Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Correlating continuously captured home-based digital biomarkers of daily function with postmortem neurodegenerative neuropathology

  • Nathan C. Hantke ,

    Roles Conceptualization, Writing – original draft, Writing – review & editing

    Affiliations Department of Neurology, Oregon Health & Science University, Portland, OR, United States of America, Oregon Center for Aging & Technology (ORCATECH), Portland, OR, United States of America, Mental Health and Clinical Neuroscience Division, VA Portland Health Care System, Portland, OR, United States of America

  • Jeffrey Kaye,

    Roles Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Writing – original draft, Writing – review & editing

    Affiliations Department of Neurology, Oregon Health & Science University, Portland, OR, United States of America, Oregon Center for Aging & Technology (ORCATECH), Portland, OR, United States of America

  • Nora Mattek,

    Roles Data curation, Formal analysis, Methodology, Project administration, Writing – original draft, Writing – review & editing

    Affiliations Department of Neurology, Oregon Health & Science University, Portland, OR, United States of America, Oregon Center for Aging & Technology (ORCATECH), Portland, OR, United States of America

  • Chao-Yi Wu,

    Roles Conceptualization, Writing – original draft, Writing – review & editing

    Affiliations Department of Neurology, Oregon Health & Science University, Portland, OR, United States of America, Oregon Center for Aging & Technology (ORCATECH), Portland, OR, United States of America, Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, United States of America

  • Hiroko H. Dodge,

    Roles Conceptualization, Funding acquisition, Investigation, Writing – original draft, Writing – review & editing

    Affiliations Department of Neurology, Oregon Health & Science University, Portland, OR, United States of America, Oregon Center for Aging & Technology (ORCATECH), Portland, OR, United States of America, Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, United States of America

  • Zachary Beattie,

    Roles Conceptualization, Methodology, Project administration, Writing – original draft, Writing – review & editing

    Affiliations Department of Neurology, Oregon Health & Science University, Portland, OR, United States of America, Oregon Center for Aging & Technology (ORCATECH), Portland, OR, United States of America

  • Randy Woltjer

    Roles Conceptualization, Formal analysis, Investigation, Writing – original draft, Writing – review & editing

    Affiliation Department of Pathology and Laboratory Medicine, Oregon Health & Science University, Portland, OR, United States of America



Outcome measures available for use in Alzheimer’s disease (AD) clinical trials are limited in ability to detect gradual changes. Measures of everyday function and cognition assessed unobtrusively at home using embedded sensing and computing generated “digital biomarkers” (DBs) have been shown to be ecologically valid and to improve efficiency of clinical trials. However, DBs have not been assessed for their relationship to AD neuropathology.


The goal of the current study is to perform an exploratory examination of possible associations between DBs and AD neuropathology in an initially cognitively intact community-based cohort.


Participants included in this study were ≥65 years of age, living independently, of average health for age, and followed until death. Algorithms, run on the continuously-collected passive sensor data, generated daily metrics for each DB: cognitive function, mobility, socialization, and sleep. Fixed postmortem brains were evaluated for neurofibrillary tangles (NFTs) and neuritic plaque (NP) pathology and staged by Braak and CERAD systems in the context of the “ABC” assessment of AD-associated changes.


The analysis included a total of 41 participants (M±SD age at death = 92.2±5.1 years). The four DBs showed consistent patterns relative to both Braak stage and NP score severity. Greater NP severity was correlated with the DB composite and reduced walking speed. Braak stage was associated with reduced computer use time and increased total time in bed.


This study provides the first data showing correlations between DBs and neuropathological markers in an aging cohort. The findings suggest continuous, home-based DBs may hold potential to serve as behavioral proxies that index neurodegenerative processes.


The neurodegenerative disorder Alzheimer’s disease (AD) currently affects approximately one in nine persons age 65 years or older in the United States of America, a number that is expected to rise as the current population ages [1]. AD is characterized by a progressive decline in cognitive function, reduction in functional abilities, and neuropathological markers that include neurofibrillary tangles (NFTs) and neuritic plaques (NPs) [25].

The expanding science behind AD pathogenesis is promising, but early detection continues to prove complex. Subtle cognitive change and decline in instrumental activities of daily living (IADLs) are often early signals of future dementia [6,7]. Monitoring changes in cognitive status is generally achieved through repeated clinical visits. Episodic clinical assessments such as cognitive screeners often lack sufficient ecological validity to generalize to real-world settings by capturing only one point in time and in a setting that does not indicate how a person functions in his/her daily environment [811]. Similarly, IADL questionnaires do not account well for within-person variability, are by their nature subjective, and often do not capture decreased efficiency for completing daily tasks.

