Dataset Information

The Reliability and Clinical Validation of Automatically-Derived Verbal Memory Features of the Verbal Learning Test in Early Diagnostics of Cognitive Impairment.

ABSTRACT:

Background

Previous research has shown that verbal memory accurately measures cognitive decline in the early phases of neurocognitive impairment. Automatic speech recognition from the verbal learning task (VLT) can potentially be used to differentiate between people with and without cognitive impairment.

Objective

Investigate whether automatic speech recognition (ASR) of the VLT is reliable and able to differentiate between subjective cognitive decline (SCD) and mild cognitive impairment (MCI).

Methods

The VLT was recorded and processed via a mobile application. Following, verbal memory features were automatically extracted. The diagnostic performance of the automatically derived features was investigated by training machine learning classifiers to distinguish between participants with SCD versus MCI/dementia.

Results

The ICC for inter-rater reliability between the clinical and automatically derived features was 0.87 for the total immediate recall and 0.94 for the delayed recall. The full model including the total immediate recall, delayed recall, recognition count, and the novel verbal memory features had an AUC of 0.79 for distinguishing between participants with SCD versus MCI/dementia. The ten best differentiating VLT features correlated low to moderate with other cognitive tests such as logical memory tasks, semantic verbal fluency, and executive functioning.

Conclusions

The VLT with automatically derived verbal memory features showed in general high agreement with the clinical scoring and distinguished well between SCD and MCI/dementia participants. This might be of added value in screening for cognitive impairment.

SUBMITTER: Possemis N

PROVIDER: S-EPMC10789344 | biostudies-literature | 2024

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

The Reliability and Clinical Validation of Automatically-Derived Verbal Memory Features of the Verbal Learning Test in Early Diagnostics of Cognitive Impairment.

Possemis Nina N Ter Huurne Daphne D Banning Leonie L Gruters Angelique A Van Asbroeck Stephanie S König Alexandra A Linz Nicklas N Tröger Johannes J Langel Kai K Blokland Arjan A Prickaerts Jos J de Vugt Marjolein M Verhey Frans F Ramakers Inez I

Journal of Alzheimer's disease : JAD 20240101 1

<h4>Background</h4>Previous research has shown that verbal memory accurately measures cognitive decline in the early phases of neurocognitive impairment. Automatic speech recognition from the verbal learning task (VLT) can potentially be used to differentiate between people with and without cognitive impairment.<h4>Objective</h4>Investigate whether automatic speech recognition (ASR) of the VLT is reliable and able to differentiate between subjective cognitive decline (SCD) and mild cognitive imp ...[more]

PMID: 38108348

Similar Datasets

Project description:BackgroundThe ability to detect one's own memory capacity and develop strategies based on daily contexts is important for daily activities. The Contextual Memory Test (CMT) assesses self-awareness, self-efficacy, self-perception/evaluation of performance, recall, and strategy use that are associated with daily contexts, and could be a potentially suitable measurement for assessing memory and meta-memory in older adults with and without cognitive impairment. Nevertheless, the test-retest reliability and minimal detectable change (MDC) remain unknown in these individuals.ObjectiveThe purpose of this study was to examine test-retest reliability and calculate MDC of the CMT in healthy older adults and those with mild cognitive impairment (MCI).MethodsEighty-three participants completed the CMT twice with a one-month interval. Test-retest reliability was examined using intraclass correlation coefficient (ICC) in all seven domains of the CMT and the recognition subtest. The standard error of measurement (SEM) and MDC were calculated. The Bland-Altman analysis was performed to evaluate the degree of agreement between measurements.ResultsThe ICC of five domains (self-awareness, self-perception/evaluation of performance, immediate/delayed/total recall) and the recognition subtest were good to excellent (ICC = 0.63-0.94) in healthy and MCI participants and the MDC% were less than 30% The ICC of the other two domains (self-efficacy and total strategy use, TSS) were low (ICC = 0.07-0.59) and the MDC% exceeded 30%. The Bland-Altman analysis showed generally better performance in the 2nd than the 1st measurement in most CMT domains.ConclusionsOur results revealed sufficient test-retest reliability and acceptable MDC in most CMT domains in healthy and MCI participants. Only the self-efficacy and TSS domains demonstrated low ICC and large MDC. Possible practice effects were found between repeated measurements. Clinicians should be cautious when evaluating self-efficacy and strategy use using the CMT in older adults. Further improvements are needed for these two domains.

Project description:RationaleHigh frequency oscillations (HFO; ripples = 80-200, fast ripples 200-500 Hz) are promising epileptic biomarkers in patients with epilepsy. However, especially in temporal epilepsies differentiation of epileptic and physiological HFO activity still remains a challenge. Physiological sleep-spindle-ripple formations are known to play a role in slow-wave-sleep memory consolidation. This study aimed to find out if higher rates of mesial-temporal spindle-ripples correlate with good memory performance in epilepsy patients and if surgical removal of spindle-ripple-generating brain tissue correlates with a decline in memory performance. In contrast, we hypothesized that higher rates of overall ripples or ripples associated with interictal epileptic spikes correlate with poor memory performance.MethodsPatients with epilepsy implanted with electrodes in mesial-temporal structures, neuropsychological memory testing and subsequent epilepsy surgery were included. Ripples and epileptic spikes were automatically detected in intracranial EEG and sleep-spindles in scalp EEG. The coupling of ripples to spindles was automatically analyzed. Mesial-temporal spindle-ripple rates in the speech-dominant-hemisphere (left in all patients) were correlated with verbal memory test results, whereas ripple rates in the non-speech-dominant hemisphere were correlated with non-verbal memory test performance, using Spearman correlation).ResultsIntracranial EEG and memory test results from 25 patients could be included. All ripple rates were significantly higher in seizure onset zone channels (p < 0.001). Patients with pre-surgical verbal memory impairment had significantly higher overall ripple rates in left mesial-temporal channels than patients with intact verbal memory (Mann-Whitney-U-Test: p = 0.039). Spearman correlations showed highly significant negative correlations of the pre-surgical verbal memory performance with left mesial-temporal spike associated ripples (rs = -0.458; p = 0.007) and overall ripples (rs = -0.475; p = 0.006). All three ripple types in right-sided mesial-temporal channels did not correlate with pre-surgical nonverbal memory. No correlation for the difference between post- and pre-surgical memory and pre-surgical spindle-ripple rates was seen in patients with left-sided temporal or mesial-temporal surgery.DiscussionThis study fails to establish a clear link between memory performance and spindle ripples. This highly suggests that spindle-ripples are only a small portion of physiological ripples contributing to memory performance. More importantly, this study indicates that spindle-ripples do not necessarily compromise the predictive value of ripples in patients with temporal epilepsy. The majority of ripples were clearly linked to areas with poor memory function.

Dataset Information

The Reliability and Clinical Validation of Automatically-Derived Verbal Memory Features of the Verbal Learning Test in Early Diagnostics of Cognitive Impairment.

Background

Objective

Methods

Results

Conclusions

Publications

The Reliability and Clinical Validation of Automatically-Derived Verbal Memory Features of the Verbal Learning Test in Early Diagnostics of Cognitive Impairment.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets