Higher cardiovascular burden and peripheral inflammation are associated with small vessel vascular disease, a predominantly dysexecutive cognitive profile, and a higher likelihood of conversion to vascular dementia. The digital clock drawing test, a digitized version of a standard neuropsychological tool, is useful in identifying cognitive dysfunction related to vascular etiology. However, little is known about the specific cognitive implications of vascular risk, peripheral inflammation, and varying levels of overall brain integrity. The current study aimed to examine the role of cardiovascular burden, peripheral inflammation, and brain integrity on digitally acquired clock drawing latency and graphomotor metrics in non-demented older adults.

The final prospectively recruited IRB-consented participant sample included 184 non-demented older adults (age: 69±6 years, education: 16±3 years, 46% female, 94% white) who completed digital clock drawing, vascular assessment, blood draw, and brain MRI. Digital clock drawing variables of interest included: total completion time (TCT), pre-first hand latency (PFHL), digit misplacement, hour hand distance from center, and clock face area (CFA). Cardiovascular burden was calculated using the revised version of the Framingham Stroke Risk Profile (FSRP-10). Peripheral inflammation was operationalized using interleukin (IL)-6, IL-8, IL-10, tumor necrosis factor alpha (TNF-a), and high sensitivity C-reactive protein (hsCRP). The brain integrity composite was comprised of bilateral entorhinal cortex volume, bilateral ventricular volume, and whole brain leukoaraiosis.

Over and above age and cognitive reserve, hierarchical regressions showed FSRP-10, inflammatory markers, and brain integrity explained an additional 13.3% of the variance in command TCT (p< 0.001), with FSRP-10 (p=0.001), IL-10 (p= 0.019), and hsCRP (p= 0.019) as the main predictors in the model. FSRP-10, inflammatory markers, and brain integrity explained an additional 11.7% of the variance in command digit misplacement (p= 0.009), with findings largely driven by FSRP-10 (p< 0.001).

Overall, in non-demented older adults, subtle behavioral nuances seen in digital clock drawing metrics (i.e., total completion time and digit misplacement) are partly explained by cardiovascular burden, peripheral inflammation, and brain integrity over and above age and cognitive reserve. These nuanced behaviors on digitally acquired clock drawing may associate with an emergent disease process or overall vulnerability.

Funding sources: Barber Fellowship; K07AG066813; R01 AG055337; R01 NR014810; American Psychological Foundation Dissertation Award; APA Dissertation Research Award

Research shows that highly educated individuals have at least 20 graphomotor features associated with clock drawing with hands set for '10 after 11' (Davoudi et al., 2021). Research has yet to understand clock drawing features in individuals with fewer years of education. In the current study, we compared older adults with < 8 years of education to those with > 9 years of education on number and pattern of graphomotor feature relationships in the clock drawing command condition.

Participants age 65+ from the University of Florida (UF) and UF Health (N= 10,491) completed both command and copy conditions of the digital Clock Drawing Test (dCDT) as a part of a federally-funded investigation. Participants were categorized into two education groups: < 8 years of education (n= 304) and > 9 years of education (n= 10,187). Propensity score matching was then used to match participants from each subgroup (n= 266 for each subgroup) on the following demographic characteristics: age, sex, race, and ethnicity (n= 532, age= 74.99±6.21, education= 10.41±4.45, female= 42.7%, non-white= 32.0%). Network models were derived using Bayesian Structure Learning (BSL) with the hill-climbing algorithm to obtain optimal directed acyclic graphs (DAGs) from all possible solutions in each subgroup for the dCDT command condition.

Both education groups retained 13 of 91 possible edges (14.29%). For the < 8 years of education group (education= 6.65±1.74, ASA= 3.08±0.35), the network included 3 clock face (CF), 7 digit, and 3 hour hand (HH) and minute hand (MH) independent, or “parent,” features connected to the retained edges (BIC= -7395.24). In contrast, the > 9 years of education group (education= 14.17±2.88, ASA= 2.90±0.46) network retained 1 CF, 6 digit, 5 HH and MH, and 1 additional parent features representing the total number of pen strokes (BIC= -6689.92). Both groups showed that greater distance from the HH to the center of the clock also had greater distance from the MH to the center of the clock [ßz(< 8 years)= 0.73, ßz(> 9 years)= 0.76]. Groups were similar in the size of the digit height relative to the distance of the digits to the CF [ßz(< 8 years)= 0.27, ßz(> 9 years)= 0.56]. Larger HH angle was associated with larger MH angle across groups [ßz(< 8 years)= 0.28, ßz(> 9 years)= 0.23].