Monitoring behavior in the home using remote sensing and digital technologies addresses many of the validity concerns of currently used methods without disrupting usual routines [12,13]. High data capture frequency from passive sensors provide digital biomarkers [DBs], defined as objective, quantifiable physiological and behavioral data collected and measured by means of digital devices [1315]. There is growing empirical evidence that passive monitoring of daily activities, such as changes in daily computer use, mobility about the home, medication-taking, sleep routines, phone use, and driving, provides insight into every day cognitive function [1620].

DBs have demonstrated an ability to assess change in daily function over time in older adults who are cognitively intact and in those with clinically diagnosed MCI [13,14], yet there remains a gap in understanding the relationship of these objective functional changes (i.e., DBs) and the underlying brain pathology. A prior cross-sectional study found a significant relationship between less daily computer use and medial temporal lobe atrophy [21], a brain region that is known to be affected early-on in AD pathologically. This finding provided indirect, in vivo evidence of a link between DBs and AD, but did not directly measure the gold standard of post-mortem pathology data [5].

Few autopsy-based studies exist that examine a direct link between objective functional activity measures and underlying neuropathology. Studies have examined the relationship between measured physical activity, cognition, and brain pathology among older adults [22,23]. Another study observed that lower accelerometer measured physical activity was associated with brain pathologies [24]. However, studies related to more complex activities of daily living assessed naturalistically are lacking. With this background in mind, we aimed to determine the association of DBs to AD neuropathology in an initially non-demented, longitudinally-monitored, community-based cohort. Secondly, we examined the association between objective DBs with antemortem global cognition via Mini Mental State Examination (MMSE), functional status via Functional Activities Questionnaire (FAQ), and AD neuropathology.



Forty-one participants were included in the analysis. Inclusion criteria at study onset was age 65 years and older, in average health for age without poorly controlled medical illnesses, not demented at study entry (Mini-Mental State Examination [MMSE] scores >24) [25], self-report of being able to use a computer proficiently, and living independently (12). Assessment of baseline health was based upon review of participants’ medical history, medication lists, and completion of the modified Cumulative Illness Rating Scale [26,27]. Medical illnesses with the potential to limit physical participation (e.g., wheelchair bound) or likely to lead to death over the course of 35 months (such as certain cancers) were study exclusions. All participants completed annual clinical evaluations, including administration of the Clinical Dementia Rating (CDR) scale [28] at initial and subsequent visits to monitor for the presence of MCI and transition to dementia, and were followed until death. All participants provided written informed consent and had been previously enrolled in ongoing longitudinal studies of aging and in-home monitoring ( Participants were recruited from the Portland, Oregon metropolitan area through advertisement and presentations at local retirement communities. The study protocols were approved by the Oregon Health & Science University Institutional Review Board (Life Laboratory (LL) IRB #2765; ISAAC IRB #2353). All procedures involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. All cited articles in this manuscript contain human and/or animal work approved by institutional review boards prior to publication. Additional details of the sensor systems and study protocols have been published elsewhere [12,29]. Data were collected between the years 2008 to 2018. During this time period, 65% of the cohort participants died and went to brain autopsy.

Digital biomarker activity metrics

All of the participants had an unobtrusive, pervasive technology platform installed in their home consisting of X10 passive infrared (PIR) motion and X10 door contact sensors, and computer use monitoring software (Fig 1) (12). Algorithms, run on the continuously-collected passive sensor data from the technology platform, generated daily functional metrics for each participant. From the array of DBs, four measures representing four domains of functioning were selected based on prior research demonstrating differences in these measures during everyday life in those with MCI compared to those with normal cognition, as well as their key roles in gauging functional ability: (1) cognitive function based on frequency of computer use [19]; (2) mobility based on daily mean walking speed [30,31]; (3) socialization based on time out of home [32]; and, (4) duration of time in bed [20].

Computer use was measured by the number of days participants used his/her computer during the past year. Daily mean walking speed (cm/s) was measured using an array of in-home sensors which passively identified how quickly and frequently participants were passing under a sensor line [31,3336]. Algorithms estimating the speed of walking from the in-home sensor data have been validated against a ‘gold standard’ gait mat [34,37]. Time out of home (ie., total time spent out of the home per day in hours) was measured using the PIR motion sensors and door contact sensors, which were able to detect activity (or lack thereof) in the home, and the door openings and closings [32,35]. Data from PIR motion sensors in each room of the home, including the bedroom, were used to measure total time in bed, in hours [38]. Only participants living alone were included in order to clearly disambiguate in-home movement of multiple residents.

