Alzheimer's & Dementia

INTRODUCTION The Virginia Memory Project (VMP) is a statewide epidemiological registry for Alzheimer's disease and related disorders (ADRD) and other neurodegenerative conditions. It aims to support dementia research, policy, and care by leveraging the Centers for Disease Control (CDC) Healthy Brain Initiative (HBI) Roadmap. METHODS To capture comprehensive data, the VMP integrates self‐enrollment and automatic enrollment using Virginia's All‐Payer Claims Database (APCD). It also adapts Behavioral Risk Factors Surveillance Survey (BRFSS) modules for self‐reported cognitive and caregiving data, offering connections to research, clinical services, and education. RESULTS Virginia successfully codified the VMP in the 2024 general assembly session. DISCUSSION The VMP demonstrates a novel approach to Alzheimer's Disease and Related Disorders (ADRD) surveillance by combining traditional registry functions with community engagement and workforce development. Future efforts will focus on increasing enrollment, especially among underrepresented groups, to enhance data‐driven dementia policy and care in Virginia. Highlights Integrated the Healthy Brain Initiative (HBI) domains into the newest statewide epidemiological dementia registry in the Commonwealth of Virginia. Collected data and identified gaps in the current research related to dementia and Alzheimer's related diseases. Aimed to mitigate barriers to dementia registry enrollment by identifying significant underdiagnosis and underrepresentation of racial and ethnic minority groups. Developed solutions to alleviate the current data and enrollment disparities and to connect individuals to research, physicians, and community groups.

INTRODUCTION Deciphering the diverse molecular mechanisms in living Alzheimer's disease (AD) patients is a big challenge but is pivotal for disease prognosis and precision medicine development. METHODS Utilizing an optimal transport approach, we conducted graph‐based mapping of transcriptomic profiles to transfer AD subtype labels from ROSMAP monocyte samples to ADNI and ANMerge peripheral blood mononuclear cells. Subsequently, differential expression followed by comparative pathway and diffusion pseudotime analysis were applied to each cohort to infer the progression trajectories. Survival analysis with real follow‐up time was used to obtain potential biomarkers for AD prognosis. RESULTS AD subtype labels were accurately transferred onto the blood samples of ADNI and ANMerge living patients. Pathways and associated genes in neutrophil degranulation‐like immune process, immune acute phase response, and IL‐6 signaling were significantly associated with AD progression. DISCUSSION The work enhanced our understanding of AD progression in different subtypes, offering insights into potential biomarkers and personalized interventions for improved patient care. Highlights We applied an innovative optimal transport‐based approach to map transcriptomic data from different Alzheimer's disease (AD) cohort studies and transfer known AD subtype labels from ROSMAP monocyte samples to peripheral blood mononuclear cell (PBMC) samples within ADNI and ANMerge cohorts. Through comprehensive trajectory and comparative analysis, we investigated the molecular mechanisms underlying different disease progression trajectories in AD. We validated the accuracy of our AD subtype label transfer and identified prognostic genetic markers associated with disease progression, facilitating personalized treatment strategies. By identifying and predicting distinctive AD subtypes and their associated pathways, our study contributes to a deeper understanding of AD heterogeneity.

INTRODUCTION Cerebrovascular dysfunction plays a critical role in the pathogenesis of dementia and related neurodegenerative disorders. Recent omics‐driven research has revealed associations between vascular abnormalities and transcriptomic alterations in brain vascular cells, particularly endothelial cells (ECs) and pericytes (PCs). However, the impact of these molecular changes on dementia remains unclear. METHODS We conducted a comparative analysis of gene expression in ECs and PCs across neurodegenerative conditions, including Alzheimer's disease (AD), Huntington's disease, and arteriovenous malformation, utilizing transcriptomic data from published postmortem human tissue studies. RESULTS We identified differentially expressed genes (DEGs) consistently dysregulated in ECs and PCs across these pathologies. Notably, several DEGs are linked to vascular cell zonation and genetic risks for AD and cerebral small vessel disease. DISCUSSION Our findings provide insights into the cellular and molecular mechanisms underlying vascular dysfunction in dementia, highlight the knowledge gaps, and suggest potential novel vascular therapeutic targets, including genes not previously investigated in this context. Highlights Systematic review of differentially expressed genes (DEGs) in vascular cells from neurodegenerative single‐nuclear RNA‐sequencing (snRNA‐seq) studies. Identify overlapping DEGs in multiple vascular cell types across studies. Examine functional relevance and associations with genetic risk for common DEGs. Outline future directions for the vascular omics field.

