Reidentification of Participants in Shared Clinical Data Sets: Experimental Study.

BACKGROUND: Large curated data sets are required to leverage speech-based tools in health care. These are costly to produce, resulting in increased interest in data sharing. As speech can potentially identify speakers (ie, voiceprints), sharing recordings raises privacy concerns. This is especially relevant when working with patient data protected under the Health Insurance Portability and Accountability Act. OBJECTIVE: We aimed to determine the reidentification risk for speech recordings, without reference to demographics or metadata, in clinical data sets considering both the size of the search space (ie, the number of comparisons that must be considered when reidentifying) and the nature of the speech recording (ie, the type of speech task). METHODS: Using a state-of-the-art speaker identification model, we modeled an adversarial attack scenario in which an adversary uses a large data set of identified speech (hereafter, the known set) to reidentify as many unknown speakers in a shared data set (hereafter, the unknown set) as possible. We first considered the effect of search space size by attempting reidentification with various sizes of known and unknown sets using VoxCeleb, a data set with recordings of natural, connected speech from >7000 healthy speakers. We then repeated these tests with different types of recordings in each set to examine whether the nature of a speech recording influences reidentification risk. For these tests, we used our clinical data set composed of recordings of elicited speech tasks from 941 speakers. RESULTS: We found that the risk was inversely related to the number of comparisons an adversary must consider (ie, the search space), with a positive linear correlation between the number of false acceptances (FAs) and the number of comparisons (r=0.69; P<.001). The true acceptances (TAs) stayed relatively stable, and the ratio between FAs and TAs rose from 0.02 at 1 × 105 comparisons to 1.41 at 6 × 106 comparisons, with a near 1:1 ratio at the midpoint of 3 × 106 comparisons. In effect, risk was high for a small search space but dropped as the search space grew. We also found that the nature of a speech recording influenced reidentification risk, with nonconnected speech (eg, vowel prolongation: FA/TA=98.5; alternating motion rate: FA/TA=8) being harder to identify than connected speech (eg, sentence repetition: FA/TA=0.54) in cross-task conditions. The inverse was mostly true in within-task conditions, with the FA/TA ratio for vowel prolongation and alternating motion rate dropping to 0.39 and 1.17, respectively. CONCLUSIONS: Our findings suggest that speaker identification models can be used to reidentify participants in specific circumstances, but in practice, the reidentification risk appears small. The variation in risk due to search space size and type of speech task provides actionable recommendations to further increase participant privacy and considerations for policy regarding public release of speech recordings.

Stricker Learning Span criterion validity: a remote self-administered multi-device compatible digital word list memory measure shows similar ability to differentiate amyloid and tau PET-defined biomarker groups as in-person Auditory Verbal Learning Test.

OBJECTIVE: The Stricker Learning Span (SLS) is a computer-adaptive digital word list memory test specifically designed for remote assessment and self-administration on a web-based multi-device platform (Mayo Test Drive). We aimed to establish criterion validity of the SLS by comparing its ability to differentiate biomarker-defined groups to the person-administered Rey's Auditory Verbal Learning Test (AVLT). METHOD: Participants (N = 353; mean age = 71, SD = 11; 93% cognitively unimpaired [CU]) completed the AVLT during an in-person visit, the SLS remotely (within 3 months) and had brain amyloid and tau PET scans available (within 3 years). Overlapping groups were formed for 1) those on the Alzheimer's disease (AD) continuum (amyloid PET positive, A+, n = 125) or not (A-, n = 228), and those with biological AD (amyloid and tau PET positive, A+T+, n = 55) vs no evidence of AD pathology (A-T-, n = 195). Analyses were repeated among CU participants only. RESULTS: The SLS and AVLT showed similar ability to differentiate biomarker-defined groups when comparing AUROCs (p's > .05). In logistic regression models, SLS contributed significantly to predicting biomarker group beyond age, education, and sex, including when limited to CU participants. Medium (A- vs A+) to large (A-T- vs A+T+) unadjusted effect sizes were observed for both SLS and AVLT. Learning and delay variables were similar in terms of ability to separate biomarker groups. CONCLUSIONS: Remotely administered SLS performed similarly to in-person-administered AVLT in its ability to separate biomarker-defined groups, providing evidence of criterion validity. Results suggest the SLS may be sensitive to detecting subtle objective cognitive decline in preclinical AD.