Neuropathologic data

Fixed postmortem brains were evaluated for NFTs and NP pathology and staged by Braak [39] and Consortium to Establish a Registry for AD (CERAD) systems [40]. Brains were fixed in neutral-buffered formaldehyde solution for at least two weeks and examined grossly, as well as microscopically. For microscopic evaluation, tissue samples were taken from all cortical lobes bilaterally or unilaterally, frontal lobe white matter, anterior cingulate gyrus, hippocampus, amygdala, bilateral striatum and thalamus, midbrain, pons, medulla, and cerebellum. Six-micrometer sections were routinely stained with hematoxylin-eosin and Luxol fast blue. Selected sections of hippocampus and neocortical regions were immunostained using PHF1 antibody to tau and additional sections were stained to determine the presence of beta-amyloid (4G8 antibody, Biolegend, San Diego, CA), alpha-synuclein (MJFR1 antibody, Abcam, Waltham, MA), and TDP-43 (1D3 antibody, Biolegend, San Diego, CA). Clinical and pathologic diagnoses were established using current consensus criteria [4145]. Information related to NP and NFT burdens, amyloid angiopathy, large vessel strokes or lacunes, presence of Lewy bodies (LBs), hippocampal sclerosis (HS), and degree of arteriosclerosis were summarized using the National Alzheimer’s Coordinating Center (NACC) Neuropathology Data reporting format [46]. The NACC Neuropathology Form changed versions (versions 9, 10, and 11) over the course of the study, resulting in hippocampal sclerosis data only being available for a limited number of subjects (n = 19).

Statistical analysis

Summary statistics were generated for participant characteristics and pathologic variables. A normally distributed composite DB measure including the four activity domains (cognition, mobility, socialization, and sleep/time in bed) was constructed by z-score normalizing the four individual domain metrics. Faster walking speed, more time out of home, more days with computer use, and less total time in bed contributed to a higher composite DB score. Data analysis was conducted using the home-monitored data from the 12-month period of available sensor information prior to death to avoid measuring acute, end of life changes in activity.

Differences in DBs according to individual neuropathological categories (e.g., Braak stage, plaque severity) and composite DB score were presented visually as box plots in Figs 24. Independent t-tests, spearman rank (non-parametric) correlations, and linear regression models were generated when appropriate to examine the association between neuropathological categories and the DB composite metric (Table 2; S1 Table), as well as neuropathological categories, antemortem global cognition (last MMSE before death), and a functional measure (FAQ at last visit before death). Due to a small sample size we were unable to control for covariates. Analyses were performed using SAS software 9.4 (Cary, NC).

Fig 2. Distributions of multi-domain activity level by neuritic plaque score and Braak score.

a. Box plots of the distribution of multi-domain activity level by neuritic plaque score (p = .01). b. Box plots of the distribution of multi-domain activity level by Braak score (p = .16).

Fig 3. Digital biomarkers and Braak scores.

(A) Cognition by Braak score, measured in days with Computer use; (B) Mobility by Braak score, measured in M walk speed (cm/s); (C) Sleep by Braak score, measured in M time spent in bed; (D) Socialization by Braak score, as measured by M time out of home (TOH).

Fig 4. Digital biomarkers and neuritic plaque scores.

(A) Cognition by plaque score, measured in days with Computer use; (B) Mobility by plaque score, measured in M walk speed (cm/s); (C) Sleep by plaque score, measured in M time spent in bed (TST); (D) Socialization by plaque score, as measured by M time out of home (TOH).


Characteristics of the 41 participants are described in Table 1. Cohort mean age at death was 92.2 years. Thirty-two percent (n = 13) of participants were ApoE ε4 carriers; the sample size and number of variables included in the analysis did not allow for additional sub analysis including ε4 status. Study participants had sensors in their home for an average of 5.8 years (2.4); median time from last DB home monitoring data and death was one day. Twenty-three participants (56%) died while their home was sensored; median last MMSE score before death was 27.0 (5.9). A subset of participants had each of the four individual DBs available; walking speed (n = 36), time out of home (n = 40), total time in bed (n = 35) and computer use (n = 29). Twenty-three participants (56%) had all four individual DBs available to calculate the DB composite score. There were no significant differences in age, gender, education or antemortem clinical diagnosis between participants with (n = 23) and those without (n = 18) DBs data available to create the composite measure.

Table 1. Participant demographic, clinical, and digital biomarker characteristics.

The composite z-score is normally distributed; K-S goodness of fit test D(23) = 0,13; p>0.15. The reasons for missing DBs included the in-home sensor technology being removed for various reasons (e.g., participant moved from independent living to assisted living) or the participant being hospitalized for the last several months of his or her life. These patients continued to be clinically followed, but did not have sensor data for that time period, which resulted in a gap between the last sensor data and death. In order to determine the potential impact of this gap in data collection, Spearman rank correlations were rerun removing outliers, defined as participants with greater than 2 years between sensor data collection and death (remaining n = 17); all results remained significant. Participants’ antemortem clinical diagnoses, based on clinician evaluation at last research visit prior to death, were: cognitively normal (46%), MCI (39%), and dementia (15%). Causes of death was available for 40 participants, and included cardiovascular-related (n = 15), pneumonia/inanition (n = 9), cancer (n = 8), unknown (n = 4), acute organ failure (n = 2), and suicide (n = 2).