Education groups differed in the ratio of dCDT parent feature types. Specifically, copy clock production in older adults with < 8 years of education relied more heavily on CF parent features. In contrast, older adults with > 9 years of education relied more heavily on HH and MH parent features. Individuals with < 8 years of education may more infrequently present the concept of time in the clock drawing command condition. This study highlights the importance of considering education level in interpreting dCDT scores and features.

Recent research has found that machine learning based analysis of patient speech can be used to classify Alzheimer’s Disease. We know of no studies, however, which systematically explore the value of pausing events in speech for detecting cognitive limitations. Using retrospectively acquired voice data from paragraph memory tests, we created two types of pause features: a) the number and duration of pauses, and b) frequency components in speech immediately following pausing. Multiple machine learning models were used to assess how these features could effectively discriminate individuals classified into two groups: Cognitively Compromised versus Cognitively Well.

Participants (age> 65 years, n= 67) completed the Newcomer paragraph memory test and a neuropsychological protocol as part of a federally funded prospective IRB approved investigation at the University of Florida. Participant vocal recordings were acquired for the immediate and delay conditions of the test. Speaker diarization was performed on the immediate free recall test condition to separate voices of patients from examiners. Features extracted from both test conditions included a) 3 pause characteristics (total number of pauses, total pause duration, and length of the longest pause), and b) 20 Mel Frequency Cepstral Coefficients (MFCC) pertaining to speech immediately (2.7 seconds) following pauses. These were combined with demographics (age, sex, race, education, and handedness) to create a total of 105 features that were used as inputs for multiple machine learning analytic models (random forest, logistic regression, naive Bayes, AdaBoost, Gradient Boost, and multi-layered perceptron). External neuropsychological metrics were used to initially classify Cognitively Compromised (i.e., < -1.0 standard deviation on > two of five test metrics: total immediate, delay, discrimination Hopkins Verbal Learning Test-Revised (HVLT-R),

Controlled Oral Word Association (COWA) test, category fluency ('animals')). Pearson Product Moment Correlations were used to assess the linear relationships between pauses and speech frequency categories and neuropsychological metrics.

Neuropsychology metric classification using -1SD cut-off identified 27% (18/67 participants) as Cognitively Compromised. The Cognitively Compromised group and the Cognitively Well group did not show any difference in distributions of individual pause/frequency features (Mann Whitney U-test, p> 0.11). A negative correlation was found between total duration of short pauses and HVLT total immediate free recall, while a positive correlation was found between MFCC-10 and HVLT total immediate free recall. The best classification model was AdaBoost Classifier which predicted the Cognitively Compromised label with 0.91 area under receiver operating curve, 0.81 accuracy, 0.43 sensitivity, 1.0 specificity, 1.0 precision, 0.6 f1 score.

Pause characteristics and frequency profiles of speech immediately following pauses from a paragraph memory test accurately identified older adults with compromised cognition, as measured by verbal learning and verbal fluency metrics. Furthermore, individuals with reduced HVLT immediate free recall generated more pauses, while individuals who recalled more words had higher power in mid-frequency bands (10th MFCC). Future research needs to replicate how paragraph recall pause characteristics and frequency the profile of speech immediately following pauses potentially provides a low resource alternative to automatic speech recognition models for detecting cognitive impairments.

Cognitive tests requiring spoken responses, such as paragraph recall, are rich in cognitive-related information that is not captured using traditional scoring methods. This study aimed to determine if linguistic features embedded in spoken responses may differentiate between individuals who are and are not cognitively impaired.

Participants in the Long Life Family Study completed a neuropsychological assessment which included the WMS-R Logical Memory I paragraph recall. For a subset of participants (N=709), test responses were digitally recorded and manually transcribed. We used Linguistic Inquiry Word Count, a text analysis program, to quantify word counts, grammatical features (e.g, prepositions, verb tenses), and the use of content words related to specific semantic categories (e.g., work-related, numbers) for immediate (IR) and delayed recall (DR). We used regression models with Generalized Estimating Equations adjusted by age, sex, education, and within-family correlation to select features associated with cognitive status (normal cognition [NC] versus cognitive impairment [CI]; Bonferroni-corrected threshold p<0.001). Next, we developed a “polyfeature score” (PFS) for both immediate and delayed recall, each calculated as a weighted sum of the selected linguistic features. We then built a logistic regression model to evaluate the predictive value of each PFS for identifying cognitively impaired individuals. In secondary analyses, we used regression models as above to identify features associated with mild cognitive impairment subtype (amnestic [aMCI] versus nonamnestic [naMCI]; threshold p< .05).