Observational studies on the cancer–dementia relationship have yielded controversial results. This study systematically reviews the evidence to clarify this association. We searched Embase, Global Health, Ovid Medline, and APA PsycInfo. Colorectal and lung cancers showed the greatest risk reduction for all‐cause dementia (ACD) and Alzheimer's disease (AD), respectively, while melanoma and colorectal cancers had the largest reduction in vascular dementia (VaD). Prostate cancer survivors on androgen deprivation therapy (ADT) had a higher risk of ACD/AD, while breast cancer patients on tamoxifen had a lower AD risk. Chemotherapy was linked to a reduced AD risk. ACD patients saw a 30% risk reduction for bladder, colorectal, and lung cancers, while AD patients had a ≈ 35% reduction for bladder and lung cancers. Our study urges clinicians to monitor cognitive function in cancer patients, especially those on ADT, tamoxifen, or chemotherapy and highlights the need for research into cancer–dementia mechanisms. Highlights Cancer survivors have an 8% to 14% lower risk of dementia, while those with dementia have a 25% lower cancer risk. Colorectal and non‐melanoma skin cancers were associated with reduced risks of all‐cause dementia (ACD; 16%/9%), Alzheimer's disease (AD; 13%/5%), and vascular dementia (VaD; 24%/9%). Lung cancer reduced AD risk by 17%, and melanoma reduced VaD risk by 27%. ACD and AD patients had lower risks of lung (30%/36%), bladder (32%/34%), breast (26%/20%), and colorectal (31%/28%) cancers. Tamoxifen and chemotherapy reduced AD risk, while androgen deprivation therapy increased ACD risk.

INTRODUCTION Alzheimer's disease (AD) in Down syndrome (DS) is associated with changes in brain structure. It is unknown if thickness and volumetric changes can identify AD stages and if they are similar to other genetic forms of AD. METHODS Magnetic resonance imaging scans were collected for 178 DS adults (106 nonclinical, 45 preclinical, and 27 symptomatic). Cortical thickness and subcortical volumes were compared between DS groups and evaluated as a staging metric using receiver operating characteristic analyses. Thickness patterns were compared to those previously reported in autosomal‐dominant AD (ADAD). RESULTS Decreased parietal and temporal lobe thickness differentiated amyloid positivity (area under the curve [AUC] = 0.83) and impairment (AUC = 0.81), and slightly outperformed subcortical volumes (AUC = 0.8/0.74). Thickness differences in DS were more widespread, severe, and had better discriminative ability than ADAD. DISCUSSION Cortical thickness can stage AD pathology in DS. Identification of brain regions affected by AD may aid in tracking disease course and evaluating treatment effects. Highlights DSAD is associated with reduced temporal and parietal cortical thickness. DSAD is associated with smaller hippocampal and striatal volumes. Thickness differences can stage DSAD better than other forms of AD. DSAD thickness differences are more extensive and severe than ADAD.

INTRODUCTION Interferon‐induced transmembrane protein 3 (IFITM3) modulates γ‐secretase in Alzheimer's Disease (AD). Although IFITM3 knockout reduces amyloid β protein (Aβ) production, its cell‐specific effect on AD remains unclear. METHODS Single nucleus RNA sequencing (snRNA‐seq) was used to assess IFITM3 expression. Adeno‐associated virus‐BI30 (AAV‐BI30) was injected to reduce IFITM3 expression in the cerebrovascular endothelial cells (CVECs). The effects on AD phenotypes in cells and AD mice were examined through behavioral tests, two‐photon imaging, flow cytometry, Western blot, immunohistochemistry, and quantitative polymerase chain reaction assay (qPCR). RESULTS IFITM3 expression was increased in the CVECs of patients with AD. Overexpression of IFITM3 in primary endothelial cells enhanced Aβ generation through regulating beta‐site APP cleaving enzyme 1 (BACE1) and γ‐secretase. Aβ further increased IFITM3 expression, creating a vicious cycle. Knockdown of IFITM3 in CVECs decreased Aβ accumulation within cerebrovascular walls, reduced Alzheimer's‐related pathology, and improved cognitive performance in AD transgenic mice. DISCUSSION Knockdown of IFITM3 in CVECs alleviates AD pathology and cognitive impairment. Targeting cerebrovascular endothelial IFITM3 holds promise for AD treatment. Highlights Interferon‐induced transmembrane protein 3 (IFITM3) expression was increased in the cerebrovascular endothelial cells (CVECs) of patients with Alzheimer's Disease (AD). Cerebrovascular endothelial IFITM3 regulates amyloid β protein (Aβ) generation through regulating beta‐site APP cleaving enzyme 1 (BACE1) and γ‐secretase. Knockdown of IFITM3 in CVECs reduces Aβ deposits and improves cognitive impairments in AD transgenic mice. Cerebrovascular endothelial IFITM3 could be a potential target for the treatment of AD.

INTRODUCTION Using an Asian cohort with high prevalence of concomitant cerebrovascular disease (CeVD), we evaluated the performance of a plasma immunoassay for tau phosphorylated at threonine 217 (p‐tau217) in detecting amyloid beta positivity (Aβ+) on positron emission tomography and cognitive decline, based on a three‐range reference, which stratified patients into low‐, intermediate‐, and high‐risk groups for Aβ+. METHODS Brain amyloid status (Aβ– [n = 142] vs Aβ+ [n = 73]) on amyloid PET scans was assessed along with the plasma ALZpath p‐tau217 assay to derive three‐range reference points for PET Aβ+ based on 90% sensitivity (lower threshold) and 90% specificity (upper threshold). RESULTS Plasma p‐tau217 (area under the curve [AUC] = 0.923) outperformed routine clinical assessments (AUC = 0.760–0.819; p ≤ 0.003) and other plasma biomarkers (AUC = 0.817–0.834; p < 0.001). The high‐risk group showed significantly higher rates of cognitive decline than the low‐risk group. DISCUSSION Risk stratification for PET Aβ+ based on a plasma p‐tau217 assay demonstrated potential diagnostic and prognostic utility in an Asian cohort with concomitant CeVD. Highlights The utility of plasma p‐tau217 to detect brain amyloid beta pathology (Aβ+) was studied in an Asian cohort with concomitant cerebrovascular disease Plasma tau phosphorylated at threonine 217 (p‐tau217) showed superior utility in detecting Aβ+ compared to neuroimaging measures, clinical workup, or other blood biomarkers including p‐tau181, glial fibrillary protein (GFAP), and Aβ Higher plasma p‐tau217 correlated with faster cognitive decline Plasma p‐tau217 shows promise as an Alzheimer's disease (AD) diagnostic and prognostic biomarker in diverse populations

