Scientific publications

Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring | February 23, 2024

Clinical classification of memory and cognitive impairment with multimodal digital biomarkers

This paper compared the ability of different types of digitally-collected data to accurately predict an individual’s level of cognitive impairment. 922 older adults completed a memory test (RAVLT) which was used to classify them into cognitively impaired and unimpaired groups. Then, participants completed two additional cognitive tests, the clock drawing test (CDT) and a verbal recall test, which were used to try and predict which group participants belonged to.

Several machine learning models were trained on the data to predict cognitive impairment status. The model that performed best was a multimodal model that was trained on the graphomotor data from the CDT and the memory, speech, and voice data from the verbal recall test. This model achieved higher accuracy (0.80), sensitivity (0.81; ability to specifically detect cognitive impairment), and specificity (0.80; ability to specifically detect absence of impairment) than single modal models.

Takeaways

• Cognitive impairment can be predicted with substantial accuracy via fast, accessible digital assessments – i.e., without the need for invasive, time-consuming, costly tests.
• Cognitive impairment prediction is most accurate when data across multiple sources, like graphomotor, speech, and voice measures, are combined together.
• Overall, for early and accurate detection of cognitive impairment, a multimodal assessment is recommended that includes the CDT and verbal recall.

Alzheimer's Research & Therapy | January 2, 2024

Digital Clock and Recall is superior to the Mini-Mental State Examination for the detection of mild cognitive impairment and mild dementia

Researchers studied 706 participants from the multisite Bio-Hermes study who were classified as either cognitively unimpaired, mild cognitive impairment (MCI), or probable mild Alzheimer’s dementia based on a review of medical history and selective cognitive and imaging tests. They evaluated cognitive classifications (MCI and early dementia) based on the Digital Clock and Recall (DCR™) and the Mini-Mental State Examination (MMSE) against cohorts based on the results of the Rey Auditory Verbal Learning Test (RAVLT), the Trail Making Test-Part B (TMT-B), and the Functional Activities Questionnaire (FAQ) while also comparing the influence of demographic variables such as race, ethnicity, and level of education.

The study found that the DCR was superior to the MMSE in classifying mild cognitive impairment and early dementia, and faster to administer as well. Additionally, the DCR was significantly less biased by ethnicity than the MMSE, with no significant difference in the DCR score between Hispanic and non-Hispanic individuals.

Frontiers in Aging Neuroscience | November 17, 2023

Using digital assessment technology to detect neuropsychological problems in primary care settings

The CCE is a powerful neurocognitive assessment tool that is sensitive to patient’s subjective concerns about possible decline in memory, mood symptoms, possible cognitive impairment, and cardiovascular risk. iPad administration ensures total reliability for test administration and scoring. The CCE is easily deployable in outpatient ambulatory primary care settings.

Frontiers in Aging Neuroscience | September 20, 2022

The functional connectivity and neuropsychology underlying mental planning operations: data from the digital clock drawing test

Performance on the DCTclock, specifically digit misplacement in the Command Clock, is associated with functional connectivity between key structures involved in executive functions and the cholinergic pathway in the brain: the basal nucleus of Meynert (BNM) and the anterior cingulate cortex (ACC). This study combined the DCTclock with resting-state functional brain imaging (fMRI) to explore a potential link between digit misplacement and the functional connectivity between the BNM and ACC.

The ACC is an important area of the brain involved with executive abilities like mental planning and the BNM is a major source of acetylcholine (ACH), a key neurotransmitter related to executive abilities often disrupted in patients with Alzheimer’s disease. The study showed an association between digital placement accuracy and BNM-ACC functional connectivity; individuals with MCI had greater digit misplacement on the Command Clock and lower BNM-ACC connectivity.

