Detection of cognitive decline using a single-channel EEG with an interactive assessment tool

Abstract

Background: Cognitive decline remains highly underdiagnosed in the community despite extensive efforts to find novel biomarkers to detect it. Finding objective screening tools for cognitive decline may improve early diagnosis of Alzheimer′s disease (AD) in the community. EEG biomarkers based on machine learning (ML) may offer a noninvasive low-coast approach for identifying cognitive decline with clinically useful accuracy. However, most of the studies use multi-electrode systems which are not vastly accessible. This study aims to evaluate the ability to extract cognitive decline biomarkers using a wearable single-channel EEG system with a short interactive cognitive assessment tool. Methods: Seniors in different clinical stages of cognitive decline (healthy to mild dementia, n=60) and young healthy participants (n=22) performed a cognitive assessment which included auditory detection and resting state tasks, while being recorded with a single-channel EEG (Aurora by Neurosteer®). Seniors′ MMSE scores were obtained by clinicians and used in allocating the groups (Healthy: MMSE>28; MCI-R: 28>MMSE>24; and MD: MMSE<24). Data analysis included standard frequency bands as well as three novel biomarkers, A0, ST4 and VC9, previously extracted from a different dataset to minimize overfitting risks. Results: Correlation between MMSE scores and reaction times was significant, validating the cognitive assessment tool. Individual MMSE scores correlated significantly with two of the EEG biomarkers: A0 and ST4. Furthermore, A0 and ST4 showed significant separation between groups of seniors with high vs. low MMSE scores, as well as the healthy young group. ST4 separated between the healthy groups (young and seniors) and the low MMSE (MD) group. Conversely, A0 differentiated between the healthy young group and all three groups of seniors. In the healthy young group, activity of Theta band and VC9 biomarker significantly increased with higher cognitive load, with both separating between the high cognitive load task and resting state. Furthermore, VC9 showed a finer separation between high and low cognitive load levels within the cognitive task. This was not shown in the senior groups, suggesting VC9 may be indicative to cognitive decline in the senior population. Conclusions: These results introduce novel biomarkers which potentially detect cognitive decline, obtained by a wearable single-channel EEG with a short interactive cognitive assessment. Such objective screening tools can be used on a large scale to detect cognitive decline and potentially allow early diagnosis of AD in every clinic.

Reduced neural activity during volatile anesthesia compared to TIVA: evidence from a novel EEG signal processing analysis

“…Next, for data analysis, higher-order features which were previously extracted (39) were provided with a second-by-second activity. These were previously shown to correlate with cognitive load (40)(41)(42) and cognitive decline severity (43).…”

“…VC9 also correlated with cognitive load and individual performance of medical interns who were undergoing a surgery simulation task (40). A0 and ST4 activity correlated with cognitive decline scores (i.e., mini-mental score, 54) and separated between groups with different levels of cognitive impairment (i.e., mild, moderate ,and severe) (39,43). ST4 correlated with individual performance for healthy participants undergoing the n-back task (39,41) and also showed some…”

“…On separate datasets, these features were found to correlate with different tasks/levels/states. For example, VC9 correlated with cognitive load using the n-back task, auditory detection task, and interruptions paradigm (39)(40)(41)(42)(43). VC9 also correlated with cognitive load and individual performance of medical interns who were undergoing a surgery simulation task (40).…”

“…For details on creating the higher-level features VC9, A0, and ST4, see Appendix A and (39)(40)(41). Validation of the cognitive properties of these features on healthy individuals appears in (39,42), and usage in different contexts appears in (40,41,43).…”

“…usefulness in differentiating between cognitive load levels (39,43). Neurosteer® provided the real-time activity of the BAFs and a dataset with a sample-per-second activity of the EEG bands (i.e., delta (0.5-4 Hz), theta (4-7 Hz), alpha (8-15 Hz), and beta (16-31 Hz)), and features (i.e., VC9, A0, and ST4).…”