Relation between Depression Dimensions and Speech Acoustic and Emotion‐based Features

Alzheimer s & Dementia · 2025-12-01

BACKGROUND: Depression is a common and highly heterogeneous disorder in older adults, often linked to faster cognitive decline. While standard questionnaires are subjective, speech analysis may offer a more objective method for characterizing depression. This study investigates the relationship between speech features and depression dimensions in participants at Mount Sinai Alzheimer's Disease Research Center (ADRC). METHOD: Participants included healthy controls (n = 31) and individuals with mild cognitive impairment (MCI= 22) and Alzheimer's Disease (AD, n = 16). They described three pictures with neutral, negative and positive themes. Speech features were analyzed using automated pipelines and emotion-based variables and acoustic features were used for analysis. Depression dimensions (dysphoria, apathy, hopelessness, and memory complaints) were assessed based on Geriatric Depression Scale-15 (GDS-15). Mixed model regression was used to assess the relationship of depression dimensions and the emotional nature of the pictures. RESULT: The 73 participants (41% male, average age 80.14±8.01) showed that dysphoria and apathy had opposite associations with emotion-based measures. Dysphoria was associated with higher valence (more positive emotion), while apathy to lower valence. Subjective memory complaint was also associated with lower valence words. Further analysis revealed that apathy was associated with lower pitch and slower speech when describing negative pictures, and dysphoria with a wider pitch range and faster speech for negative pictures. Patients with memory complaints used a narrower pitch range in both positive and negative tasks. CONCLUSION: Dysphoria was associated with heightened emotional reactivity, while apathy showed decreased reactivity. Apathy and memory complaints shared similar speech features. Our preliminary results support the use of speech features in distinguishing different depression dimensions even at the subclinical level, offering promising opportunities for use of technology to enhance both understanding and diagnosis of depression.

Automatic quantification of syntactic complexity in natural spontaneous speech of people with primary progressive aphasia

Aphasiology · 2025-02-17 · 2 citations

Developing digital health technologies for frontotemporal degeneration

Alzheimer s & Dementia · 2025-04-01 · 3 citations

Frontotemporal degeneration (FTD) is a rare neurodegenerative disease in which patients can present with cognitive, behavioral, motor, and speech impairment. Currently, there are no approved therapies available to slow or halt disease progression. Detection and monitoring of patient symptoms is challenging for this heterogeneous disease and has negatively impacted progress in FTD clinical trials. Rapid technological advancements can promote the development of digital health technologies (DHTs) capable of capturing even the most subtle clinical impairments. DHTs are computing platforms being designed to measure meaningful aspects of disease onset and progression. Here we present some of the numerous tools currently being developed to measure changes in the functional domains that become impaired in FTD, challenges faced by developers, and a proposed roadmap for developing fit-for-purpose DHTs that will aid in the development of effective therapies for FTD. HIGHLIGHTS: DHTs are being developed to assess FTD onset and progression. Tool developers must overcome numerous challenges in creating effective applications. Guidance to tool developers aims to benefit FTD drug development and patient care.

Relation between Depression Dimensions and Speech Acoustic and Emotion‐based Features

Alzheimer s & Dementia · 2025-12-01

BACKGROUND: Depression is a common and highly heterogeneous disorder in older adults, often linked to faster cognitive decline. While standard questionnaires are subjective, speech analysis may offer a more objective method for characterizing depression. This study investigates the relationship between speech features and depression dimensions in participants at Mount Sinai Alzheimer's Disease Research Center (ADRC). METHOD: Participants included healthy controls (n = 31) and individuals with mild cognitive impairment (MCI= 22) and Alzheimer's Disease (AD, n = 16). They described three pictures with neutral, negative and positive themes. Speech features were analyzed using automated pipelines and emotion-based variables and acoustic features were used for analysis. Depression dimensions (dysphoria, apathy, hopelessness, and memory complaints) were assessed based on Geriatric Depression Scale-15 (GDS-15). Mixed model regression was used to assess the relationship of depression dimensions and the emotional nature of the pictures. RESULT: The 73 participants (41% male, average age 80.14±8.01) showed that dysphoria and apathy had opposite associations with emotion-based measures. Dysphoria was associated with higher valence (more positive emotion), while apathy to lower valence. Subjective memory complaint was also associated with lower valence words. Further analysis revealed that apathy was associated with lower pitch and slower speech when describing negative pictures, and dysphoria with a wider pitch range and faster speech for negative pictures. Patients with memory complaints used a narrower pitch range in both positive and negative tasks. CONCLUSION: Dysphoria was associated with heightened emotional reactivity, while apathy showed decreased reactivity. Apathy and memory complaints shared similar speech features. Our preliminary results support the use of speech features in distinguishing different depression dimensions even at the subclinical level, offering promising opportunities for use of technology to enhance both understanding and diagnosis of depression.

