Background Story recall is a simple and sensitive cognitive test that is commonly used to measure changes in episodic memory function in early Alzheimer disease (AD). Recent advances in digital technology and natural language processing methods make this test a candidate for automated administration and scoring. Multiple parallel test stimuli are required for higher-frequency disease monitoring. Objective This study aims to develop and validate a remote and fully automated story recall task, suitable for longitudinal assessment, in a population of older adults with and without mild cognitive impairment (MCI) or mild AD. Methods The “Amyloid Prediction in Early Stage Alzheimer’s disease” (AMYPRED) studies recruited participants in the United Kingdom (AMYPRED-UK: NCT04828122) and the United States (AMYPRED-US: NCT04928976). Participants were asked to complete optional daily self-administered assessments remotely on their smart devices over 7 to 8 days. Assessments included immediate and delayed recall of 3 stories from the Automatic Story Recall Task (ASRT), a test with multiple parallel stimuli (18 short stories and 18 long stories) balanced for key linguistic and discourse metrics. Verbal responses were recorded and securely transferred from participants’ personal devices and automatically transcribed and scored using text similarity metrics between the source text and retelling to derive a generalized match score. Group differences in adherence and task performance were examined using logistic and linear mixed models, respectively. Correlational analysis examined parallel-forms reliability of ASRTs and convergent validity with cognitive tests (Logical Memory Test and Preclinical Alzheimer’s Cognitive Composite with semantic processing). Acceptability and usability data were obtained using a remotely administered questionnaire. Results Of the 200 participants recruited in the AMYPRED studies, 151 (75.5%)—78 cognitively unimpaired (CU) and 73 MCI or mild AD—engaged in optional remote assessments. Adherence to daily assessment was moderate and did not decline over time but was higher in CU participants (ASRTs were completed each day by 73/106, 68.9% participants with MCI or mild AD and 78/94, 83% CU participants). Participants reported favorable task usability: infrequent technical problems, easy use of the app, and a broad interest in the tasks. Task performance improved modestly across the week and was better for immediate recall. The generalized match scores were lower in participants with MCI or mild AD (Cohen d=1.54). Parallel-forms reliability of ASRT stories was moderate to strong for immediate recall (mean rho 0.73, range 0.56-0.88) and delayed recall (mean rho=0.73, range=0.54-0.86). The ASRTs showed moderate convergent validity with established cognitive tests. Conclusions The unsupervised, self-administered ASRT task is sensitive to cognitive impairments in MCI and mild AD. The task showed good usability, high parallel-forms reliability, and high convergent validity with established cognitive tests. Remote, low-cost, low-burden, and automatically scored speech assessments could support diagnostic screening, health care, and treatment monitoring.
Introduction Artificial intelligence (AI) systems leveraging speech and language changes could support timely detection of Alzheimer's disease (AD). Methods The AMYPRED study (NCT04828122) recruited 133 subjects with an established amyloid beta (Aβ) biomarker (66 Aβ+, 67 Aβ–) and clinical status (71 cognitively unimpaired [CU], 62 mild cognitive impairment [MCI] or mild AD). Daily story recall tasks were administered via smartphones and analyzed with an AI system to predict MCI/mild AD and Aβ positivity. Results Eighty‐six percent of participants (115/133) completed remote assessments. The AI system predicted MCI/mild AD (area under the curve [AUC] = 0.85, ±0.07) but not Aβ (AUC = 0.62 ±0.11) in the full sample, and predicted Aβ in clinical subsamples (MCI/mild AD: AUC = 0.78 ±0.14; CU: AUC = 0.74 ±0.13) on short story variants (immediate recall). Long stories and delayed retellings delivered broadly similar results. Discussion Speech‐based testing offers simple and accessible screening for early‐stage AD.
INTRODUCTION: Longitudinal data is key to identifying cognitive decline and treatment response in Alzheimer's disease (AD). METHODS: The Automatic Story Recall Task (ASRT) is a novel, fully automated test that can be self-administered remotely. In this longitudinal case-control observational study, 151 participants (mean age: 69.99 (range 54-82), 73 mild cognitive impairment/mild AD and 78 cognitively unimpaired) completed parallel ASRT assessments on their smart devices over 7-8 days. Responses were automatically transcribed and scored using text similarity metrics. RESULTS: Participants reported good task usability. Adherence to optional daily assessment was moderate. Parallel forms correlation coefficients between ASRTs were moderate-high. ASRTs correlated moderately with established tests of episodic memory and global cognitive function. Poorer performance was observed in participants with MCI/Mild AD. DISCUSSION: Unsupervised ASRT assessment is feasible in older and cognitively impaired people. This automated task shows good parallel forms reliability and convergent validity with established cognitive tests.
