Automated detection of progressive speech changes in early Alzheimer's disease.

Speech and language changes occur in Alzheimer's disease (AD), but few studies have characterized their longitudinal course. We analyzed open-ended speech samples from a prodromal-to-mild AD cohort to develop a novel composite score to characterize progressive speech changes. Participant speech from the Clinical Dementia Rating (CDR) interview was analyzed to compute metrics reflecting speech and language characteristics. We determined the aspects of speech and language that exhibited significant longitudinal change over 18 months. Nine acoustic and linguistic measures were combined to create a novel composite score. The speech composite exhibited significant correlations with primary and secondary clinical endpoints and a similar effect size for detecting longitudinal change. Our results demonstrate the feasibility of using automated speech processing to characterize longitudinal change in early AD. Speech-based composite scores could be used to monitor change and detect response to treatment in future research. HIGHLIGHTS: Longitudinal speech samples were analyzed to characterize speech changes in early AD.Acoustic and linguistic measures showed significant change over 18 months.A novel speech composite score was computed to characterize longitudinal change.The speech composite correlated with primary and secondary trial endpoints.Automated speech analysis could facilitate remote, high frequency monitoring in AD.

Correlating natural language processing and automated speech analysis with clinician assessment to quantify speech-language changes in mild cognitive impairment and Alzheimer's dementia.

BACKGROUND: Language impairment is an important marker of neurodegenerative disorders. Despite this, there is no universal system of terminology used to describe these impairments and large inter-rater variability can exist between clinicians assessing language. The use of natural language processing (NLP) and automated speech analysis (ASA) is emerging as a novel and potentially more objective method to assess language in individuals with mild cognitive impairment (MCI) and Alzheimer's dementia (AD). No studies have analyzed how variables extracted through NLP and ASA might also be correlated to language impairments identified by a clinician. METHODS: Audio recordings (n=30) from participants with AD, MCI, and controls were rated by clinicians for word-finding difficulty, incoherence, perseveration, and errors in speech. Speech recordings were also transcribed, and linguistic and acoustic variables were extracted through NLP and ASA. Correlations between clinician-rated speech characteristics and the variables were compared using Spearman's correlation. Exploratory factor analysis was applied to find common factors between variables for each speech characteristic. RESULTS: Clinician agreement was high in three of the four speech characteristics: word-finding difficulty (ICC = 0.92, p<0.001), incoherence (ICC = 0.91, p<0.001), and perseveration (ICC = 0.88, p<0.001). Word-finding difficulty and incoherence were useful constructs at distinguishing MCI and AD from controls, while perseveration and speech errors were less relevant. Word-finding difficulty as a construct was explained by three factors, including number and duration of pauses, word duration, and syntactic complexity. Incoherence was explained by two factors, including increased average word duration, use of past tense, and changes in age of acquisition, and more negative valence. CONCLUSIONS: Variables extracted through automated acoustic and linguistic analysis of MCI and AD speech were significantly correlated with clinician ratings of speech and language characteristics. Our results suggest that correlating NLP and ASA with clinician observations is an objective and novel approach to measuring speech and language changes in neurodegenerative disorders.

Comparing Pre-trained and Feature-Based Models for Prediction of Alzheimer's Disease Based on Speech.

Introduction: Research related to the automatic detection of Alzheimer's disease (AD) is important, given the high prevalence of AD and the high cost of traditional diagnostic methods. Since AD significantly affects the content and acoustics of spontaneous speech, natural language processing, and machine learning provide promising techniques for reliably detecting AD. There has been a recent proliferation of classification models for AD, but these vary in the datasets used, model types and training and testing paradigms. In this study, we compare and contrast the performance of two common approaches for automatic AD detection from speech on the same, well-matched dataset, to determine the advantages of using domain knowledge vs. pre-trained transfer models. Methods: Audio recordings and corresponding manually-transcribed speech transcripts of a picture description task administered to 156 demographically matched older adults, 78 with Alzheimer's Disease (AD) and 78 cognitively intact (healthy) were classified using machine learning and natural language processing as "AD" or "non-AD." The audio was acoustically-enhanced, and post-processed to improve quality of the speech recording as well control for variation caused by recording conditions. Two approaches were used for classification of these speech samples: (1) using domain knowledge: extracting an extensive set of clinically relevant linguistic and acoustic features derived from speech and transcripts based on prior literature, and (2) using transfer-learning and leveraging large pre-trained machine learning models: using transcript-representations that are automatically derived from state-of-the-art pre-trained language models, by fine-tuning Bidirectional Encoder Representations from Transformer (BERT)-based sequence classification models. Results: We compared the utility of speech transcript representations obtained from recent natural language processing models (i.e., BERT) to more clinically-interpretable language feature-based methods. Both the feature-based approaches and fine-tuned BERT models significantly outperformed the baseline linguistic model using a small set of linguistic features, demonstrating the importance of extensive linguistic information for detecting cognitive impairments relating to AD. We observed that fine-tuned BERT models numerically outperformed feature-based approaches on the AD detection task, but the difference was not statistically significant. Our main contribution is the observation that when tested on the same, demographically balanced dataset and tested on independent, unseen data, both domain knowledge and pretrained linguistic models have good predictive performance for detecting AD based on speech. It is notable that linguistic information alone is capable of achieving comparable, and even numerically better, performance than models including both acoustic and linguistic features here. We also try to shed light on the inner workings of the more black-box natural language processing model by performing an interpretability analysis, and find that attention weights reveal interesting patterns such as higher attribution to more important information content units in the picture description task, as well as pauses and filler words. Conclusion: This approach supports the value of well-performing machine learning and linguistically-focussed processing techniques to detect AD from speech and highlights the need to compare model performance on carefully balanced datasets, using consistent same training parameters and independent test datasets in order to determine the best performing predictive model.