Realtime Diagnosis from Electrocardiogram Artificial Intelligence-Guided Screening for Atrial Fibrillation with Long Follow-Up (REGAL): Rationale and design of a pragmatic, decentralized, randomized controlled trial.

BACKGROUND: Atrial fibrillation (AF) is associated with increased risks of stroke and dementia. Early diagnosis and treatment could reduce the disease burden, but AF is often undiagnosed. An artificial intelligence (AI) algorithm has been shown to identify patients with previously unrecognized AF; however, monitoring these high-risk patients has been challenging. Consumer wearable devices could be an alternative to enable long-term follow-up. OBJECTIVES: To test whether Apple Watch, used as a long-term monitoring device, can enable early diagnosis of AF in patients who were identified as having high risk based on AI-ECG. DESIGN: The Realtime diagnosis from Electrocardiogram (ECG) Artificial Intelligence (AI)-Guided Screening for Atrial Fibrillation (AF) with Long Follow-up (REGAL) study is a pragmatic trial that will accrue up to 2,000 older adults with a high likelihood of unrecognized AF determined by AI-ECG to reach our target of 1,420 completed participants. Participants will be 1:1 randomized to intervention or control and will be followed up for 2 years. Patients in the intervention arm will receive or use their existing Apple Watch and iPhone and record a 30-second ECG using the watch routinely or if an abnormal heart rate notification is prompted. The primary outcome is newly diagnosed AF. Secondary outcomes include changes in cognitive function, stroke, major bleeding, and all-cause mortality. The trial will utilize a pragmatic, digitally-enabled, decentralized design to allow patients to consent and receive follow-up remotely without traveling to the study sites. SUMMARY: The REGAL trial will examine whether a consumer wearable device can serve as a long-term monitoring approach in older adults to detect AF and prevent cognitive function decline. If successful, the approach could have significant implications on how future clinical practice can leverage consumer devices for early diagnosis and disease prevention. CLINICALTRIALS: GOV: : NCT05923359.

Neural network process simulations support a distributed memory system and aid design of a novel computer adaptive digital memory test for preclinical and prodromal Alzheimer's disease.

OBJECTIVE: Growing evidence supports the importance of learning as a central deficit in preclinical/prodromal Alzheimer's disease. The aims of this study were to conduct a series of neural network simulations to develop a functional understanding of a distributed, nonmodular memory system that can learn efficiently without interference. This understanding is applied to the development of a novel digital memory test. METHOD: Simulations using traditional feed forward neural network architectures to learn simple logic problems are presented. The simulations demonstrate three limitations: (a) inefficiency, (b) an inability to learn problems consistently, and (c) catastrophic interference when given multiple problems. A new mirrored cascaded architecture is introduced to address these limitations, with support provided by a series of simulations. RESULTS: The mirrored cascaded architecture demonstrates efficient and consistent learning relative to feed forward networks but also suffers from catastrophic interference. Addition of context values to add the capability of distinguishing features as part of learning eliminates the problem of interference in the mirrored cascaded, but not the feed forward, architectures. CONCLUSIONS: A mirrored cascaded architecture addresses the limitations of traditional feed forward neural networks, provides support for a distributed memory system, and emphasizes the importance of context to avoid interference. These process models contributed to the design of a digital computer-adaptive word list learning test that places maximum stress on the capability to distinguish specific episodes of learning. Process simulations provide a useful method of testing models of brain function and contribute to new approaches to neuropsychological assessment. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

A novel computer adaptive word list memory test optimized for remote assessment: Psychometric properties and associations with neurodegenerative biomarkers in older women without dementia.

Introduction: This study established the psychometric properties and preliminary validity of the Stricker Learning Span (SLS), a novel computer adaptive word list memory test designed for remote assessment and optimized for smartphone use. Methods: Women enrolled in the Mayo Clinic Specialized Center of Research Excellence (SCORE) were recruited via e-mail or phone to complete two remote cognitive testing sessions. Convergent validity was assessed through correlation with previously administered in-person neuropsychological tests (n = 96, ages 55-79) and criterion validity through associations with magnetic resonance imaging measures of neurodegeneration sensitive to Alzheimer's disease (n = 47). Results: SLS performance significantly correlated with the Auditory Verbal Learning Test and measures of neurodegeneration (temporal meta-regions of interest and entorhinal cortical thickness, adjusting for age and education). Test-retest reliabilities across two sessions were 0.71-0.76 (two-way mixed intraclass correlation coefficients). Discussion: The SLS is a valid and reliable self-administered memory test that shows promise for remote assessment of aging and neurodegenerative disorders.