An automated method for quantifying syntactic complexity of spontaneous speech in Primary Progressive Aphasia

Alzheimer s & Dementia · 2024-12-01

Abstract Background Reduced syntactic complexity is the predominant linguistic impairment unique to the nonfluent agrammatic variant of PPA (naPPA). Reliable, objective and automatic methods for assessing syntactic complexity are currently lacking. Here, we introduce an automatic method for quantifying syntactic complexity in speech. We show that this method differentiates naPPA from other subtypes and monitors disease progression and severity. Method We analyzed speech samples of picture descriptions that were collected from PPA patients and healthy controls (HC) at the University of Pennsylvania during 2000‐2021, including a subset of patients with longitudinal data. Ten syntactic features were automatically extracted from the speech samples and combined into a composite score, using principal component analysis, to represent syntactic complexity. Regression models compared syntactic complexity between PPA phenotypes, covarying for age, sex, education and MMSE. Linear mixed‐effects models tested change in syntactic complexity over time. The correlation of syntactic complexity with neurofilament light chain (NfL) levels in the cerebrospinal fluid (CSF) was tested in naPPA and svPPA patients. Result We examined a total of 136 participants: HC (n=36, 31% males, age 69±8y, MMSE 29±1), naPPA (n=31, 51% males, age 71±8y, MMSE 24±5), lvPPA (n=32, 50% males, age 68±9y, MMSE 22±6) and svPPA (n=37, 49% males, age 64±7y, MMSE 22±7). Z‐scored syntactic complexity scores showed significant group differences, with naPPA scoring lower than HC (β=1.08, p =.004), lvPPA (β=1.19, p =.001) and svPPA (β=.93, p =.008) [Fig‐1]. Sensitivity to disease progression and severity was found only within the naPPA group. The syntactic score of naPPA significantly decreased over time (‐.5 per year, p =.005), significantly different from the 0 decrease in lvPPA ( p =.001) or the .1 decrease in svPPA ( p =.001) [Fig‐2]. naPPA patients had a significant inverse correlation of syntactic complexity score and CSF NfL levels (r=‐.58, p =.03), while svPPA patients did not (r=.39, p =.9) [Fig‐3]. Conclusion We introduce a novel measure of syntactic complexity which can be extracted automatically from brief naturalistic speech samples. We validated this measure by relating it to naPPA and its biological underpinning. We provided objective evaluation of syntax, which has great potential as a non‐invasive biomarker in the dementia research and clinic.

Automatic classification of AD pathology in FTD phenotypes using natural speech

Alzheimer s & Dementia · 2024-04-04 · 11 citations

INTRODUCTION: Screening for Alzheimer's disease neuropathologic change (ADNC) in individuals with atypical presentations is challenging but essential for clinical management. We trained automatic speech-based classifiers to distinguish frontotemporal dementia (FTD) patients with ADNC from those with frontotemporal lobar degeneration (FTLD). METHODS: We trained automatic classifiers with 99 speech features from 1 minute speech samples of 179 participants (ADNC = 36, FTLD = 60, healthy controls [HC] = 89). Patients' pathology was assigned based on autopsy or cerebrospinal fluid analytes. Structural network-based magnetic resonance imaging analyses identified anatomical correlates of distinct speech features. RESULTS: 0.04 AUC for patients versus HC. Noun frequency and pause rate correlated with gray matter volume loss in the limbic and salience networks, respectively. DISCUSSION: Brief naturalistic speech samples can be used for screening FTD patients for underlying ADNC in vivo. This work supports the future development of digital assessment tools for FTD. HIGHLIGHTS: We trained machine learning classifiers for frontotemporal dementia patients using natural speech. We grouped participants by neuropathological diagnosis (autopsy) or cerebrospinal fluid biomarkers. Classifiers well distinguished underlying pathology (Alzheimer's disease vs. frontotemporal lobar degeneration) in patients. We identified important features through an explainable artificial intelligence approach. This work lays the groundwork for a speech-based neuropathology screening tool.

