Development of a screening algorithm for Alzheimer's disease using categorical verbal fluency

CERAD (Consortium to Establish a Registry for Alzheimer's Disease) Neuropsychology Assessment Battery: 35 Years and Counting

BACKGROUND

In 1986, the Consortium to Establish a Registry for Alzheimer's Disease (CERAD) was mandated to develop a brief neuropsychological assessment battery (CERAD-NAB) for AD, for uniform neuropsychological assessment, and information aggregation. Initially used across the National Institutes of Aging-funded Alzheimer's Disease Research Centers, it has become widely adopted wherever information is desired on cognitive status and change therein, particularly in older populations.

OBJECTIVE

Our purpose is to provide information on the multiple uses of the CERAD-NAB since its inception, and possible further developments.

METHODS

Since searching on "CERAD neuropsychological assessment battery" or similar terms missed important information, "CERAD" alone was entered into PubMed and SCOPUS, and CERAD-NAB use identified from the resulting studies. Use was sorted into major categories, e.g., psychometric information, norms, dementia/differential dementia diagnosis, epidemiology, intervention evaluation, genetics, etc., also translations, country of use, and alternative data gathering approaches.

RESULTS

CERAD-NAB is available in ∼20 languages. In addition to its initial purpose assessing AD severity, CERAD-NAB can identify mild cognitive impairment, facilitate differential dementia diagnosis, determine cognitive effects of naturally occurring and experimental interventions (e.g., air pollution, selenium in soil, exercise), has helped to clarify cognition/brain physiology-neuroanatomy, and assess cognitive status in dementia-risk conditions. Surveys of primary and tertiary care patients, and of population-based samples in multiple countries have provided information on prevalent and incident dementia, and cross-sectional and longitudinal norms for ages 35-100 years.

CONCLUSION

CERAD-NAB has fulfilled its original mandate, while its uses have expanded, keeping up with advances in the area of dementia.

Selecting the Most Important Features for Predicting Mild Cognitive Impairment from Thai Verbal Fluency Assessments

Mild cognitive impairment (MCI) is an early stage of cognitive decline or memory loss, commonly found among the elderly. A phonemic verbal fluency (PVF) task is a standard cognitive test that participants are asked to produce words starting with given letters, such as “F” in English and “ก” /k/ in Thai. With state-of-the-art machine learning techniques, features extracted from the PVF data have been widely used to detect MCI. The PVF features, including acoustic features, semantic features, and word grouping, have been studied in many languages but not Thai. However, applying the same PVF feature extraction methods used in English to Thai yields unpleasant results due to different language characteristics. This study performs analytical feature extraction on Thai PVF data to classify MCI patients. In particular, we propose novel approaches to extract features based on phonemic clustering (ability to cluster words by phonemes) and switching (ability to shift between clusters) for the Thai PVF data. The comparison results of the three classifiers revealed that the support vector machine performed the best with an area under the receiver operating characteristic curve (AUC) of 0.733 (N = 100). Furthermore, our implemented guidelines extracted efficient features, which support the machine learning models regarding MCI detection on Thai PVF data.

Reliability and Validity of Alzheimer's Disease Screening With a Semi-automated Smartphone Application Using Verbal Fluency

Introduction: This study aimed to examine the reliability and validity of Alzheimer's disease (AD) screening with a self-administered categorical verbal fluency test using a semi-automated Android application (app; tCVFT). Furthermore, its diagnostic accuracy concerning AD was compared with both that of a conventional categorical verbal fluency test (cCVFT) administered by a health professional and the Mini-Mental State Examination (MMSE).

Materials and Methods: Participants included 100 community-dwelling patients with early AD, whose Clinical Dementia Rating was either 0.5 or 1, and a further 100 sex-matched cognitively normal controls. The internal consistency and test-retest reliability of the tCVFT weighted sum score (tCVFT-WS) was examined using Cronbach's alpha and Pearson's correlation analyses (adjusted for age and education), respectively. The concurrent validity of the tCVFT-WS was examined by testing its correlations with the cCVFT weighted sum score (cCVFT-WS) and MMSE using Pearson's correlation tests. The diagnostic accuracies for early AD of the tCVFT-WS, cCVFT-WS, and MMSE were estimated and compared using receiver operating characteristic (ROC) analyses.

Results: The tCVFT-WS exhibited strong internal consistency (Cronbach's alpha = 0.79). However, its test-retest reliability was moderate (r = 0.54) owing to the low test-retest reliability of the second-half responses. The patient group exhibited a higher tCVFT-WS than the control group (p < 0.001). Correlations between the tCVFT-WS, cCVFT-WS, and MMSE were significant. The tCVFT-WS's area under the ROC curve for AD was 0.861. At its optimal cutoff, the sensitivity and specificity for AD were 0.78 and 0.77, respectively.

Conclusions: The self-administered tCVFT-WS, using an Android app, proved valid and reliable at distinguishing people with early AD from cognitively normal controls.

Deep learning based low-cost high-accuracy diagnostic framework for dementia using comprehensive neuropsychological assessment profiles

BACKGROUND

The conventional scores of the neuropsychological batteries are not fully optimized for diagnosing dementia despite their variety and abundance of information. To achieve low-cost high-accuracy diagnose performance for dementia using a neuropsychological battery, a novel framework is proposed using the response profiles of 2666 cognitively normal elderly individuals and 435 dementia patients who have participated in the Korean Longitudinal Study on Cognitive Aging and Dementia (KLOSCAD).

METHODS

The key idea of the proposed framework is to propose a cost-effective and precise two-stage classification procedure that employed Mini Mental Status Examination (MMSE) as a screening test and the KLOSCAD Neuropsychological Assessment Battery as a diagnostic test using deep learning. In addition, an evaluation procedure of redundant variables is introduced to prevent performance degradation. A missing data imputation method is also presented to increase the robustness by recovering information loss. The proposed deep neural networks (DNNs) architecture for the classification is validated through rigorous evaluation in comparison with various classifiers.

RESULTS

The k-nearest-neighbor imputation has been induced according to the proposed framework, and the proposed DNNs for two stage classification show the best accuracy compared to the other classifiers. Also, 49 redundant variables were removed, which improved diagnostic performance and suggested the potential of simplifying the assessment. Using this two-stage framework, we could get 8.06% higher diagnostic accuracy of dementia than MMSE alone and 64.13% less cost than KLOSCAD-N alone.

CONCLUSION

The proposed framework could be applied to general dementia early detection programs to improve robustness, preciseness, and cost-effectiveness.