Comparison of the traditional recall-based versus a new list-based method for computing semantic clustering on the California Verbal Learning Test: evidence from Alzheimer's disease.

For over 50 years, cognitive psychologists and neuropsychologists have relied almost exclusively on a method for computing semantic clustering on list-learning tasks (recall-based formula) that was derived from an outdated assumption about how learning occurs. A new procedure for computing semantic clustering (list-based formula) was developed for the CVLT-II to correct the shortcomings of the traditional method. In the present study we compared the clinical utility of the traditional recall-based method versus the new list-based method using results from the original CVLT administered to 87 patients with Alzheimer's disease and 86 matched normal control participants. Logistic regression and score distribution analyses indicated that the new list-based method enhances the detection of differences in semantic-clustering ability between the groups.

Disability and self-management practices of people with bipolar disorder: a web-based survey.

In a web-based survey asking adults diagnosed with bipolar disorder about illness management, we obtain frequency of self-reported usage and perceived helpfulness of 27 self-management strategies. We correlated the strategy use and perceived helpfulness with demographic and clinical characteristics, along with the Illness Intrusiveness Scale total score. Completed surveys were obtained from 1,024 individuals. Perceived helpfulness of 18 of 27 strategies was correlated negatively with illness intrusiveness at the P < 0.001 level. Given limitations of web-based surveys, our study underscores the substantial negative impact of bipolar disorder, along with the potential of the Internet to enhance the use of self-management strategies.

Short-term course of neuropsychological abilities in middle-aged and older adults with bipolar disorder.

OBJECTIVES: There are few longitudinal studies of neurocognition in bipolar disorder, and the short-term course of cognitive deficits in later-life bipolar disorder is unknown. METHODS: We administered a battery of neurocognitive tests, repeated 1-3 years after baseline, to 35 community-dwelling outpatients with bipolar disorder (mean age = 58), and compared their performance on a composite measure of cognitive functioning to that of demographically matched samples of normal comparison subjects (NCs; n = 35) and patients with schizophrenia (n = 35). Using regression analyses, we examined group differences in baseline performance, trajectory of change over time, and variability in performance across time. Within the bipolar group, we examined the impact of baseline severity and change in severity of psychiatric symptoms on intra-individual change in neurocognitive performance. RESULTS: At baseline, the group with bipolar disorder differed in overall neurocognitive functioning from the NCs, but did not differ significantly from the schizophrenia group. The bipolar group did not differ from the NCs or schizophrenia group in the mean trajectory of change between time-points, but the bipolar patients showed more intra-individual variability over time than the NCs or schizophrenia group. In the bipolar group, change in neurocognitive function was not related to baseline or change in psychiatric symptom severity. CONCLUSIONS: Middle-aged and older community-dwelling adults with bipolar disorder have greater short-term variability in level of neurocognitive functioning relative to NCs or people with schizophrenia. The developmental course of and risk factors for cognitive deficits in bipolar disorder should be examined in future longitudinal studies.

Flanker compatibility effects in patients with Parkinson's disease: impact of target onset delay and trial-by-trial stimulus variation.

Parkinson's disease (PD) patients and healthy controls were administered a flanker task that consisted of the presentation of colored targets and distractors. Participants were required to attend to the center target and identify its color. The stimulus displays were either congruent (i.e., the target and flankers were the same color) or incongruent. The time between the onset of the flanker and the target color (the target onset delay) was either short or long. Results indicated that PD patients and controls did not differ in the magnitude of the flanker effect within individual trials in that both groups demonstrated a typical flanker effect at the short target onset delay and neither group demonstrated a flanker effect at the longer delay. However, when performance was examined on a trial-by-trial basis, PD patients demonstrated a slowing of reaction time relative to controls when having to make the same response across consecutive trials at longer inter-trial intervals when the flankers were incongruent across consecutive trials and the display on the second of two trials was incongruent. These results indicate that PD patients are impaired in inhibiting the distractors over an extended delay and that this deficit may impact motor responding in these patients, suggesting that the basal ganglia contribute to the interface of attention and action.