Limbic predominant age‐related TDP‐43 encephalopathy neuropathologic change (LATE‐NC) is highly prevalent in late life and a common co‐pathology with Alzheimer's disease neuropathologic change (ADNC). LATE‐NC is a slowly progressive, amnestic clinical syndrome. Alternatively, when present with ADNC, LATE‐NC is associated with a more rapid course. With the emergence of anti‐amyloid therapeutics, discrimination of LATE‐NC from ADNC is critical and will lead to greater clinical recognition of amnestic patients without ADNC. Furthermore, co‐pathology with LATE‐NC may influence outcomes of these therapeutics. Thus there is a need to identify patients during life with likely LATE‐NC. We propose criteria for clinical diagnosis of LATE as an initial framework for further validation. In the context of progressive memory loss and substantial hippocampal atrophy, criteria are laid out for probable (amyloid negative) or possible LATE (amyloid biomarkers are unavailable or when amyloid is present, but hippocampal neurodegeneration is out of proportion to expected pure ADNC). Highlights Limbic‐predominant age‐related TDP‐43 encephalopathy (LATE) is a highly prevalent driver of memory loss in late life. LATE neuropathic change (LATE‐NC) is a common co‐pathology with Alzheimer's disease neuropathic change (ADNC) and may influence outcomes with emerging disease‐modifying medicines. We provide initial clinical criteria for diagnosing LATE during life either when LATE‐NC is the likely primary driver of symptoms or when observed in conjunction with AD. Definitions of possible and probable LATE are provided.

INTRODUCTION Transcranial pulse stimulation (TPS) is increasingly being investigated as a promising potential treatment for Alzheimer's disease (AD). Although the safety and preliminary clinical efficacy of TPS short pulses have been supported by neuropsychological scores in treated AD patients, its fundamental mechanisms are uncharted. METHODS Herein, we used a multi‐modal preclinical imaging platform combining real‐time volumetric optoacoustic tomography, contrast‐enhanced magnetic resonance imaging, and ex vivo immunofluorescence to comprehensively analyze structural and hemodynamic effects induced by TPS. Cohorts of healthy and AD transgenic mice were imaged during and after TPS exposure at various per‐pulse energy levels. RESULTS TPS enhanced the microvascular network, whereas the blood–brain barrier remained intact following the procedure. Notably, higher pulse energies were necessary to induce hemodynamic changes in AD mice, arguably due to their impacted vessels. DISCUSSION These findings shed light on cerebrovascular dynamics induced by TPS treatment, and hence are expected to assist improving safety and therapeutic outcomes. Highlights ·Transcranial pulse stimulation (TPS) facilitates transcranial wave propagation using short pulses to avoid tissue heating. ·Preclinical multi‐modal imaging combines real‐time volumetric optoacoustic (OA) tomography, contrast‐enhanced magnetic resonance imaging (CE‐MRI), and ex vivo immunofluorescence to comprehensively analyze structural and hemodynamic effects induced by TPS. ·Blood volume enhancement in microvascular networks was reproducibly observed with real‐time OA imaging during TPS stimulation. ·CE‐MRI and gross pathology further confirmed that the brain architecture was maintained intact without blood–brain barrier (BBB) opening after TPS exposure, thus validating the safety of the procedure. ·Higher pulse energies were necessary to induce hemodynamic changes in AD compared to wild‐type animals, arguably due to their pathological vessels.

INTRODUCTION Successful cognitive aging is related to both maintaining brain structure and avoiding Alzheimer's disease (AD) pathology, but how these factors interplay is unclear. METHODS A total of 109 cognitively normal older adults (70+ years old) underwent amyloid beta (Aβ) and tau positron emission tomography (PET) imaging, structural magnetic resonance imaging (MRI), and cognitive testing. Cognitive aging was quantified using the cognitive age gap (CAG), subtracting chronological age from predicted cognitive age. RESULTS Lower CAG (younger cognitive age) was related to slower decline in episodic memory, multi‐domain cognition, and atrophy of the midcingulate cortex (MCC). Lower entorhinal cortical tau was linked to slower decline in episodic memory, multi‐domain cognition, and hippocampal atrophy. DISCUSSION These results suggest that aging outcomes may be influenced by two independent pathways: one associated with tau accumulation, affecting primarily memory and hippocampal atrophy, and another involving tau‐independent structural preservation of the MCC, benefiting multi‐domain cognition over time. Highlights Younger cognitive age (lower cognitive age gap [CAG]) is related to slower cognitive decline. Lower CAG is linked to slower midcingulate cortex (MCC) atrophy. Reduced tau in the entorhinal cortex is related to less hippocampal atrophy and cognitive decline. Structural preservation of the MCC benefits multi‐domain cognition over time. Two independent pathways influence cognitive aging: tau accumulation and MCC preservation.