On neuropathological evaluation, no participants were Braak stage zero. For the statistical analysis, Braak stages were categorized into three groups: I/II (n = 7), III/IV (n = 29), and V/VI (n = 5). Among this cohort, 83% were found to have Braak stage III or higher NFTs on autopsy. Twenty percent (n = 8) were found to have moderate/frequent neuritic plaques while 80% had none or sparse neuritic plaques (Table 1). Other pathologies were relatively infrequent: large vessel stroke or lacunar stroke (17%), hippocampal sclerosis (5%) and Lewy bodies (7%).

The DB composite score significantly predicted NP severity (R2 = 0.36, F(2, 20) = 5.66, p = 0.01). Fig 2A; S1 Table), but not Braak staging (R2 = 0.14, F(2, 20) = 1.68, p = 0.21; Fig 2B). In the model examining DB composite score by NP severity, while those with sparse plaques (Beta = -0.43, SE = 0.26, t = -1.68, p = 0.11) were not significantly different than the control group (no plaques), those with moderate/frequent neuritic plaques had a significantly lower / poorer DB composite score (Beta = -0.90, SE = 0.27, t = -3.34, p<0.01). Global cognition at death (as measured by latest annual MMSE score) did not discriminate between NP severity (R2 = 0.09, F(2, 38) = 1.90, p = 0.16) or Braak stages (p = 0.42). Functional status (as measure by last FAQ score) also did not discriminate between NP severity or Braak stages.

When the postmortem pathology variables were treated as ordinal variables, higher (worse) Braak stage was significantly correlated with fewer number of days with computer use (ρ = -0.438, p = 0.018) and more total time in bed (ρ = 0.395, p = 0.019; Table 2). Higher (worse) NP severity was significantly correlated with slower walking speed (ρ = -0.379 p = 0.023) and a lower DB composite score (ρ = -0.555 p = 0.006). Braak score and plaque severity were not different among those with computer use (n = 29) DBs and those without (n = 12). The DB composite score did not significantly differ between participant groups with or without post mortem evidence of infarction or stroke (n = 7 with infarction; t(6) = -0.57, p = 0.57). Other pathologies noted above (hippocampal sclerosis and presence of Lewy Bodies) were too infrequent within the sample to be engaged in further analysis.

Table 2. Correlations between digital biomarker activity metrics and postmortem pathology.


The current exploratory study provides the first data examining correlations between digital biomarkers (DBs) of daily functioning and neuropathological markers in an aging cohort, extending beyond established associations of DBs with clinical diagnoses [19] and providing a potentially important keystone in examining decline in older adults via passive monitoring.

A composite of DBs of daily function, as well as individual DBs, were correlated with neuropathological findings even in individuals whose cognition was not significantly impaired at the time of measurement. These correlations were not present between MMSE and Braak stage, which has been reported in other studies [46]. This lack of correlation in our study is likely a reflection of the predominantly low to intermediate stage of neurofibrillary tangle pathology in our sample. Specifically, the majority (70.7%) of the participants in the current study were in an intermediate, Braak stage III/IV, with only 5% in Braak stage V/VI. Other studies which have examined the relationship of Braak stage to MMSE have also not found a relationship between MMSE and Braak stage III/IV [47].

Taken together, these preliminary results suggest DBs may be more sensitive at detecting neuropathological findings than commonly used cognitive screeners, self-report questionnaires, or clinical diagnosis, with the potential to provide useful information in clinical and research settings. Thus, DBs, particularly DB composite metrics, may hold significant promise in detecting incipient behavioral or functional changes in AD. However, the present cross-sectional findings require additional longitudinal studies in order to confirm these findings and importantly, to determine the trajectory and timing of DB changes relative to underlying neuropathologic change. Given the growing availability of in vivo AD neuropathological biomarkers (blood-based and imaging), the correlation between DBs and early AD pathologic change during life is suggested as a promising future avenue of study to substantiate the clinical utility of these DBs to reflect early stage AD pathology.