Frontiers in Aging Neuroscience | April 15, 2022

Associations Between the Digital Clock Drawing Test and Brain Volume: Large Community-Based Prospective Cohort (Framingham Heart Study)

DCTclock scores are associated with brain atrophy – an important physical marker of cognitive impairment – as measured by MRI, demonstrating its use as a practical tool to assess subtle cognitive changes occurring in MCI or early dementia stages. In a large community-based cohort (n=1656), linear regression models were used to assess the associations between 18 DCTclock composite scores and brain MRI measures related to atrophy, including total cerebral brain volume (TCBV) and both cerebral gray and white matter volumes. Therefore, DCTclock composite scores can predict imaging markers of cognitive impairment and have the potential to be used as a cognitive assessment tool in the clinical diagnosis of MCI.

Journal of International Neuropsychological Society | February 21, 2022

Dissociating Statistically Determined Normal Cognitive Abilities and Mild Cognitive Impairment Subtypes with DCTclock

DCTclock metrics not only accurately identify cognitively normal older adults from those with mild cognitive impairment (MCI), but also distinguish between various cognitive impairment subtypes: Subtle Cognitive Impairment (SCI); Amnestic Mild Cognitive Impairment (aMCI); and Mixed/dysexecutive Mild Cognitive Impairment (mx/dysMCI.). In a study of 103 participants, extensive examination and analysis (neuropsychological testing, MRI, bloodwork, and subjective interviews) was done to determine cognitive status (MCI vs. cognitively normal) as well as MCI subtype. Participants then completed the DCTclock assessment. Results of the DCTclock showed significant alignment with classification of MCI subtypes, suggesting that digital technology can contribute to more efficient, scalable diagnostic decision-making at critical early disease stages.

Frontiers in Digital Health | October 15, 2021

DCTclock: Clinically-Interpretable and Automated Artificial Intelligence Analysis of Drawing Behavior for Capturing Cognition

DCTclock offers significant improvement in the detection of early stages of cognitive impairment when compared to the Mini Mental State Exam (MMSE). In a large study (n= 591), participants who were known to be either cognitively normal or have mild cognitive impairment (MCI) took both the MMSE and the DCTclock assessments. When compared with MMSE, DCTclock demonstrated higher sensitivity for detecting MCI, accurately identifying many individuals with MCI that the MMSE mis-classified as cognitively normal.  Results suggest that when compared to MMSE, DCTclock may be a beneficial, more efficient, and more sensitive aid in the diagnostic process for mild cognitive impairment.

Journal of Medical Internet Research | August 6, 2021

Association Between the Digital Clock Drawing Test and Neuropsychological Test Performance: Large Community-Based Prospective Cohort (Framingham Heart Study)

The digital clock drawing test is a reliable assessment tool in detecting mild cognitive impairment. Composite scores derived from the digitial clock drawing test (dCDT) were significantly associated with both neuropsychological tests and mild cognitive impairment, demonstrating the potential use of the dCDT as a tool for cognitive assessment in large community-based populations. 

Parkinsonism & Related Disorders | August 4, 2021

Quantitative digital clock drawing test as a sensitive tool to detect subtle cognitive impairments in early stage Parkinson's disease

The use of quantitative digital cognitive assessment showed greater sensitivity in identifying subtle cognitive decline than the current standardized tests. Differences in cognitive profiles were observed based on genotype. The identification of early cognitive decline may improve the clinical management of PD patients and be useful for cognitive related clinical trials.

Journal of Alzheimer's Disease | June 29, 2021

Classifying Non-Dementia and Alzheimer's Disease/Vascular Dementia Patients Using Kinematic, Time-Based, and Visuospatial Parameters: The Digital Clock Drawing Test

The digital clock drawing test provides additional neurocognitive biomarkers that may be able to identify and track dementia syndromes. The digital clock drawing test obtains and operationally defines graphomotor output — such as kinematic, time-based, and visuospatial behavior — output not measured using traditional paper and pencil test methods with older healthy controls and participants with dementia.

Journal of Parkinson's Disease | April 13, 2021

Parkinson’s Disease Cognitive Phenotypes Show Unique Clock Drawing Features When Measured with Digital Technology

The DCTclock captures subtle clock drawing behaviors that can help distinguish between various Parkinson's diagnoses. Subtle digital clock drawing behaviors—such as drawing latency and graphomotor metrics—differ between Parkinson's Disease (PD) and non-PD peers, and distinguish between the PD cognitive phenotypes low executive (PDExe), cognitively well (PDWell), and low memory (PDMem).

