Harnessing the Power of Voice: A Deep Neural Network Model for Alzheimer's Disease Detection.

Background and Purpose: Voice, reflecting cerebral functions, holds potential for analyzing and understanding brain function, especially in the context of cognitive impairment (CI) and Alzheimer's disease (AD). This study used voice data to distinguish between normal cognition and CI or Alzheimer's disease dementia (ADD). Methods: This study enrolled 3 groups of subjects: 1) 52 subjects with subjective cognitive decline; 2) 110 subjects with mild CI; and 3) 59 subjects with ADD. Voice features were extracted using Mel-frequency cepstral coefficients and Chroma. Results: A deep neural network (DNN) model showed promising performance, with an accuracy of roughly 81% in 10 trials in predicting ADD, which increased to an average value of about 82.0%±1.6% when evaluated against unseen test dataset. Conclusions: Although results did not demonstrate the level of accuracy necessary for a definitive clinical tool, they provided a compelling proof-of-concept for the potential use of voice data in cognitive status assessment. DNN algorithms using voice offer a promising approach to early detection of AD. They could improve the accuracy and accessibility of diagnosis, ultimately leading to better outcomes for patients.

qEEG as Biomarker for Alzheimer's Disease: Investigating Relative PSD Difference and Coherence Analysis.

Purpose: Electroencephalography (EEG) is a non-intrusive technique that provides comprehensive insights into the electrical activities of the brain's cerebral cortex. The brain signals obtained from EEGs can be used as a neuropsychological biomarker to detect different stages of Alzheimer's disease (AD) through quantitative EEG (qEEG) analysis. This paper investigates the difference in the abnormalities of resting state EEG (rEEG) signals between eyes-open (EOR) and eyes-closed (ECR) in AD by analyzing 19-scalp electrode EEG signals and making a comparison with healthy controls (HC). Participants and Methods: The rEEG data from 534 subjects (ages 40-90) consisting of 269 HC and 265 AD subjects in South Korea were used in this study. The qEEG for EOR and ECR states were performed separately for HC and AD subjects to measure the relative power spectrum density (PSD) and coherence with functional connectivity to evaluate abnormalities. The rEEG data were preprocessed and analyzed using EEGlab and Brainstorm toolboxes in MATLAB R2021a software, and statistical analyses were carried out using ANOVA. Results: Based on the Welch method, the relative PSD of the EEG EOR and ECR states difference in the AD group showed a significant increase in the delta frequency band of 19 EEG channels, particularly in the frontal, parietal, and temporal, than the HC groups. The delta power band on the source level was increased for the AD group and decreased for the HC group. In contrast, the source activities of alpha, beta, and gamma frequency bands were significantly reduced in the AD group, with a high decrease in the beta frequency band in all brain areas. Furthermore, the coherence of rEEG among different EEG electrodes was analyzed in the beta frequency band. It showed that pair-wise coherence between different brain areas in the AD group is remarkably increased in the ECR state and decreased after subtracting out the EOR state. Conclusion: The findings suggest that examining PSD and functional connectivity through coherence analysis could serve as a promising and comprehensive approach to differentiate individuals with AD from normal, which may benefit our understanding of the disease.

A machine-learning algorithm for predicting brain age using Rey-Osterrieth complex figure tests of healthy participants.

OBJECTIVE: Neuropsychologists widely use the Rey-Osterrieth complex figure test (RCFT) as part of neuropsychological test batteries to evaluate cognitive function and assess constructional ability, with age being the most significant factor. Our study investigated a supervised machine learning (ML) algorithm to predict brain age gap using RCFT drawings from the healthy elderly community for early dementia detection. PARTICIPANTS AND METHODS: RCFT drawings from 1,970 healthy subjects (ages 45-90 years) were collected from the Korean elderly community. We recorded subject demographic information including: age, gender, and education level. We trained the ML model with RCFT copies, immediate recall, delayed recall, and education level of the healthy subjects using CNN regression algorithm from Keras (https://keras.io/) with the Tensorflow library. RESULTS: The performance was evaluated by the mean absolute error (MAE) and root mean squared error (RMSE) between the predicted age and the chronological age based on a test dataset of 300 healthy subjects. The CNN regression model achieved an MAE of 7.2 years in predicting the brain age gap of the subjects, with an RMSE of 8.9 years. CONCLUSION: The MAE and RMSE accuracies of the CNN regression model predicting the brain age gap showed the model could be a potential biomarker for individual brain aging and a cost-effective method for early dementia detection.

Relationship Between Calcium Channel Blockers Therapy and Cognitive Function Improvement in Cognitive Decline Patients with Cerebrovascular Disease.

INTRODUCTION: Amlodipine belongs to a class of calcium channel blockers that relax blood vessels to allow easier flow of blood. Higher blood pressure (BP) is associated with cerebrovascular disease and is an important contributor to cognitive decline and dementia. AIM: This study aimed to evaluate the effect of 24 weeks of S-amlodipine besylate therapy on cognitive function in patients with hypertension and cerebrovascular disease. METHODS: The data were obtained from a study of post-market surveillance of S-amlodipine besylate. RESULTS: A total of 545 subjects (mean age 67 ± 9.68 years) with hypertension and ischemic cerebrovascular disease were enrolled. Patients with a baseline Mini-Mental State Examination (MMSE) score above 26 were assigned to the cognitive normal (CN) (n = 294) group, and those with MMSE score less than 26 were in the cognitive decline (CD) (n = 251) group. After 24 weeks of treatment with S-amlodipine besylate 5 mg, MMSE and Global Deterioration Scale (GDS) were evaluated again. Changes in MMSE were compared in the target BP reached (TBPR) and non-reached (NTBPR) groups and for CN and CD groups. Treatment with 5 mg of S-amlodipine besylate for 24 weeks improved MMSE and GDS scores (p < 0.001). The CD group showed improvement in MMSE score regardless of whether target BP was obtained (TBPR: p < 0.001, NTBPR: p < 0.01). However, the CN classification was not significant for either TBPR or NTBPR groups. CONCLUSIONS: S-amlodipine besylate improved cognition of the CD group with hypertension and cerebrovascular disease regardless of obtaining target BP.