BackgroundChanges in speech, language, and episodic and semantic memory are documented in Alzheimer’s disease (AD) years before routine diagnosis.AimsDevelop an Artificial Intelligence (AI) system detecting amyloid-confirmed prodromal and preclinical AD from speech collected remotely via participants’ smartphones.MethodA convenience sample of 133 participants with established amyloid beta and clinical diagnostic status (66 Aβ+, 67 Aβ-; 71 cognitively unimpaired (CU), 62 with mild cognitive impairment (MCI) or mild AD) completed clinical assessments for the AMYPRED study (NCT04828122). Participants completed optional remote assessments daily for 7-8 days, including the Automatic Story Recall Task (ASRT), a story recall paradigm with short and long variants, and immediate and delayed recall phases. Vector-based representations from each story source and transcribed retelling were produced using ParaBLEU, a paraphrase evaluation model. Representations were fed into logistic regression models trained with tournament leave-pair-out cross-validation analysis, predicting Aβ status and MCI/mild AD within the full sample and Aβ status in clinical diagnostic subsamples.FindingsAt least one full remote ASRT assessment was completed by 115 participants (mean age=69.6 (range 54-80); 63 female/52 male; 66 CU and 49 MCI/mild AD, 56 Aβ+ and 59 Aβ-). Using an average of 2.7 minutes of automatically transcribed speech from immediate recall of short stories, the AI system predicted MCI/mild AD in the full sample (AUC=0.85 +/- 0.08), and amyloid in MCI/mild AD (AUC=0.73 +/- 0.14) and CU subsamples (AUC=0.71 +/- 0.13). Amyloid classification within the full sample was no better than chance (AUC=0.57 +/- 0.11). Broadly similar results were reported for manually transcribed data, long ASRTs and delayed recall.InterpretationCombined with advanced AI language models, brief, remote speech-based testing offers simple, accessible and cost-effective screening for early stage AD.FundingNovoic.Research in contextEvidence before this studyRecent systematic reviews have examined the use of speech data to detect vocal and linguistic changes taking place in Alzheimer’s dementia. Most of this research has been completed in the DementiaBank cohort, where subjects are usually in the (more progressed) dementia stages and without biomarker confirmation of Alzheimer’s disease (AD). Whether speech assessment can be used in a biomarker-confirmed, early stage (preclinical and prodromal) AD population has not yet been tested. Most prior work has relied on extracting manually defined “features”, e.g. the noun rate, which has too low a predictive value to offer clinical utility in an early stage AD population. In recent years, audio- and text-based machine learning models have improved significantly and a few studies have used such models in the context of classifying AD dementia. These approaches could offer greater sensitivity but it remains to be seen how well they work in a biomarker-confirmed, early stage AD population. Most studies have relied on controlled research settings and on manually transcribing speech before analysis, both of which limit broader applicability and use in clinical practice.Added value of this studyThis study tests the feasibility of advanced speech analysis for clinical testing of early stage AD. We present the results from a cross-sectional sample in the UK examining the predictive ability of fully automated speech-based testing in biomarker-confirmed early stage Alzheimer’s disease. We use a novel artificial intelligence (AI) system, which delivers sensitive indicators of AD-at-risk or subtle cognitive impairment. The AI system differentiates amyloid beta positive and amyloid beta negative subjects, and subjects with mild cognitive impairment (MCI) or mild AD from cognitively healthy subjects. Importantly the system is fully remote and self-contained: participants’ own devices are used for test administration and speech capture. Transcription and analyses are automated, with limited signal loss. Overall the results support the real-world applicability of speech-based assessment to detect early stage Alzheimer’s disease. While a number of medical devices have recently been approved using image-based AI algorithms, the present research is the first to demonstrate the use case and promise of speech-based AI systems for clinical practice.Implications of all the available evidencePrior research has shown compelling evidence of speech- and language-based changes occurring in more progressed stages of Alzheimer’s disease. Our study builds on this early work to show the clinical utility and feasibility of speech-based AI systems for the detection of Alzheimer’s disease in its earliest stages. Our work, using advanced AI systems, shows sensitivity to a biomarker-confirmed early stage AD population. Speech data can be collected with self-administered assessments completed in a real world setting, and analysed automatically. With the first treatment for AD entering the market, there is an urgent need for scalable, affordable, convenient and accessible testing to screen at-risk subject candidates for biomarker assessment and early cognitive impairment. Sensitive speech-based biomarkers may help to fulfil this unmet need.