Neurology | April 6, 2021

Association of Digital Clock Drawing With PET Amyloid and Tau Pathology in Normal Older Adults

DCTclock has a demonstrated association with key biomarkers of Alzheimer's Disease (AD), amyloid and tau (via PET scan), in clinically asymptomatic older adults. In addition, DCTclock showed excellent discrimination between diagnostic groups – clinically normal and mild cognitive impairment/early AD  – while requiring a fraction of the time than the PACC, a collection of standard neuropsychological tests researchers compared the DCTclock to in this 300-person study. Furthermore, among clinically normal individuals specifically, the DCTclock showed better discrimination between amyloid positive and negative participants than the PACC. Having set out to determine if a digital clock drawing test improves upon standard cognitive assessments for discriminating diagnostic groups and detecting biomarker evidence of amyloid and tau pathology in asymptomatic adults, researchers concluded that "DCTclock discriminates between diagnostic groups and improves upon traditional cognitive tests for detecting [these] biomarkers" among this group, emphasizing its accuracy and relatively short administration time.

Anesthesia & Analgesia | July 2019

Clock Drawing Performance Slows for Older Adults After Total Knee Replacement Surgery

The DCTclock provides insight into cognitive changes in patients who underwent knee surgery with general anethesia. Study findings show that digital clock drawing behaviors significantly slowed for individuals electing total knee arthroplasty surgery with general anesthesia, demonstrating that clock drawing digital technology can provide insight into perioperative cognitive changes.

Journal of Alzheimer's Disease | August 21, 2017

Age and Graphomotor Decision Making Assessed with the Digital Clock Drawing Test: The Framingham Heart Study

The digital clock drawing test (dCDT) detects biomarkers of Alzheimer's disease and other neurocognitive disorders. The dCDT examines neurocognitive constructs related to higher-order neurocognitive decision-making and information processing speed, as longer age-related decision-making latencies may reflect a greater need for working memory — a potential neurocognitive biomarker of AD.

Neuropsychologia | May 2016

Cognitive and connectome properties detectable through individual differences in graphomotor organization

Graphomotor organization, defined as using "anchor" digits (i.e., 12, 3, 6, 9) before any other digits while completing a digital Clock Drawing Test, was associated with differences in cognitive and connectome properties. Individuals who used anchor digits displayed 1) better performance in specific cognitive domains, 2) higher local efficiency suggesting better regional integration, and 3) more sophisticated modular integration involving the ventral (‘what’) visuospatial processing stream.

Machine Learning | October 20, 2015

Learning classification models of cognitive conditions from subtle behaviors in the digital Clock Drawing Test

The digital clock drawing test demonstrates the possibility of substantial improvement in detecting earlier cognitive impairment, through new machine learning methods and predictive modeling. The analysis suggests the potential use of the dCDT to make significant improvements in both cognitive screening and diagnosis, as well as the potential use of these methods and models in clinical practice.

Journal of International Neuropsychological Society | October 20, 2014

Digital clock drawing: Differentiating ‘thinking’ versus ‘doing’ in younger and older adults with depression

Regardless of age, the digital clock drawing test can detect signs of depression. Digital clock drawing metrics can detect signs of depression as indexed by psychomotor slowing. The dCDT metrics in younger adults with depression are similar to those in older adults without depression.

Journal of Multiple Sclerosis | September 1, 2014

Deficits in Processing Speed and Decision Making in Relapsing-Remitting Multiple Sclerosis: The Digit Clock Drawing Test (dCDT)

The digital clock drawing test can identify executive dysfunction known to be present in multiple sclerosis. The dCDT complements traditional clock scoring methods, captures behavior previously unobtainable, such as time elapsed between clock drawing components, average time between drawing numbers, and total drawing time, and is related to processing speed and dysexecutive impairment known to be present in MS.