Speech markers of depression dimensions across cognitive status

Alzheimer s & Dementia Diagnosis Assessment & Disease Monitoring · 2024-07-01 · 1 citations

Abstract Introduction Depression and its components significantly impact dementia prediction and severity, necessitating reliable objective measures for quantification. Methods We investigated associations between emotion‐based speech measures (valence, arousal, and dominance) during picture descriptions and depression dimensions derived from the geriatric depression scale (GDS, dysphoria, withdrawal‐apathy‐vigor (WAV), anxiety, hopelessness, and subjective memory complaint). Results Higher WAV was associated with more negative valence (estimate = ‐0.133, p = 0.030). While interactions of apolipoprotein E (APOE) 4 status with depression dimensions on emotional valence did not reach significance, there was a trend for more negative valence with higher dysphoria in those with at least one APOE4 allele (estimate = –0.404, p = 0.0846). Associations were similar irrespective of dementia severity. Discussion Our study underscores the potential utility of speech biomarkers in characterizing depression dimensions. In future research, using emotionally charged stimuli may enhance emotional measure elicitation. The role of APOE on the interaction of speech markers and depression dimensions warrants further exploration with greater sample sizes. Highlights Participants reporting higher apathy used more negative words to describe a neutral picture. Those with higher dysphoria and at least one APOE4 allele also tended to use more negative words. Our results suggest the potential use of speech biomarkers in characterizing depression dimensions.

Automated Measures of Syntactic Complexity in Natural Speech Production: Older and Younger Adults as a Case Study

Journal of Speech Language and Hearing Research · 2024-01-12 · 14 citations

PURPOSE: Multiple methods have been suggested for quantifying syntactic complexity in speech. We compared eight automated syntactic complexity metrics to determine which best captured verified syntactic differences between old and young adults. METHOD: = 36, ages 53-89 years) healthy participants, manually transcribed and segmented into sentences. We manually verified that older participants produced fewer complex structures. We developed a metric of syntactic complexity using automatically extracted syntactic structures as features in a multidimensional metric. We compared our metric to seven other metrics: Yngve score, Frazier score, Frazier-Roark score, developmental level, syntactic frequency, mean dependency distance, and sentence length. We examined the success of each metric in identifying the age group using logistic regression models. We repeated the analysis with automatic transcription and segmentation using an automatic speech recognition (ASR) system. RESULTS: Our multidimensional metric was successful in predicting age group (area under the curve [AUC] = 0.87), and it performed better than the other metrics. High AUCs were also achieved by the Yngve score (0.84) and sentence length (0.84). However, in a fully automated pipeline with ASR, the performance of these two metrics dropped (to 0.73 and 0.46, respectively), while the performance of the multidimensional metric remained relatively high (0.81). CONCLUSIONS: Syntactic complexity in spontaneous speech can be quantified by directly assessing syntactic structures and considering them in a multivariable manner. It can be derived automatically, saving considerable time and effort compared to manually analyzing large-scale corpora, while maintaining high face validity and robustness. SUPPLEMENTAL MATERIAL: https://doi.org/10.23641/asha.24964179.

