Explainable Machine Learning-Based Classification of Clinical Dementia Ratings Using a Tablet-Based Cognitive Assessment

Topic: Risk, prevention and care in population studies

Dementia is a substantial public health issue with increasing prevalence globally. Fast and accurate diagnosis is vital for successful management and intervention. This article presents a new machine-learning model that uses data from a tablet computer cognitive test to label clinical dementia ratings (CDR). Traditional cognitive assessments often depend on subjective evaluations, which may introduce bias and limit applicability across diverse populations. We evaluated five different machine-learning algorithms for dementia classification, integrating advanced feature extraction techniques, and to enhance the interpretability of our model, we employed explainable artificial intelligence (XAI) techniques, specifically local interpretable model-agnostic explanations (LIME) and Shapley Additive exPlanations (SHAP) with ensemble methods to capture subtle cognitive deficits and these methods elucidate the contributions of individual features to the model’s predictions, fostering trust and transparency in clinical applications. Area Under the Receiver-Operating Characteristic (AUROC) is used to measure the model’s ability to distinguish between classes across various threshold settings. Among different ML models, the deep learning model demonstrated superior predictive performance, achieving an accuracy of 100% and 95.8% for training and testing, respectively. Additionally, it attained area under the receiver-operating characteristic (AUROC) scores of 1.00 (95% CI 1.00–1.00) for training and 0.98 (95% CI 0.92–1.00) for testing, indicating strong discrimination between different CDR levels. Our analysis identified features, such as delay, mini-mental state examination (MMSE) score, gender, yong_overtimes, and mouse_times as the top five predictors. On the other hand, angle degree, spiral tremble, and yong tremble were the least predictive features. These findings are consistent with the previous literature, highlighting the importance of these features in the assessment of dementia. Our approach illustrates the promise of fusing state-of-the-art machine-learning algorithms with explainability methods to enhance the accuracy and interpretability of dementia classification. Such integration is central to the development of trustworthy digital tools that can support early diagnosis and management of dementia, eventually leading to improved patient outcomes.

Key findings

• Among different ML models, the deep learning model demonstrated superior predictive performance, achieving an accuracy of 100% and 95.8% for training and testing, respectively.
• These findings are consistent with the previous literature, highlighting the importance of these features in the assessment of dementia.

Why it matters

This study developed a machine-learning model using tablet-based cognitive test data to classify clinical dementia ratings, with a deep learning model achieving high accuracy. It concerns automated classification for diagnosis and does not test any treatment, cure or prevention.

New to this subject? Start with our plain-English guide: Getting a diagnosis →

Study details

• Effect size95.8%
• Evidence strengthLow

Read the original

• DOI: 10.1007/s44196-025-01019-8
• Original source →
• Free full text: International Journal of Computational Intelligence Systems →

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

@article{ningthoujam2025-661d0e63,
  author  = {Ningthoujam, Avichandra Singh and Huang, Yo-Ping and Thiyagarajan, Gomathi and Sharma, Shilpa},
  title   = {{Explainable Machine Learning-Based Classification of Clinical Dementia Ratings Using a Tablet-Based Cognitive Assessment}},
  journal = {International Journal of Computational Intelligence Systems},
  year    = {2025},
  doi     = {10.1007/s44196-025-01019-8},
  url     = {https://doi.org/10.1007/s44196-025-01019-8}
}

Studies this one cites

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

• Interpretable machine learning for dementia: A systematic review Alzheimer's & dementia : the journal of the Alzheimer's Association · 2023-02-03 · usefulness score 66/100 · Digital health tools
• Machine Learning for Dementia Prediction: A Systematic Review and Future Research Directions Journal of medical systems · 2023-02-01 · usefulness score 62/100 · Digital health tools
• Generalizable deep learning model for early Alzheimer's disease detection from structural MRIs Scientific reports · 2022-10-17 · usefulness score 54/100 · Long-term cohort studies
• Screening for dementia in primary care: a review of the use, efficacy and quality of measures International psychogeriatrics · 2008-06-05 · usefulness score 44/100 · Cognitive tests and assessment

Related studies in Getting a diagnosis

• A Machine Learning-Based Framework for differential screening of Neurodegenerative Dementias using routine clinical data Research Square · 2026-10-07 · usefulness score 16/100
• Machine Learning–Based Cognitive Assessment With The Autonomous Cognitive Examination: Randomized Controlled Trial Journal of Medical Internet Research · 2025-07-30 · usefulness score 47/100
• A chart-based method for identification of delirium: validation compared with interviewer ratings using the confusion assessment method Journal of the American Geriatrics Society · 2005-02-01 · usefulness score 49/100
• Clinical Utility and Diagnostic Accuracy of the Tablet-Based Seoul Cognitive Status Test: Evidence for Scalable Cognitive Assessment Dementia and Neurocognitive Disorders · 2025-01-01 · usefulness score 43/100
• Eye-tracking-based tablet application for enhanced cognitive assessment: a clinical trial for software as a medical device (SaMD) GeroScience · 2025-10-20 · usefulness score 35/100
• Investigating dementia patterns by machine learning and graph-based marker analysis for multiclass dementia diagnosis Discover Computing · 2026-05-18 · usefulness score 35/100

Similar studies

• Leveraging Explainable AI for Dementia Classification: A Machine Learning Approach 2025 · Digital health tools
• Machine learning algorithms and traditional statistical models for detection of dementia: a population-based study 2025 · Digital health tools
• An explainable machine learning approach for Alzheimer's disease classification 2024 · Digital health tools
• Explainable machine learning for predicting longitudinal dementia status: Establishing a leakage-free benchmark 2026 · Digital health tools
• Investigating the Utility of Explainable Artificial Intelligence for Neuroimaging‐Based Dementia Diagnosis and Prognosis 2026 · Digital health tools
• Developing an explainable machine learning and fog computing-based visual rating scale for the prediction of dementia progression 2025 · Digital health tools

More on Getting a diagnosis

• A review that covers this ground The Impact of Mild Cognitive Impairment on Gait and Balance: A Systematic Review and Meta-Analysis of Studies Using Instrumented Assessment Gerontology · 2016-05-13 · usefulness score 64/100
• Newest in this subject The Contribution of Routinely Collected Clinical Data to Dementia Identification in Primary Care: A Cross-Sectional Study Research Square · 2026-10-28 · Preprint — not yet peer-reviewed · usefulness score 16/100
• Newest in this subject Beyond “Denial”: Understanding and Responding to Help-Refusal in Dementia Preprints.org · 2026-10-26 · Preprint — not yet peer-reviewed · usefulness score 16/100

See all studies in Getting a diagnosis →