INTRODUCTION This study describes the implementation outcomes and evaluation of DEM‐SKY, a community‐based dementia screening program developed in rural Kenya with the support of community health care workers (CHWs). METHODS DEM‐SKY was delivered to 3546 older adults in Makueni County, Kenya, over a 6‐month period. Using a mixed‐methods design, we explored implementation outcomes with stakeholders through surveys and interviews. RESULTS The program demonstrated good acceptability, adoption, and fidelity and was effective in instigating behavior change. Individuals who screened positive for dementia were 28.7 times more likely to intend to speak to a doctor. Qualitative data showed that participants valued the program but indicated scope for improvement, particularly further down the diagnostic pathway. DISCUSSION DEM‐SKY was successful across several implementation metrics. Although the program demonstrates that community‐based screening can be conducted effectively with minimal resources, future research needs to explore the long‐term benefits of dementia screening in Kenya. Highlights Community‐based dementia screening is feasible in rural Africa. Involving community health workers strengthens trust in health care systems. Empowering community health workers enhances the community capacity to address dementia Screening promotes proactive health seeking among older adults.

BACKGROUND We sought to characterize the cognitive profile among individuals with mild cognitive impairment with Lewy bodies (MCI‐LB) to help guide future clinical criteria. METHODS Systematic review and meta‐analysis included MCI‐LB studies with cognitive data from PubMed, Embase, Web of Science, and PsycINFO (January 1990 to March 2023). MCI‐LB scores were compared to controls, MCI due to Alzheimer's disease (MCI‐AD), and dementia with Lewy bodies (DLB) groups with random‐effects models. RESULTS We included 26 studies and 2823 participants. Across all domains, the MCI‐LB group performed worse than controls and better than DLB. Compared to MCI‐AD, the MCI‐LB group performed worse in attention/processing speed (g = –0.24, 95% confidence interval [CI]: –0.35, –0.12), attention/executive (g = –0.42, 95% CI: –0.56, –0.28); better in verbal immediate recall (g = 0.37; 95% CI: 0.15, 0.59) and delayed memory (g = 0.40; 95% CI: 0.22, 0.58). DISCUSSION The cognitive profiles in MCI‐LB and MCI‐AD are consistent with established profiles in DLB and AD. Neuropsychological assessment may be helpful in differential diagnosis, even in early disease states. Highlights We performed a systematic review and meta‐analysis for cognition in mild cognitive impairment with Lewy bodies (MCI‐LB). Compared to MCI due to Alzheimer's disease (MCI‐AD), MCI‐LB had worse attention, executive function, and processing speed. Compared to MCI‐AD, MCI‐LB had better verbal immediate and delayed recall. The MCI‐LB group was worse on all cognitive domains than controls, and better than dementia with Lewy bodies. Studies used different tests and there is a need for global efforts for harmonization.

Background The early detection of neurologic damage at the microscopic level when the disease is subclinical would facilitate intervention preventing progression or potentially reversing the condition. The early determination of drug efficacy could shorten the length of drug studies, thereby reducing research costs. The eye is the only place in the body where an artery, vein, and nerve can be directly visualized The nerve fiber layer of the retina is an outgrowth of the brain. Method Dynamic Light Scattering (DLS) Spectroscopy measures the thermal random movement (Brownian Motion) of particles by analyzing the temporal fluctuations of scattered light. A proof‐of‐concept instrument for making retinal measurements provides optical power well below the maximal permissible exposure recommended by the ANSI Z136.1 standard. Scattered light is analyzed by a digital autocorrelator with an extended delay option for baseline determination. The testing duration is 5 seconds. DLS measurements were made from the macular area of 4 patients diagnosed with mild cognitive impairment and 4 age‐matched controls with no history of neurologic disease. Both groups of patients had normal retinal examinations, including normal Optical Coherence Tomography, (OCT). Result Figure 1 Screenshot ‐ 62‐year‐old female recently with mild cognitive impairment, subsequently diagnosed with Alzheimer’s disease 1 year after DLS testing. Figure 2 Screenshot ‐ Age‐matched control Conclusion The 4 patients with mild cognitive impairment were subsequently diagnosed with Alzheimer’s disease approximately 1 year after the abnormal DLS measurement. The differences between the patients subsequently diagnosed with Alzheimer’s disease, compared to the normal age‐matched controls, was consistent across the tested patients. Alzheimer’s measurements begin above 0.6 g 2(t)‐1 and exhibit flattening of the initial line as opposed to the curve seen in non‐Alzheimer’s patients. The development of an early, noninvasive, quantitative test to diagnose Alzheimer’s disease will lead to breakthroughs in drug development and hopefully, to the successful treatment of patients before dementia is irreversible.