Digital speech markers of cognitive impairment in ALS‐FTD spectrum disorders

Alzheimer s & Dementia · 2024-12-01

Abstract Background Comorbid cognitive impairment in amyotrophic lateral sclerosis (i.e., ALS‐FTD), is associated with adverse clinical outcomes and survival. It often exhibits clinical features of behavioral variant frontotemporal dementia (bvFTD) and/or non‐fluent agrammatic primary progressive aphasia (naPPA). Traditional cognitive assessments, such as the Edinburgh Cognitive Assessment Scale (ECAS), require trained personnel for administration. An additional repeatable, objective approach involves automatic analysis of picture descriptions. Prior research identified acoustic and lexical‐semantic impairments in bvFTD, PPAs, and ALS‐FTD (acoustics). This study extends the analysis to the complete speech feature set (n=30) in ALS‐FTD spectrum disorders. Method ANCOVAs (FDR adjusted) compared picture descriptions in n=51 ALS, n=22 ALS‐FTD (n=10 bvFTD‐dominant, n=4 naPPA‐dominant), and n=12 ALS with ECAS‐defined mild cognitive/behavioural impairment (ALS‐MCI). Backward selection identified the best feature subset. Post‐hoc assessments examined associations with ECAS behavioral, language, and executive scores. Longitudinal analysis compared within‐individual change over time in n=11 ALS vs. n=9 ALS+cog (combining ALS‐MCI with ALS‐FTD) with available longitudinal data. All models adjusted for bulbar motor scores. Result ALS‐FTD significantly differed from ALS and ALS‐MCI on twelve features at first visit: slow speaking and articulatory rates, reduced word count, less percent speech time, shorter speech segments, and longer and more frequent pausing. ALS‐FTD also exhibited fewer adjectives and prepositions, more repetitions, lower lexical diversity, and words with a lower mean age of acquisition (AoA). The best model comprising three features (percent speech, prepositions, and lexical diversity) strongly distinguished ALS‐FTD from ALS (AUC=0.90). ALS‐MCI did not differ from ALS on any speech measure. Acoustic measures were all significantly associated with behavioral, language, and executive scores. Fewer prepositions related to worse behavioral and language scores, and more repetitions and lower lexical diversity related to worse language scores. Mean AoA and articulatory rate were not linked to any ECAS sub‐domains. Longitudinally, ALS+cog exhibited increased partial words and reduced word AoA, while ALS showed no significant change. Conclusion Automated speech analysis strongly distinguishes ALS‐FTD from ALS, declines with cognitive severity, and captures distinctive longitudinal changes in ALS‐FTD. These results highlight the potential of automated speech analysis as an objective tool for assessing cognitive impairment in ALS.

Automated analysis of story recall tasks produced by AD and MCI patients

Alzheimer s & Dementia · 2024-12-01

Abstract Background Story recall tasks are often employed in clinical settings to measure verbal episodic memory in individuals with mild cognitive impairment (MCI) and Alzheimer’s disease (AD). Traditional analyses (e.g. total number of recalled words) provide useful but limited information compared to detailed analyses of lexical performance. In this study, we analyzed story recall data produced by patients with AD or MCI using an automated method. Method We analyzed digitized recordings of the Craft story recall task produced by 15 individuals with MCI (age=65.8±10.14, females=40%, education=16.07±2.94), 27 individuals with AD (age=65.19±6.82, females=37%, education=15.42±3.01) and 14 healthy age‐matched individuals (HC; age=58.86±16.16, females=50%, education=17.29±2.16). The groups did not differ in any demographic characteristics. Our automated pipeline identified the number of verbatim and paraphrase recalls, calculated semantic similarities between the original story and recalled immediate and delayed stories using word embeddings and rated several lexical characteristics, including word frequency, concreteness, and word length. Two‐way ANOVA tests examined the effect of Group (HC vs. MCI vs. AD), Task (immediate vs. delayed), and their interactions. As the interactions of Group and Task were not significant, post‐hoc analyses examined significant pairwise group differences in immediate and delayed recalls, separately. Result Participants with AD produced the fewest verbatim and total recalled words, followed by MCI patients and HC in both immediate and delayed recalls (Figure 1A–B, Table 1). AD patients produced fewer paraphrase recalls compared to the other groups in the delayed recall only (Figure 1C). AD group’s recalled story was also the least similar to the original story, followed by MCI and HC (Figure 1D). Also, individuals with AD produced words that were more frequent and less concrete during immediate and delayed recalls compared to MCI patients and HC (Figure 1E–F). Words produced by AD patients were also shorter than those produced by the MCI and HC groups during the immediate recall (Figure 1G). Conclusion Our method highlights that digitized language‐based tests of verbal episodic memory can be automatically analyzed. These capture traditional language features that cannot be extracted manually without massive effort and promise to provide novel, granular, digitized biomarkers that traditional assessment methods of neuropsychological tests cannot show.