Although we identified specific DBs which were significantly correlated with one neuropathology but not the other (e.g., computer use with Braak stage but not NP severity), in general the sample sizes of the individual groups available for analysis limit the ability to make definitive statements about these relationships at this time. Nevertheless, we note that the neuropathologic change in this sample was not severe nor extensive; 80% had none or sparse neuritic plaques, and 88% were below Braak stages V/VI (neocortical neurofibrillary tangle involvement). Thus, these DB observations have been made in older adults with relatively mild to moderate pathological change consistent with the current conceptualization of amyloid and tau accumulation likely occurring well before the presence of functional and cognitive changes are detected with conventional clinical tests [5,48,49]. Changes in specific DBs that may preferentially utilize a number of brain networks, are likely to reflect disruption of these networks as the complex, slowly evolving, and regionally progressing neuropathological process plays out over time. Thus for example, the interplay of tau or neurofibrillary change with amyloid deposition would be expected to lead to possible bidirectional effects on sleep behaviors where there is a balancing between amyloid and tau aggregation [50] that depending on the timing and distribution of these processes, may result in disturbed sleep that is manifested by time in bed or other measures such as restlessness [51] or sleep efficiency [52].

NFT count is a stronger predictor of functioning than amyloid accumulation [53,54], which may be reflected in the present finding of decrease in the cognitively demanding task of computer use correlated with Braak staging. The relationship between specific DB and neuropathology type is worth consideration of exploration in future studies.

This study includes several limitations that can be addressed in future studies. First, the diversity of the sample is limited in terms of race, gender, and educational attainment. Second, while DBs have been shown to potentially yield clinically significant outcomes in longitudinal studies with relatively small samples [37], the sample size of the reported study is small and findings should thus be considered preliminary. A larger sample would have allowed for additional analyses, such as examining the effect of potential covariates and important predictors of cognitive decline that could be investigated further in the context of the noted DBs, including but not limited to family history of neurodegenerative disease, cardio- and cerebrovascular risk factors, APOE genotyping, and polypharmacy. It is also possible that some DBs have a more complex relationship with daily functioning than examined in the current models. Future DB-pathological correlational studies may consider these alternative models and consider changes in home-based activity measured as intradaily stability, variability, as well as spatio-temporal extent captured over time [5557].

Third, the obtainment of DBs requires several factors that may limit accessibility, such requiring participants to have reliable internet, which may be problematic in some rural settings. Fourth, staging of AD pathology is dependent on the utilized neuropathological scales. This study used the Braak and CERAD systems, which is a combination recommended by the National Institute on Aging and Reagan Institute. However, staging may vary should investigators use the Poly-Pathology AD assessment (PPAD9), which focuses more intently on cytoarchitectural disorder and gliosis, microvacuolization, and degree of neuronal degeneration in nine cerebral areas, along with NPs and NFTs [58].

Overall, findings of this novel study suggest that DBs of daily function hold potential to serve as behavioral proxies for assessing pre-dementia pathological findings. In the context of suboptimal conventions for early detection of cognitive dysfunction, functional decline, and clinical diagnosis, DBs may bridge an important gap in the detection and treatment of neurodegenerative processes in pre-dementia phases.

Supporting information

S1 Table.

a. Linear regression model showing association between DB composite score and Braak stages (n = 23). b. Linear regression model showing association between DB composite score and Neuritic plaque severity (n = 23).