The sample included 599 participants with NC and 110 with CI (mean age = 72.3 ± 11.0 years, 54% female). The regression identified 8 linguistic features for IR and 7 for DR that significantly predicted cognitive status. Decreased use of content words related to work (e.g., employed, school, police) and biological processes (e.g., cook, cafeteria, eat) and the use of negations (e.g., no, not, can’t) were predictive of cognitive impairment in both recall conditions. In contrast, the use of other content word categories were predictive of cognitive status in only one recall condition (IR: leisure, cognitive processes, space; DR: drives, number). The use of fewer prepositions in IR, more first-person pronouns in DR, and fewer words in the past tense in DR were each associated with cognitive impairment. Word count was not predictive of cognitive status. Both PFSs were highly associated with cognitive status (PFS_IR ß= 0.74, p< 0.001; PFS_DR ß= 0.86, p= 0.001) with high discriminative value (PFS_IR AUC= 0.93, sensitivity = 0.81, specificity= 0.91; PFS_DR AUC= 0.95, sensitivity= 0.77, specificity= 0.88). In the CI subset, linguistic features differed between those classified as aMCI (n= 24) and naMCI (n= 40). Two function word categories predicted aMCI in IR whereas decreased word count, two function word categories, and two content word categories predicted aMCI in DR (all p< .05)

Linguistic features from paragraph recall provide high predictive value for classifying cognitive status increasing its potential as a cognitive screener in clinical settings. Additionally, each recall condition identified unique linguistic features associated with cognitive impairment which may aid differentiation of cognitive impairment subtypes and elucidate processes underlying deficits in learning and recall.

Research shows that highly educated individuals have at least 20 graphomotor features associated with clock drawing with hands set for '10 after 11' (Davoudi et al., 2021). Research has yet to understand clock drawing features in individuals with fewer years of education. In the current study, we compared older adults with < 8 years of education to those with > 9 years of education on number and pattern of graphomotor feature relationships in the clock drawing copy condition.

Participants age 65+ from the University of Florida (UF) and UF Health (N= 10,491) completed command and copy digital Clock Drawing Tests (dCDT) as a part of a federally-funded investigation. Participants were categorized into two groups: < 8 years of education (n= 304) and > 9 years of education (n= 10,187). Propensity score matching was used to match participants from each subgroup (n= 266 for each subgroup) on the following: age, sex, race, and ethnicity (n= 532, age= 74.99±6.21, education= 10.41±4.45, female= 42.7%, non-white= 32.0%). Network models were derived using Bayesian Structure Learning (BSL) with the hill-climbing algorithm to obtain optimal directed acyclic graphs (DAGs) from all possible solutions in each subgroup for the dCDT copy condition.

The < 8 years of education group (education= 6.65±1.74, ASA= 3.08±0.35), retained 12 of 91 possible edges (13.19%, BIC= -7775.50). The network retained 2 clock face (CF), 5 digit, and 5 hour hand (HH) and minute hand (MH) independent, or “parent,” features connected to the retained edges. In contrast, the > 9 years of education group (education= 14.17±2.88, ASA= 2.90±0.46) network retained 15 of 91 possible edges (16.48%, BIC= -8261.484). The network retained 2 CF, 6 digit, 4 HH and MH, and an additional 3 total stroke parent features. Both groups showed that greater distance from the HH to the clock center also had greater distance from the MH to the clock center (ßz= 0.73, both). Groups were similar in digit width size relative to digit height [ßz(< 8 years)= 0.72, ßz(> 9 years)= 0.74]. Digit height size related to CF area [ßz(< 8 years)= 0.44, ßz(> 9 years)= 0.62] and CF area related to the digit distance to the CF across groups [ßz(< 8 years)= 0.39, ßz(> 9 years)= 0.46]. Greater distance from the MH to the clock center was associated with smaller MH angle [ßz(< 8 years)= -0.35, ßz(> 9 years)= -0.31], whereas greater digit misplacement was associated with larger MH angle across groups [ßz(< 8 years)= 0.14, ßz(> 9 years)= 0.29].

Education groups differed in the ratio of dCDT parent feature types. Specifically, copy clock production in older adults with < 8 years of education relied more evenly across CF, digit, and MH and HH parent features. In contrast, those with > 9 years of education differed in the additional reliance on total stroke parent features. Individuals with < 8 years of education may more heavily rely upon visual referencing when copying a clock. This study highlights the importance of considering education level in interpreting dCDT scores and features.