Background MATCH (Music Attuned Technology – Care via eHealth) is a music and health application that supports caregivers of people living with dementia to use music strategically to better manage care through virtual training and intuitive music technology. This study trialled a prototype version of the MATCH app with family caregivers and people with dementia residing in the community. Method 16 Dyads trialled the prototype MATCH app. After completing the virtual training in the app, family caregivers created personalised playlists within the app and used strategies learnt during training with the person they care for at least twice per week over 8 weeks. Pre‐post measures on the Neuropsychiatric Inventory Questionnaire (NPI‐Q), Cohen‐Mansfield Agitation Inventory and Acceptability of the mobile app and training were captured to detect change in symptoms and feasibility of the application. Qualitative measures including a participation diary and qualitative interviews captured participant perspectives and usage of the app. Results Significant pre‐post changes in symptom severity (‐4.23 [4.9], 95% CI ‐7.2 to ‐1.3, p = 0.009) and symptom distress (‐5.69 [5.8], 95% CI ‐9.2 to ‐2.2, p = 0.004) on the NPI‐Q were found, but not on the Cohen‐Mansfield Agitation Inventory (‐4.38 [11.6], 95% CI ‐9.9 to 1.34, p = 0.109). Subscales of the NPI‐Q found significant changes to mood symptoms (‐1.23 [1.9], 95% CI ‐2.36 to ‐0.01, p = 0.036) and frontal symptoms (‐1.45 [23], 95% CI ‐2.87 to ‐0.05, p = 0.043), but not agitation/aggression (‐1.1 (1.9], 95% CI ‐2.21 to 0.07, p = 0.062). 85% of carers were satisfied with the MATCH app features, and 92% were satisfied with the music training. Qualitative findings revealed that participants found the strategies to be effective, and that they were able to integrate music‐based strategies into their care routines. Conclusion The MATCH app was found to be feasible and acceptable; family caregivers reported that MATCH training was understandable and useful. There is preliminary evidence that music interventions implemented by trained family caregivers are effective in reducing neuropsychiatric symptoms.

Background There is ample evidence that music can boost brain activity and jog deeply embedded memories. Literature indicates a significant improvement in autobiographical memory (ABM) recall for different individuals during background music sessions. Existing research is based solely on qualitative data, although music has a significant impact on physiological activity. Thus, it’s important to explore the connection between memory recall and physiological activities. Method To better understand memory recall, the electroencephalogram (EEG) and electrodermal activity (EDA) data were gathered from healthy participants using wearable sensors. Physiological signals such as the electroencephalogram (EEG) and electrodermal activity (EDA) were recorded as quantitative data using various wearable sensors from 40 participants of different age groups while playing different background music sessions. The study involved listening to nine music sessions (three happy, three sad, and three neutral). Immediately after each piece of music, a post‐study survey was conducted to gauge if the participants recalled any autobiographical memories. A machine learning algorithm was developed to train a model using features collected from physiological data to determine if the memory recall was successful. The purpose of the study was to identify an EEG biomarker. Result The results of the EEG and EDA data analysis revealed that for all four EEG channels, there was a consistent increase in the alpha power (on average 16.2%) during the memory “recall” scenario (F3: p = 0.0066, F7: p = 0.0386, F4: p = 0.0023, and F8: p = 0.0288) compared to the “no‐recall” control. There was also a significant surge in the Beta power for two channels (F3: p = 0.0100 and F4: p = 0.0210) but not for the control (F7: p = 0.6792 and F8: p = 0.0814). Additionally, the EDA data analysis revealed significant differences in the phasic standard deviation (p = 0.0260), phasic max (p = 0.0011), phasic energy (p = 0.0478), tonic min (p = 0.0092), tonic standard deviation (p = 0.0171), and phasic energy (p = 0.0478). This implies that the memory recall biomarker is alpha power (8‐12 Hz). Conclusion The results indicate that the biomarker for memory recall is alpha power (8‐12Hz).

Background Past studies examining sleep‐cognition relationships mostly employed univariate approaches, which are subject to problems such as multicollinearity and multiple comparisons. Further, results from small sample univariate analyses are difficult to compare, precluding the identification of the aspects of sleep health associated with a particular cognitive domain(s). The current study used a multivariate approach to identify key sleep metrics and cognitive domains that contribute to the maximum sleep‐cognition covariance in healthy older adults. From the reduced list of sleep metrics and cognitive domains, novel associations between different aspects of sleep health and cognitive domains can be uncovered. Method The current study is part of the ongoing SG70: Toward Healthy Longevity in Singapore study, which aims to comprehensively assess factors that affect aging health in over 1000 Singaporean older adults. From the SG70 study, 440 healthy older adults wore an Oura Ring (OuraHealth OY) for 14‐30 days. Twenty‐three metrics encompassing 4 major aspects of objective sleep health: duration, timing, regularity, and continuity were extracted (Figure 1). Cognition was assessed using a comprehensive battery that encompassed 7 domains. The overall covariance between sleep and cognition was examined by a partial least square correlation (PLSC) analysis. Sleep metrics and cognitive domains that contributed significantly to significant PLSC components were identified by bootstrapping. Result PLSC analysis identified a component that explained 68% of covariance between sleep and cognition matrices (r = 0.17, p<0.001). Bootstrapping tests further identified 10 sleep continuity and regularity metrics and 4 corresponding cognitive domains that contributed significantly to the observed covariance (Figure 2). Post‐hoc univariate analyses showed that sleep continuity metrics correlated with speed of processing, while sleep regularity metrics correlated with verbal memory, visual‐spatial ability, executive functions, and speed of processing (Figure 3). Conclusion Our results suggest that sleep continuity and regularity may be more sensitive markers of impairments across multiple cognitive domains in healthy aging compared to sleep duration and timing. In addition, they support the utility of multivariate analyses in uncovering significant association patterns between sleep and cognition.