  1. 1. Hebert LE, Weuve J, Scherr PA, Evans DA. Alzheimer disease in the United States (2010–2050) estimated using the 2010 census. Neurology. 2013;80(19):1778–83. pmid:23390181
  2. 2. Bejanin A, Schonhaut DR, La Joie R, Kramer JH, Baker SL, Sosa N, et al. Tau pathology and neurodegeneration contribute to cognitive impairment in Alzheimer’s disease. Brain. 2017;140(12):3286–300. pmid:29053874
  3. 3. Cicognola C, Brinkmalm G, Wahlgren J, Portelius E, Gobom J, Cullen NC, et al. Novel tau fragments in cerebrospinal fluid: relation to tangle pathology and cognitive decline in Alzheimer’s disease. Acta Neuropathol. 2019;137(2):279–96. pmid:30547227
  4. 4. McGurran H, Glenn JM, Madero EN, Bott NT. Prevention and Treatment of Alzheimer’s Disease: Biological Mechanisms of Exercise. J Alzheimers Dis. 2019;69(2):311–38. pmid:31104021
  5. 5. Jack CR jr, Bennett DA, Blennow K, Carrillo MC, Dunn B, Haeberlein SB, et al. NIA-AA Research Framework: Toward a biological definition of Alzheimer’s disease. Alzheimers Dement. 2018;14(4):535–62. pmid:29653606
  6. 6. Peres K, Helmer C, Amieva H, Orgogozo JM, Rouch I, Dartigues JF, et al. Natural history of decline in instrumental activities of daily living performance over the 10 years preceding the clinical diagnosis of dementia: a prospective population-based study. J Am Geriatr Soc. 2008;56(1):37–44. pmid:18028344
  7. 7. Tomaszewski Farias S, Giovannetti T, Payne BR, Marsiske M, Rebok GW, Schaie KW, et al. Self-perceived Difficulties in Everyday Function Precede Cognitive Decline among Older Adults in the ACTIVE Study. J Int Neuropsychol Soc. 2018;24(1):104–12. pmid:28797312
  8. 8. Burgess PW, Alderman N, Forbes C, Costello A, Coates LM, Dawson DR, et al. The case for the development and use of "ecologically valid" measures of executive function in experimental and clinical neuropsychology. J Int Neuropsychol Soc. 2006;12(2):194–209. pmid:16573854
  9. 9. Poncet F, Swaine B, Dutil E, Chevignard M, Pradat-Diehl P. How do assessments of activities of daily living address executive functions: A scoping review. Neuropsychol Rehabil. 2017;27(5):618–66. pmid:28075219
  10. 10. Edmonds EC, Delano-Wood L, Galasko DR, Salmon DP, Bondi MW, Alzheimer’s Disease Neuroimaging I. Subtle Cognitive Decline and Biomarker Staging in Preclinical Alzheimer’s Disease. J Alzheimers Dis. 2015;47(1):231–42. pmid:26402771
  11. 11. Farias ST, Mungas D, Reed BR, Cahn-Weiner D, Jagust W, Baynes K, et al. The measurement of everyday cognition (ECog): scale development and psychometric properties. Neuropsychology. 2008;22(4):531–44. pmid:18590364
  12. 12. Kaye JA, Maxwell SA, Mattek N, Hayes TL, Dodge H, Pavel M, et al. Intelligent Systems For Assessing Aging Changes: home-based, unobtrusive, and continuous assessment of aging. J Gerontol B Psychol Sci Soc Sci. 2011;66 Suppl 1:i180–90. pmid:21743050
  13. 13. Lussier M, Lavoie M, Giroux S, Consel C, Guay M, Macoir J, et al. Early Detection of Mild Cognitive Impairment With In-Home Monitoring Sensor Technologies Using Functional Measures: A Systematic Review. IEEE J Biomed Health Inform. 2019;23(2):838–47. pmid:29994013
  14. 14. Piau A, Wild K, Mattek N, Kaye J. Current State of Digital Biomarker Technologies for Real-Life, Home-Based Monitoring of Cognitive Function for Mild Cognitive Impairment to Mild Alzheimer Disease and Implications for Clinical Care: Systematic Review. J Med Internet Res. 2019;21(8):e12785. pmid:31471958
  15. 15. Babrak LM, Menetski J, Rebhan M, Nisato G, Zinggeler M, Brasier N, et al. Traditional and Digital Biomarkers: Two Worlds Apart? Digit Biomark. 2019;3(2):92–102. pmid:32095769
  16. 16. Seelye A, Hagler S, Mattek N, Howieson DB, Wild K, Dodge HH, et al. Computer mouse movement patterns: A potential marker of mild cognitive impairment. Alzheimers Dement (Amst). 2015;1(4):472–80. pmid:26878035
  17. 17. Seelye A, Mattek N, Howieson D, Austin D, Wild K, Dodge H, et al. Embedded online questionnaire measures are sensitive to identifying mild cognitive impairment. Alzheimer’s Disease & Associated Disorders, Epub ahead of print, 2015.
  18. 18. Seelye A, Mattek N, Sharma N, Riley T, Austin J, Wild K, et al. Weekly observations of online survey metadata obtained through home computer use allow for detection of changes in everyday cognition before transition to mild cognitive impairment. Alzheimer’s & dementia: the journal of the Alzheimer’s Association. 2017. pmid:29107052