Is Functional Connectivity after a First Unprovoked Seizure Different Based on Subsequent Seizures and Future Diagnosis of Epilepsy?

Background and Purpose: There are no highly sensitive biomarkers for epilepsy to date. Recently, promising results regarding functional connectivity analysis have been obtained, which may improve epilepsy diagnosis even in the absence of visible abnormality in electroencephalography. We aimed to investigate the differences in functional connectivity after a first unprovoked seizure between patients diagnosed with epilepsy within 1 year due to subsequent seizures and those who were not. Methods: We compared quantitative electroencephalography power spectra and functional connectivity between 12 patients who were diagnosed with epilepsy (two or more unprovoked seizures) within 1 year and 17 controls (those not diagnosed within 1 year) using iSyncBrain® (iMediSync Inc., Suwon, Korea; https://isyncbrain.com/). In the source-level analysis, the current distribution across the brain was assessed using the standardized low-resolution brain electromagnetic tomography technique, to compare relative power values in 68 regions of interest and connectivity (the imaginary part of coherency) between regions of interest. Results: In the epilepsy group, quantitative electroencephalography showed lower alpha2 band power in left frontal, central, superior temporal, and parietal regions and higher beta2 power in both frontal, central, temporal, occipital, and left parietal regions compared with the control group. Additionally, epilepsy patients had significantly lower connectivity in alpha2 and beta2 bands than the controls. Conclusions: Patients experiencing their first unprovoked seizure presented different brain function according to whether they have subsequent seizures and future epilepsy. Our results propose the potential clinical ability to diagnose epilepsy after the first unprovoked seizure in the absence of interictal epileptiform discharges.

Power Spectral Changes of Quantitative EEG in the Subjective Cognitive Decline: Comparison of Community Normal Control Groups.

PURPOSE: The purpose of this study is to compare and analyze the power spectral changes between subjective cognitive decline (SCD) subjects and normal controls (NC) while checking the preclinical stage of AD in the SCD subjects and to use the derived data for biomarker research that can diagnose early-stage AD in the future. METHODS: We recruited 23 SCD patients and 23 normal control subjects and QEEG analysis including power spectral density (PSD) and source-level analysis were performed. An automated preprocessing procedure and statistical analysis were performed by iSync Brain® (iMediSync Inc., Republic of Korea) (https://isyncbrain.com/) using the international standard 10-20 system (19 electrodes). RESULTS: Absolute PSD, there was no statistically significant difference in all of the EEG power measurements of the 19 channels. In the relative PSD analysis, the average delta band power of the SCD group was significantly higher in Fp2, F4, and F8 than NC. Alpha1 band power of the O1 channel was 22.56±16.05 for the SCD group and 33.19±19.05 for the NC (p-value <0.05). Source-level analysis did not show a statistically significant difference. CONCLUSION: SCD subjects showed a partial increase of delta waves in the frontal lobe region and a partial decrease in alpha1, a fast wave in the occipital region, compared to the NC. SCD is considered one of the earliest clinical symptoms of AD and it is predicted to be related to minor nerve damage. We were able to observe the power spectral changes in SCD subjects in this cross-sectional study, a large number of subjects and longitudinal studies are needed to evaluate their predictability for future deterioration such as conversion to MCI.

Differences between memory encoding and retrieval failure in mild cognitive impairment: results from quantitative electroencephalography and magnetic resonance volumetry.

BACKGROUND: The memory impairments in mild cognitive impairment (MCI) can be classified into encoding (EF) and retrieval (RF) failure, which can be affected by underlying pathomechanism. We explored the differences structurally and functionally. METHODS: We compared quantitative electroencephalography (qEEG) power spectra and connectivity between 87 MCI patients with EF and 78 MCI with RF using iSyncBrain® (iMediSync Inc., Republic of Korea) ( https://isyncbrain.com/ ). Voxel-based morphometric analysis of the gray matter (GM) in the MCI groups and 71 cognitive normal controls was also done using the Computational Anatomy Toolbox 12 ( http://www.neuro.uni-jena.de/cat/ ). RESULTS: qEEG showed higher frontal theta and lower beta2 band power, and higher theta connectivity in the EF. There was no statistically significant difference in GM volume between the EF and RF. However, when compared to normal control, GM volume reductions due to EF in the left thalamus and bilateral hippocampi and reductions due to RF in the left thalamus, right superior frontal lobe, right superior temporal lobe, and right middle cingulum were observed (p < 0.05, family-wise error correction). CONCLUSIONS: MCI differs functionally and structurally according to their specific memory impairments. The EF findings are structurally and functionally more consistent with the prodromal Alzheimer's disease stage than the RF findings. Since this study is a cross-sectional study, prospective follow-up studies are needed to investigate whether different types of memory impairments can predict the underlying pathology of amnestic MCI. Additionally, insufficient sample size may lead to ambiguous statistical findings in direct comparisons, and a larger patient cohort could more robustly identify differences in GM volume reductions between the EF and the RF group.