Background Regulated breathing is increasingly recognized as a vital component in enhancing cognition. Scientific studies suggest that intentional and controlled breathing techniques, such as deep and rhythmic breathing, can promote relaxation, reduce stress, and improve oxygen flow to the brain [1, 2]. Consequently, this may contribute to heightened cognitive function, better concentration, and increased mental clarity, accentuating the importance of incorporating mindful breathing practices as a potential avenue for cognitive improvement [3]. The increasing recognition of the cognitive benefits of regulated breathing is leading to the development of innovative technologies, such as robots, for better breathing practices. These technologies can be used to potentially enhance the cognitive well‐being and mindfulness of people in their daily routines. Methods In order to accomplish a breathing robot, the entire design consists of two linear actuators controlled by an L298N Motor Driver Interface and a Raspberry Pi 3B. The Raspberry Pi coordinates both linear actuators to extend, hold, and contract in opposite directions for 4 seconds at each stage. Additionally, the mechanical design utilizes two “push caps” that push at the sides of a contained space, such as a stuffed animal, simulating breathing (see Fig. 1). This design simulates a breathing pattern in any stuffed animal (see Fig. 2). Results The developed robot underwent rigorous testing in both a lab and poster presentation setting. Various case studies were examined during the poster presentation, revealing the effectiveness of a tool designed to aid in regulated breathing. The present robot demonstrated increased participant engagement in mindful breathing practices, which suggests a promising avenue for leveraging robot‐assisted therapies to enhance cognitive well‐being in people. Conclusion The present project aimed to build a robot that fosters cognitive well‐being through regulated breathing. The case study findings of this project highlight the effectiveness of a cuddly bear‐shaped robot as a valuable tool for promoting regulated breathing, offering positive implications for integrating such robotic assistance in enhancing cognitive well‐being.

Background Self‐administered cognitive assessments demonstrate usability and ability to detect cognitive decline in Alzheimer’s disease, but usability in other neurodegenerative diseases is understudied. We investigated whether Mayo Test Drive (MTD), a self‐administered multi‐device compatible cognitive assessment platform, demonstrates usability and correlation with traditional neuropsychological tests in a pilot study of individuals with progressive supranuclear palsy (PSP). Method Eleven individuals with PSP (mean age = 69.6±7.2 years; mean education = 15.3±2.7 years; 72.7% male; 90.9% white; PSP Rating Scale [PSPRS] = 32.4±15.5) enrolled in a Neurodegenerative Research Group study at Mayo Clinic participated. Participants completed neurological evaluation, traditional neuropsychological assessment, in‐person self‐administered MTD comprised of the Stricker Learning Span and Symbols Test, and a MTD usability questionnaire with quantitative and qualitative questions. Participants received diagnoses of PSP clinical variants per Movement Disorders Society‐PSP criteria. Seven participants had PSP‐Richardson, two had PSP‐postural instability, and two had PSP‐parkinsonism. We used descriptive analyses to examine usability questions, qualitative analyses for free text responses, and Spearman correlations to analyze associations between MTD and traditional tests of global cognition and five cognitive domains. Result Ten of eleven patients completed MTD (average completion time [minutes] = 22.2±6.5). Study staff terminated one participant’s session early due to an appointment conflict; we excluded this participant from analyses. Six participants completed MTD with tablets and four used smartphones. Participants reported high ease completing MTD (9.00±1.76 on 0‐10 scale) and comfortability using their devices (9.40±1.58). All participants reported they could complete MTD independently, with one participant indicating need for assistance with hyperlinks and three reporting need for assistance with QR codes. Qualitative data suggested three participants had motoric difficulty using the tablets; these individuals had the greatest PSPRS symptom severity, although MTD performance was not correlated with the PSPRS (rho = ‐.122; p>.05). MTD performance significantly correlated with the MoCA (rho = 0.88) and tests of attention, executive, and visuospatial function (Table). Conclusion In a MTD usability pilot for participants with PSP, participants reported high ease and comfort completing MTD independently; those with greater motor symptoms had more difficulty navigating technology. Additionally, performances on MTD were correlated with the MoCA and other in‐person assessments, demonstrating convergent validity.