  19. 19. Kaye J, Mattek N, Dodge HH, Campbell I, Hayes T, Austin D, et al. Unobtrusive measurement of daily computer use to detect mild cognitive impairment. Alzheimers Dement. 2014;10(1):10–7. pmid:23688576
  20. 20. Beattie Z, Miller LM, Almirola C, Au-Yeung WM, Bernard H, Cosgrove KE, et al. The Collaborative Aging Research Using Technology Initiative: An Open, Sharable, Technology-Agnostic Platform for the Research Community. Digit Biomark. 2020;4(Suppl 1):100–18. pmid:33442584
  21. 21. Silbert LC, Dodge HH, Lahna D, Promjunyakul NO, Austin D, Mattek N, et al. Less Daily Computer Use is Related to Smaller Hippocampal Volumes in Cognitively Intact Elderly. J Alzheimers Dis. 2016;52(2):713–7. pmid:26967228
  22. 22. Dawe RJ, Yu L, Leurgans SE, James BD, Poole VN, Arfanakis K, et al. Physical activity, brain tissue microstructure, and cognition in older adults. PLoS One. 2021;16(7):e0253484. pmid:34232955
  23. 23. Won J, Callow DD, Pena GS, Jordan LS, Arnold-Nedimala NA, Nielson KA, et al. Hippocampal Functional Connectivity and Memory Performance After Exercise Intervention in Older Adults with Mild Cognitive Impairment. J Alzheimers Dis. 2021;82(3):1015–31. pmid:34151792
  24. 24. Buchman AS, Yu L, Wilson RS, Lim A, Dawe RJ, Gaiteri C, et al. Physical activity, common brain pathologies, and cognition in community-dwelling older adults. Neurology. 2019;92(8):e811–e22. pmid:30651386
  25. 25. Folstein MF, Folstein SE, McHugh PR. "Mini-mental state". A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12(3):189–98. pmid:1202204
  26. 26. Linn BS, Linn MW, Gurel L. Cumulative illness rating scale. J Am Geriatr Soc. 1968;16(5):622–6. pmid:5646906
  27. 27. Miller MD, Paradis CF, Houck PR, Mazumdar S, Stack JA, Rifai AH, et al. Rating chronic medical illness burden in geropsychiatric practice and research: application of the Cumulative Illness Rating Scale. Psychiatry Res. 1992;41(3):237–48. pmid:1594710
  28. 28. Morris JC. The Clinical Dementia Rating (CDR): current version and scoring rules. Neurology. 1993;43(11):2412–4. pmid:8232972
  29. 29. Lyons BE, Austin D, Seelye A, Petersen J, Yeargers J, Riley T, et al. Pervasive Computing Technologies to Continuously Assess Alzheimer’s Disease Progression and Intervention Efficacy. Front Aging Neurosci. 2015;7:102. pmid:26113819
  30. 30. Sullivan KJ, Ranadive R, Su D, Neyland BR, Hughes TM, Hugenschmidt CE, et al. Imaging-based indices of Neuropathology and gait speed decline in older adults: the atherosclerosis risk in communities study. Brain Imaging Behav. 2021;15(5):2387–96. pmid:33439369
  31. 31. Dodge HH, Mattek NC, Austin D, Hayes TL, Kaye JA. In-home walking speeds and variability trajectories associated with mild cognitive impairment. Neurology. 2012;78(24):1946–52. pmid:22689734
  32. 32. Petersen J, Austin D, Kaye JA, Pavel M, Hayes TL. Unobtrusive in-home detection of time spent out-of-home with applications to loneliness and physical activity. IEEE J Biomed Health Inform. 2014;18(5):1590–6. pmid:25192570
  33. 33. Kaye J, Mattek N, Dodge H, Buracchio T, Austin D, Hagler S, et al. One walk a year to 1000 within a year: continuous in-home unobtrusive gait assessment of older adults. Gait Posture. 2012;35(2):197–202. pmid:22047773
  34. 34. Hagler S, Austin D, Hayes TL, Kaye J, Pavel M. Unobtrusive and ubiquitous in-home monitoring: a methodology for continuous assessment of gait velocity in elders. IEEE Trans Biomed Eng. 2010;57(4):813–20. pmid:19932989
  35. 35. Austin J, Dodge HH, Riley T, Jacobs PG, Thielke S, Kaye J. A Smart-Home System to Unobtrusively and Continuously Assess Loneliness in Older Adults. IEEE J Transl Eng Health Med. 2016;4:2800311. pmid:27574577
  36. 36. Hayes TL, Abendroth F, Adami A, Pavel M, Zitzelberger TA, Kaye JA. Unobtrusive assessment of activity patterns associated with mild cognitive impairment. Alzheimers Dement. 2008;4(6):395–405. pmid:19012864
  37. 37. Dodge HH, Zhu J, Mattek NC, Austin D, Kornfeld J, Kaye JA. Use of High-Frequency In-Home Monitoring Data May Reduce Sample Sizes Needed in Clinical Trials. PLoS One. 2015;10(9):e0138095. pmid:26379170
  38. 38. Hayes TL, Riley T, Pavel M, Kaye JA. Estimation of rest-activity patterns using motion sensors. Annu Int Conf IEEE Eng Med Biol Soc. 2010;2010:2147–50. pmid:21097221