Early detection of Alzheimer’s disease is required to identify patients suitable for disease-modifying medications and to improve access to non-pharmacological preventative interventions. Prior research shows detectable changes in speech in Alzheimer’s dementia and its clinical precursors. The current study assesses whether a fully automated speech-based artificial intelligence system can detect cognitive impairment and amyloid beta positivity, which characterize early stages of Alzheimer’s disease. Two hundred participants (age 54–85, mean 70.6; 114 female, 86 male) from sister studies in the UK (NCT04828122) and the USA (NCT04928976), completed the same assessments and were combined in the current analyses. Participants were recruited from prior clinical trials where amyloid beta status (97 amyloid positive, 103 amyloid negative, as established via PET or CSF test) and clinical diagnostic status was known (94 cognitively unimpaired, 106 with mild cognitive impairment or mild Alzheimer’s disease). The automatic story recall task was administered during supervised in-person or telemedicine assessments, where participants were asked to recall stories immediately and after a brief delay. An artificial intelligence text-pair evaluation model produced vector-based outputs from the original story text and recorded and transcribed participant recalls, quantifying differences between them. Vector-based representations were fed into logistic regression models, trained with tournament leave-pair-out cross-validation analysis to predict amyloid beta status (primary endpoint), mild cognitive impairment and amyloid beta status in diagnostic subgroups (secondary endpoints). Predictions were assessed by the area under the receiver operating characteristic curve for the test result in comparison with reference standards (diagnostic and amyloid status). Simulation analysis evaluated two potential benefits of speech-based screening: (i) mild cognitive impairment screening in primary care compared with the Mini-Mental State Exam, and (ii) pre-screening prior to PET scanning when identifying an amyloid positive sample. Speech-based screening predicted amyloid beta positivity (area under the curve = 0.77) and mild cognitive impairment or mild Alzheimer’s disease (area under the curve = 0.83) in the full sample, and predicted amyloid beta in subsamples (mild cognitive impairment or mild Alzheimer’s disease: area under the curve = 0.82; cognitively unimpaired: area under the curve = 0.71). Simulation analyses indicated that in primary care, speech-based screening could modestly improve detection of mild cognitive impairment (+8.5%), while reducing false positives (−59.1%). Furthermore, speech-based amyloid pre-screening was estimated to reduce the number of PET scans required by 35.3% and 35.5% in individuals with mild cognitive impairment and cognitively unimpaired individuals, respectively. Speech-based assessment offers accessible and scalable screening for mild cognitive impairment and amyloid beta positivity.
BackgroundVocal and linguistic changes in Alzheimer’s dementia have been documented. The current study assesses whether a fully automated speech‐based artificial intelligence (AI) system can detect early clinical impairment and amyloid positivity, which characterise the earliest stages of Alzheimer’s disease (AD).MethodTwo studies were completed in the UK and USA: AMYPRED‐UK (NCT04828122) and AMYPRED‐US (NCT04928976). 200 participants with established amyloid beta (Aβ) and clinical diagnostic status (97 Aβ+, 103 Aβ‐ from prior PET scan or CSF tests; 94 cognitively unimpaired (CU), 106 with mild cognitive impairment (MCI) or mild AD) were recruited and completed automated assessments with the Automatic Story Recall Task (ASRT) either in‐clinic or via telemedicine appointment. The AI text‐pair evaluation model ParaBLEU produced vector‐based representations of the abstract, generalized patterns that differed between the original story text and transcribed retellings. These were fed into logistic regression models trained with tournament leave‐pair‐out cross‐validation analysis to predict Aβ status and MCI/mild AD. Potential benefits of screening using the ASRT system were examined via simulation, including: (1) identifying MCI in primary care, and (2) reducing the number of PET scans required in clinical research studies and trials.ResultUsing an average of only 6.56 minutes of speech per participant, the ASRT system predicted Aβ positivity in the full sample (area under the receiver operating characteristic curve (AUC) = 0.77) and in diagnostic subsamples (MCI/mild AD: AUC = 0.82; CU: AUC = 0.71), as well as MCI/mild AD (AUC = 0.83) in the full sample. Simulation indicated that screening with the ASRT system could: (1) increase correct (+8.5%) and reduce incorrect referrals (‐59.1%) compared with the Mini‐Mental State Exam; and (2) enrich samples for Aβ positivity prior to PET scan (‐35.3% and ‐35.5% fewer scans required in MCI and CU individuals, respectively).ConclusionWith the first disease‐modifying treatment for AD now available, there is an urgent need for improved screening to identify individuals at risk for AD dementia. The ASRT system is a brief, effective, automated, speech‐based cognitive assessment offering scalable screening for MCI/mild AD and amyloid beta biomarker positivity.
We propose a method for learning de-identified prosody representations from raw audio using a contrastive self-supervised signal. Whereas prior work has relied on conditioning models on bottlenecks, we introduce a set of inductive biases that exploit the natural structure of prosody to minimize timbral information and decouple prosody from speaker representations. Despite aggressive downsampling of the input and having no access to linguistic information, our model performs comparably to state-of-the-art speech representations on DAMMP, a new benchmark we introduce for spoken language understanding. We use minimum description length probing to show that our representations have selectively learned the subcomponents of non-timbral prosody, and that the product quantizer naturally disentangles them without using bottlenecks. We derive an information-theoretic definition of speech de-identifiability and use it to demonstrate that our prosody representations are less identifiable than other speech representations.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.