Background We currently lack in the dementia field accurate, noninvasive, quick, and affordable screening tools for brain dysfunctions associated with early subtle risk of mild cognitive impairment (MCI). Our Kentucky aging cohort demonstrates that asymptomatic older individuals with MCI‐like frontal memory‐related brainwave patterns convert to MCI within a short 5‐year period, as opposed to individuals with NC‐like patterns (1) that remain normal 10 years later (2). Astrocyte reactivity influences amyloid‐ß effects on tau pathology in preclinical Alzheimer’s disease (3). Leveraging blood‐based AD and astrocyte biomarkers and the cognitive electroencephalogram (EEG) signatures (4), we test the hypothesis that predictive frontal memory‐related EEG changes correlate with preclinical and early AD plasma biomarkers. Method 34 (19 women) older volunteers with or without MCI, average age 79 (SD 8.53) years old, from a longitudinal cohort followed by University of Kentucky ADRC participated. Each participant’s EEG was recorded (64‐ or 14‐channels) during a working memory (modified delayed match‐to‐sample) task. Principal component analysis (PCA) was performed on 64‐channel EEG data to create PC scores (PC1 & PC2). For multiple linear regression of EEG PC scores on multiple neurodegenerative plasma biomarkers including Aß42/40, pTau181, total Tau, and GFAP (Astrocyte reactivity), we adjusted age, sex, education, and gap years between collection dates. Result The 61% of variance in frontal signals can be explained by PC1 in normal cognition (NC) and MCI individuals, and PC2 counts for 35% of variance (Figure 1). The decreased brainwaves (MCI‐like) seen in left frontal sites significantly correlate with increased pTau181, GFAP, and PC2 (Figure 2). Curiously, right frontal EEG relations with pTau181, GFAP showed the opposite trend. Bilateral frontal signals showed negative correlations with Aß42/40 and positive correlations with total Tau. Conclusion Our results indicate that GFAP & pTau181 trend in similar asymmetry ways with frontal cognitive brainwaves, but Aß42/40 & total Tau correlate to a different component of frontal EEG. That is, distinct cognitive brainwaves correlate with astrocyte reactivity differentially that influence pathologies of beta‐amyloid accumulations and Tau development. Cognitive pathophysiological signatures and AD–Astrocyte plasma biomarkers have great potential for predicting subtle cognitive decline and specific dementia risk in healthy normal individuals.

There is growing recognition of the importance of changes in behavior as an early clinical marker of the onset of Alzheimer’s disease. Behavior symptoms may precede the onset of cognitive symptoms by as many as three years. However, these symptoms are often confused for primary psychiatric pathology and there is an urgent need for markers that can help quantify behaviors with the eventual goal of helping distinguish behavior changes related to psychiatric pathology from behavior changes related to the onset of dementia. This talk will summarize the state of knowledge on mild behavioral impairment, and present data from three studies demonstrating how a range of sensor‐derived biomarkers can quantify subtle changes in behavior with precision. The talk will include case presentations around detection of apathy, agitation, pacing. It will also demonstrate how these markers can identify response to medications, which may be one way of distinguishing the presence of psychiatric versus neurodegenerative pathology. Finally, the talk will present data from ongoing clinical trials around how these biomarkers may translate into research and eventually practice.

Background When designing cutting‐edge technology, particularly humanoid social robots, an essential consideration is understanding how individuals naturally engage in social interactions, shaping their relationships with technology and media. Method In pursuit of insights into the application of natural human behavior, specifically reciprocation, in human‐robot interaction, an experiment involving 72 participants, involving facial electromyography, focusing on zygomatic and corrugator muscles, served as a tool to gauge users' emotional valence during interactions. The study assessed users' willingness to reciprocate a favor and measured compliance by tracking the number of raffle tickets purchased by users at the robot’s request. Events were recorded during the participant’s interaction with Pepper the humanoid robot, analyzing facial EMG data in five‐second intervals to capture changes in muscle activity before and after key events. The study comprised three scripted events, including the participant’s first sighting of Pepper, Pepper offering water (favor condition only), and a simulated glitch on Pepper’s tablet. Post‐interaction, participants filled out a final survey, contributing to the comprehensive examination of their experiences. The research involved 72 undergraduate students from diverse majors. The study aimed to assess initial impressions of humanoid robots, utilizing a phasic study design with a five‐second response time, aligning with previous research practices. The participants, largely lacking prior experience with social, service, or humanoid robots, provided valuable insights into human‐robot interaction dynamics. Results In testing two hypotheses on participants' interaction with a Humanoid Social Robot (HSR), we found no significant differences in ticket purchases between favor and non‐favor conditions using statistical tests, including chi‐square and Mann‐Whitney U. However, an independent samples t‐test, after outlier removal, revealed a significant difference, supporting both hypotheses. Pearson’s correlation coefficients showed positive correlations between emotional responses, but no significant relation with self‐reported affect or ticket purchases. These findings enhance understanding of human‐robot interaction dynamics and inform future research directions. Conclusion This study emphasizes the importance of incorporating natural human behaviors, like reciprocation, in designing humanoid social robots, indicating that initiating positive interactions significantly impacts users' compliance in subsequent interactions, advancing the field of human‐robot interaction with theoretical implications and proposing future research avenues despite acknowledged study limitations.

