools
• Optimizing Neuropsychological Assessments for Cognitive, Behavioral, and Functional Impairment Classification: A Machine Learning Study Behavioural neurology · 2017-01-31 · usefulness score 44/100 · Digital health tools
• Downstream Revenue Generated After Visits to Neurologists: A Claims-Based Analysis of Medicare Data Neurology. Clinical practice · 2025-08-11 · usefulness score 40/100 · Health systems and care delivery

Related studies in Digital health tools

• Enhanced Multi-Model Machine Learning-Based Dementia Detection Using a Data Enrichment Framework: Leveraging the Blessing of Dimensionality Bioengineering · 2025-05-30 · usefulness score 34/100
• Enhanced in silico QSAR-based screening of butyrylcholinesterase inhibitors using multi-feature selection and machine learning SAR and QSAR in environmental research · 2025-02-01 · usefulness score 34/100
• Early diagnosis of Alzheimer's disease using machine learning: a multi-diagnostic, generalizable approach Alzheimer's research & therapy · 2022-08-03 · usefulness score 57/100
• Diagnostic performance of actigraphy in Alzheimer’s disease using a machine learning classifier – a cross-sectional memory clinic study Alzheimer s Research & Therapy · 2025-05-21 · usefulness score 51/100
• Developing machine learning-enhanced WHODAS 2.0 short forms for persons with dementia Journal of the Formosan Medical Association · 2026-05-01 · usefulness score 53/100
• Machine learning–enhanced screening funnel for clinical trials in Alzheimer's disease Alzheimer s & Dementia Translational Research & Clinical Interventions · 2025-04-01 · usefulness score 33/100

Similar studies

• Normative data of the Spanish version of the Montreal Cognitive Assessment (MoCA) in older individuals from Peru 2025 · Getting a diagnosis
• A Usability Study of MoCA Solo, a Digital AI‐Assisted Version of the MoCA Test 2025 · Cognitive tests and assessment
• Montreal Cognitive Assessment (MoCA) Scale: Strengths, Limitations, and Implication for Clinical Practice 2025 · Getting a diagnosis
• Screening utility of the Montreal Cognitive Assessment (MoCA): in place of--or as well as--the MMSE? 2012 · Cognitive tests and assessment
• Montreal Cognitive Assessment (MoCA)-Italian version: regression based norms and equivalent scores 2015 · Cognitive tests and assessment
• Montreal Cognitive Assessment (MoCA): validation study for vascular dementia 2012 · Cognitive tests and assessment

More on Digital health tools

• A review that covers this ground Deep Learning in Alzheimer's Disease: Diagnostic Classification and Prognostic Prediction Using Neuroimaging Data Frontiers in aging neuroscience · 2019-08-20 · usefulness score 72/100
• Newest in this subject Speaker-Wise Dementia Screening fromNaturalistic Speech with Lexical Entropy, Semantic Drift, and Multimodal Fusion Research Square · 2026-10-21 · Preprint — not yet peer-reviewed · usefulness score 16/100
• Newest in this subject Integrating Multisource Target Evidence with Machine-Learning Potency Prediction for Alzheimer’s Disease Drug Discovery bioRxiv · 2026-10-20 · Preprint — not yet peer-reviewed · usefulness score 16/100

See all studies in Digital health tools →