Related Experiment Video
Updated: Apr 17, 2026

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
Three screening methods for cognitive dysfunction using the Mini-Mental State Examination and Korean Dementia
Seong Hye Choi1, Moon Ho Park2
1Department of Neurology, Inha University College of Medicine, Incheon, Korea.
This study compared three cognitive screening methods for detecting cognitive dysfunction. The combination of Mini-Mental State Examination (MMSE) and Korean dementia screening questionnaire (KDSQ) showed the best performance, especially using a decision tree approach for higher accuracy.
Area of Science:
- Neurology
- Geriatrics
- Psychometrics
Background:
- Cognitive dysfunction screening is crucial for early diagnosis and management.
- Traditional cognitive tests and informant reports may not always be accessible in clinical settings.
- Investigating alternative screening methods is essential for broader application.
Purpose of the Study:
- To evaluate and compare three distinct screening methods for cognitive dysfunction.
- To assess the efficacy of the Mini-Mental State Examination (MMSE) alone, a combined MMSE and Korean dementia screening questionnaire (KDSQ) approach, and a decision tree model.
- To determine the optimal screening strategy based on diagnostic accuracy and sensitivity.
Main Methods:
- A cohort of 2861 dementia patients, 3519 mild cognitive impairment patients, and 1375 controls were recruited.
- Three screening methods were tested: MMSE alone, MMSE+KDSQ (conventional cut-off), and MMSE+KDSQ (decision tree).
- Diagnostic performance was evaluated using receiver operating characteristic curves and classification accuracy.
Main Results:
- The MMSE+KDSQ (cut-off) method demonstrated the highest accuracy in discriminating dementia from controls (AUC 0.899) and any cognitive dysfunction from controls (AUC 0.784).
- The MMSE alone method showed the highest accuracy for discriminating mild cognitive impairment from controls (AUC 0.683).
- The MMSE+KDSQ (decision tree) method achieved the highest overall classification accuracy (70.0%) and sensitivity across all discriminations.
Conclusions:
- The three evaluated screening methods possess distinct strengths for identifying and staging cognitive dysfunction.
- The choice of screening method can be tailored to specific clinical settings and situational requirements.
- The MMSE+KDSQ decision tree approach offers a highly sensitive and accurate option for cognitive screening.
More Related Videos
04:33Author Spotlight: Unveiling the Connection Between Sleep Disorders and Cognitive Symptoms in Depression
Published on: April 26, 2024
06:23The 4 Mountains Test: A Short Test of Spatial Memory with High Sensitivity for the Diagnosis of Pre-dementia Alzheimer's Disease
Published on: October 13, 2016
Related Concept Videos
Dementia
The progression of dementia is generally gradual....
Dementia l: Introduction