Background Structural and functional heterogeneity in the brains of patients with Alzheimer’s disease (AD) leads to diagnostic and prognostic uncertainty and confounds clinical treatment planning. Normative modelling, where individual‐level deviations in brain measures from a reference sample are computed to infer personalized effects of disease, allows parsing of disease heterogeneity. In this study, GAN based normative modelling technique quantifies individual level neuroanatomical abnormality thereby facilitating measurement of personalized disease related effects in AD patients. Method We adapt the pix2pix GAN to translate a subject with disease to a corresponding subject without disease. We train this model using a dataset comprising of healthy controls and synthetically simulated patients. Healthy controls (n = 6000) are selected from the ISTAGING consortium. Our neuroanatomical brain measures are the 10 region of interest (ROI) volumes (covering left and right frontal, parietal, occipital, temporal lobes as well as deep brain structures) computed using a multi‐atlas segmentation technique. To simulate patients, for each healthy control we introduce 10‐30% atrophy or expansion in a random combination of ROIs while preserving clinical covariate effects. The model learns to synthesize patient‐specific controls by removing disease‐related variations from patient’s brain measures (Fig.1). Deviation of the patient from the synthesized disease‐free control acts as an image‐based biomarker that is sensitive to disease effects and severity. For performance assessment, we select 200 controls (CN) and 200 AD participants (PT) from the OASIS dataset and compute their deviations across the 10 ROI volumes using the model pretrained on the ISTAGING dataset. Logistic regression is used to assess the overall discriminative power of the derived deviations in AD classification. Result Larger deviations in PT compared to CN suggest disease related abnormality in brain measures (Fig.2.a). Additionally, GAN’s deviations provide a considerable gain over raw ROI volumes in AD classification as quantified using the 5‐fold AUC scores (Fig.2.b). Conclusion GAN‐based normative modelling technique introduced here is a useful tool to parse heterogeneity in brain measures at an individual level. We see that self‐supervised training of the model using pseudo‐synthetically simulated patient data that is agnostic to disease patterns can help detect real disease related effects.

Background Game‐based Cognitive Assessment – 3‐Minute Version (G3) is a WeChat mini‐program developed for the early screening of cognitive impairments in Chinese older adults. It consists of three game‐like digital cognitive tests, namely “Number Ordering”, “Species Sorting”, and “Gold Finding”. This study aims to assess the sensitivity, specificity, and diagnostic accuracy of G3 in detecting mild cognitive impairment (MCI) and early‐stage dementia in Chinese older adults. Method Between February and December 2023, a total of 475 older adults (aged 50 to 89 years) with subjective cognitive complains were recruited. Participants underwent G3 assessment, neuropsychological assessments, neuroimaging, and blood tests. Among them, 230 individuals were diagnosed with cognitive impairments, including MCI and early‐stage dementia, while the remaining 245 older adults exhibited healthy cognition. We compared the G3 performance between the two groups. And the receiver operating characteristic (ROC) curve of G3 were evaluated with sensitivity and specificity calculated. Result The average G3 scores (mean ± standard deviation) in the healthy cognition group (63.2±13.04) were significantly higher than impaired cognition group (41.3±12.50, p<0.001). In both groups, the mean G3 scores tended to decline with increasing age. Meanwhile, the decline was more significant in the impaired cognition group compared to the healthy cognition group (Figure 1). G3 demonstrated high diagnostic accuracy in detecting individuals with MCI or early‐stage dementia, with an area under the ROC curve (AUC) of 0.887 (95% confidence interval [CI] = 0.859‐0.916, Figure 2). The optimal cut‐off score for G3 was determined to be 59.5, with a sensitivity of 0.913 and specificity of 0.624. Conclusion G3 WeChat mini‐program is a precise and user‐friendly screening tool that exhibits high sensitivity and diagnostic accuracy in detecting early‐stage cognitive impairment among community‐dwelling seniors. Further multicenter research is needed to verify the diagnostic value of G3 for different stages and types of cognitive impairments.

Background Persons with cognitive impairment may experience difficulties with language and cognition that interfere with their ability to make and communicate decisions. We developed an online visual tool to facilitate conversations about their preferences concerning supportive care. Methods We conducted Zoom interviews with persons with mild cognitive impairment (MCI) and mild to moderate dementia, using storytelling and a virtual tool designed to facilitate discussion. Each interview sought to discuss decision‐making in the context of three scenarios, one concerning a past decision about supportive care, and two hypothetical scenarios based on decreasing abilities and increasing care needs. Interviewers shared their screen with the visual tool, which used the visual diagramming platform draw.io (https://www.drawio.com/) to create an online “canvas.” Interviewers could drag icons (representing people, places, safety, emotions, activities, and issues) or type text onto the canvas to communicate concepts. We developed the tool with a multidisciplinary team of designers and researchers, and with input from people with cognitive impairment. Key user‐centered design goals for the tool included focusing on simplicity, minimizing cognitive load, and using concepts and images that resonated with participants. We also used pilot interviews to refine interviewers' techniques for eliciting preferences with the tool. We examined the utility of the tool for engaging participants in dialogic interactions with the interviewer, and the cognitive and communicative processes exhibited by participants. Results We conducted fifteen interviews with persons with MCI, and mild or moderate dementia [mean age 77.8 years ± 7.1 (SD), 9 male, 6 female]. With respect to dialogic interactions, interviewers actively used the visualization technique to facilitate the conversation; participants engaged with the interface to varying degrees. Common communicative issues included participants being unresponsive to the question or providing unclear responses. During the sessions participants also exhibited uncertainty, introspection, and self‐awareness. Conclusion We present a visual technique to engage persons with cognitive impairment in dialogue about complex decisions. We hope to explore the potential use of this technique in research, clinical, and community settings.