  39. 39. Braak H, Braak E. Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol. 1991;82(4):239–59. pmid:1759558
  40. 40. Mirra SS, Heyman A, McKeel D, Sumi SM, Crain BJ, Brownlee LM, et al. The Consortium to Establish a Registry for Alzheimer’s Disease (CERAD). Part II. Standardization of the neuropathologic assessment of Alzheimer’s disease. Neurology. 1991;41(4):479–86. pmid:2011243
  41. 41. Bennett DA, Schneider JA, Bienias JL, Evans DA, Wilson RS. Mild cognitive impairment is related to Alzheimer disease pathology and cerebral infarctions. Neurology. 2005;64(5):834–41. pmid:15753419
  42. 42. Beekly DL, Ramos EM, van Belle G, Deitrich W, Clark AD, Jacka ME, et al. The National Alzheimer’s Coordinating Center (NACC) Database: an Alzheimer disease database. Alzheimer Dis Assoc Disord. 2004;18(4):270–7. pmid:15592144
  43. 43. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology. 1984;34(7):939–44. pmid:6610841
  44. 44. Chui HC, Mack W, Jackson JE, Mungas D, Reed BR, Tinklenberg J, et al. Clinical criteria for the diagnosis of vascular dementia: a multicenter study of comparability and interrater reliability. Arch Neurol. 2000;57(2):191–6. pmid:10681076
  45. 45. Ferman TJ, Boeve BF, Smith GE, Lin SC, Silber MH, Pedraza O, et al. Inclusion of RBD improves the diagnostic classification of dementia with Lewy bodies. Neurology. 2011;77(9):875–82. pmid:21849645
  46. 46. Yu L, Boyle P, Schneider JA, Segawa E, Wilson RS, Leurgans S, et al. APOE epsilon4, Alzheimer’s disease pathology, cerebrovascular disease, and cognitive change over the years prior to death. Psychol Aging. 2013;28(4):1015–23.
  47. 47. Serrano-Pozo A, Qian J, Muzikansky A, Monsell SE, Montine TJ, Frosch MP, et al. Thal Amyloid Stages Do Not Significantly Impact the Correlation Between Neuropathological Change and Cognition in the Alzheimer Disease Continuum. J Neuropathol Exp Neurol. 2016;75(6):516–26. pmid:27105663
  48. 48. Monsell SE, Mock C, Hassenstab J, Roe CM, Cairns NJ, Morris JC, et al. Neuropsychological changes in asymptomatic persons with Alzheimer disease neuropathology. Neurology. 2014;83(5):434–40. pmid:24951474
  49. 49. Hardy J, Allsop D. Amyloid deposition as the central event in the aetiology of Alzheimer’s disease. Trends Pharmacol Sci. 1991;12(10):383–8. pmid:1763432
  50. 50. Wang C, Holtzman DM. Bidirectional relationship between sleep and Alzheimer’s disease: role of amyloid, tau, and other factors. Neuropsychopharmacology. 2020;45(1):104–20. pmid:31408876
  51. 51. Hayes TL, Riley T, Mattek N, Pavel M, Kaye JA. Sleep habits in mild cognitive impairment. Alzheimer Dis Assoc Disord. 2014;28(2):145–50. pmid:24145694
  52. 52. Ju YE, McLeland JS, Toedebusch CD, Xiong C, Fagan AM, Duntley SP, et al. Sleep quality and preclinical Alzheimer disease. JAMA Neurol. 2013;70(5):587–93. pmid:23479184
  53. 53. Pichet Binette A, Franzmeier N, Spotorno N, Ewers M, Brendel M, Biel D, et al. Amyloid-associated increases in soluble tau relate to tau aggregation rates and cognitive decline in early Alzheimer’s disease. Nat Commun. 2022;13(1):6635. pmid:36333294
  54. 54. Lagarde J, Olivieri P, Tonietto M, Tissot C, Rivals I, Gervais P, et al. Tau-PET imaging predicts cognitive decline and brain atrophy progression in early Alzheimer’s disease. J Neurol Neurosurg Psychiatry. 2022;93(5):459–67. pmid:35228270
  55. 55. Musiek ES, Bhimasani M, Zangrilli MA, Morris JC, Holtzman DM, Ju YS. Circadian Rest-Activity Pattern Changes in Aging and Preclinical Alzheimer Disease. JAMA Neurol. 2018;75(5):582–90. pmid:29379963
  56. 56. Wu CY, Dodge HH, Reynolds C, Barnes LL, Silbert LC, Lim MM, et al. In-Home Mobility Frequency and Stability in Older Adults Living Alone With or Without MCI: Introduction of New Metrics. Front Digit Health. 2021;3:764510. pmid:34766104
  57. 57. Wu CY, Dodge HH, Gothard S, Mattek N, Wright K, Barnes LL, et al. Unobtrusive Sensing Technology Detects Ecologically Valid Spatiotemporal Patterns of Daily Routines Distinctive to Persons With Mild Cognitive Impairment. J Gerontol A Biol Sci Med Sci. 2022;77(10):2077–84. pmid:34608939
  58. 58. Brunnström H, Englund E. Comparison of four neuropathological scales for Alzheimer’s disease. Clin Neuropathol. 2011;30(2):56–69. pmid:21329614