To determine whether the DCTclock can detect differences across groups of patients seen in the memory clinic for suspected dementia.

Patients (n = 123) were classified into the following groups: cognitively normal (CN), subtle cognitive impairment (SbCI), amnestic cognitive impairment (aMCI), and mixed/dysexecutive cognitive impairment (mx/dysMCI). Nine outcome variables included a combined command/copy total score and four command and four copy indices measuring drawing efficiency, simple/complex motor operations, information processing speed, and spatial reasoning.

Total combined command/copy score distinguished between groups in all comparisons with medium to large effects. The mx/dysMCI group had the lowest total combined command/copy scores out of all groups. The mx/dysMCI group scored lower than the CN group on all command indices (p < .050, all analyses); and lower than the SbCI group on drawing efficiency (p = .011). The aMCI group scored lower than the CN group on spatial reasoning (p = .019). Smaller effect sizes were obtained for the four copy indices.

These results suggest that DCTclock command/copy parameters can dissociate CN, SbCI, and MCI subtypes. The larger effect sizes for command clock indices suggest these metrics are sensitive in detecting early cognitive decline. Additional research with a larger sample is warranted.

To determine how well machine learning algorithms can classify mild cognitive impairment (MCI) subtypes and Alzheimer’s disease (AD) using features obtained from the digital Clock Drawing Test (dCDT).

dCDT protocols were administered to 163 patients diagnosed with AD(n = 59), amnestic MCI (aMCI; n = 26), combined mixed/dysexecutive MCI (mixed/dys MCI; n = 43), and patients without MCI (non-MCI; n = 35) using standard clock drawing command and copy procedures, that is, draw the face of the clock, put in all of the numbers, and set the hands for “10 after 11.” A digital pen and custom software recorded patient’s drawings. Three hundred and fifty features were evaluated for maximum information/minimum redundancy. The best subset of features was used to train classification models to determine diagnostic accuracy.

Neural network employing information theoretic feature selection approaches achieved the best 2-group classification results with 10-fold cross validation accuracies at or above 83%, that is, AD versus non-MCI = 91.42%; AD versus aMCI = 91.49%; AD versus mixed/dys MCI = 84.05%; aMCI versus mixed/dys MCI = 84.11%; aMCI versus non-MCI = 83.44%; and mixed/dys MCI versus non-MCI = 85.42%. A follow-up two-group non-MCI versus all MCI patients analysis yielded comparable results (83.69%). Two-group classification analyses were achieved with 25–125 dCDT features depending on group classification. Three- and four-group analyses yielded lower but still promising levels of classification accuracy.

Early identification of emergent neurodegenerative illness is criterial for better disease management. Applying machine learning to standard neuropsychological tests promises to be an effective first line screening method for classification of non-MCI and MCI subtypes.

Previous research in mild cognitive impairment (MCI) suggests that visual episodic memory impairment may emerge before analogous verbal episodic memory impairment. The current study examined working memory (WM) test performance in MCI to assess whether patients present with greater visual versus verbal WM impairment. WM performance was also assessed in relation to hippocampal occupancy (HO), a ratio of hippocampal volume to ventricular dilation adjusted for demographic variables and intracranial volume.

Jak et al. (2009) (The American Journal of Geriatric Psychiatry, 17, 368–375) and Edmonds, Delano-Wood, Galasko, Salmon, & Bondi (2015) (Journal of Alzheimer’s Disease, 47(1), 231–242) criteria classify patients into four groups: little to no cognitive impairment (non-MCI); subtle cognitive impairment (SCI); amnestic MCI (aMCI); and a combined mixed/dysexecutive MCI (mixed/dys MCI). WM was assessed using co-normed Wechsler Adult Intelligence Scale-IV (WAIS-IV) Digit Span Backwards and Wechsler Memory Scale-IV (WMS-IV) Symbol Span Z-scores.

Between-group analyses found worse WMS-IV Symbol Span and WAIS-IV Digit Span Backwards performance for mixed/dys MCI compared to non-MCI patients. Within-group analyses found no differences for non-MCI patients; however, all other groups scored lower on WMS-IV Symbol Span than WAIS-IV Digit Span Backwards. Regression analysis with HO as the dependent variable was statistically significant for WMS-IV Symbol Span performance. WAIS-IV Digit Span Backwards performance failed to reach statistical significance.

Worse WMS-IV Symbol Span performance was observed in patient groups with measurable neuropsychological impairment and better WMS-IV Symbol Span performance was associated with higher HO ratios. These results suggest that visual WM may be particularly sensitive to emergent illness compared to analogous verbal WM tests.