The impact of sleep deprivation and task difficulty on networks of fMRI brain response.

Previous fMRI research has found altered brain response after total sleep deprivation (TSD), with TSD effects moderated by task difficulty. Specific models of the impact of sleep deprivation and task difficulty on brain response have yet to be developed. Differences in networks of fMRI measured brain response during verbal encoding in sleep deprived and well-rested individuals were examined with structural equation modeling (SEM). During fMRI scanning, 23 healthy volunteers memorized words either easy or difficult to recall, 12 (well-rested) and 36 hours (sleep deprived) after awaking. A priori models that linked specified regions of interest were evaluated, with the focus on the extent to which two left parietal regions interacted with the left inferior frontal gyrus (Model 1) or with the right inferior frontal gyrus (Model 2). Task difficulty, not TSD, determined which model fit the brain response data; Model 2 fit best for hard words before and after TSD, whereas Model 1 fit best for easy words. TSD altered the patterns of interaction within each of the best fitting models: prefrontal interactions with the left inferior parietal lobe were diminished and intra-parietal interactions increased. Sleep deprivation and item difficulty produce different effects on brain networks involved in verbal learning.

Semantic and cross-case identity priming in patients with Parkinson's disease.

Semantic and cross-case identity priming were investigated in nondemented patients with Parkinson's disease (PD) and controls using the Lexical Decision Task. Three conditions were administered that consisted of the presentation of prime and target word pairs. In the semantic priming condition the word pairs were semantically related (e.g., table-CHAIR), in the cross-case identity priming condition the word pairs consisted of the same word (e.g., noise-NOISE), and in the unrelated condition the word pairs were not related semantically (e.g., guns-DEEP). A fourth condition was also administered that consisted of the presentation of a prime word and a pronounceable nonword target (e.g., starved-FORVE). Participants were asked to indicate whether the target was a real word or a nonword. The prime and target were separated by either a short or long (250 ms or 1000 ms) stimulus onset asynchrony (SOA). Results indicated that PD patients displayed normal semantic priming (i.e., faster responding to the target in the semantic condition as compared to the unrelated condition) at both the short and long SOA. Similarly, PD patients displayed normal cross-case identity priming (i.e., faster responding to the target in the identity condition relative to the unrelated condition) at the long SOA. At the short SOA, however, PD patients displayed hyper identity priming relative to controls (134 ms vs. 50 ms). These results suggest that semantic processes are normal in nondemented PD patients but that the processes involved in accessing lexical information may be overly activated in these patients.

New semantic and serial clustering indices for the California Verbal Learning Test-Second Edition: background, rationale, and formulae.

The original California Verbal Learning Test (CVLT) employed a semantic clustering index that used the words recalled during a given trial as the baseline for calculating expected values of chance clustering (recall-based expectancy). Although commonly used in cognitive psychology, clustering indices that use recall-based calculations of expectancy are implied by the assumption that organizational processes do not occur until after words are retrieved from memory. This assumption contradicts the generally held assumptions among neuropsychologists that (1) organization is an antecedent to recall, and (2) increases in the use of organizational strategies will result in better recall performance. After reviewing a brief history of clustering metrics, we used Monte Carlo simulations, informative examples, and patient data to examine clustering indices that use the word list as a baseline for calculating expectancy and propose these list-based expectancy measures as a refinement of the clustering indices used on the original CVLT. These indices are used on the recently published CVLT-II.

Visual perception without awareness in a patient with posterior cortical atrophy: impaired explicit but not implicit processing of global information.

A patient with progressive posterior cortical atrophy (PCA) was examined on several tests of visual cognition. The patient displayed multiple visual cognitive deficits, which included problems identifying degraded stimuli, attending to two or more stimuli simultaneously, recognizing faces, tracing simple visual stimuli, matching simple shapes, and copying objects. The patient was also impaired in identifying visual targets contained at the global level within global-local stimuli (i.e., smaller letters that compose a larger letter). Although the patient denied any conscious awareness of the global form, he nevertheless displayed a normal pattern of global interference when asked to identify local level targets. Thus, the patient processed the global information despite not being consciously aware of such information. These results suggest that global-local processing can take place in the absence of awareness. Possible neurocognitive mechanisms explaining this dissociation are discussed.