Automatic speech analysis to early detect functional cognitive decline in elderly population

Detailed Analysis of Responses from Older Adults through Natural Speech: Comparison of Questions by AI Agents and Humans

In recent years, an increasing number of studies have begun to use conversational data in spontaneous speech to estimate cognitive function in older people. The providers of spontaneous speech with older people used to be physicians and licensed psychologists, but it is now possible to have conversations with fully automatic AI agents. However, it has not yet been clarified what differences exist in conversational communication with older people when the examiner is either a human or an AI agent. In this study, elderly people living in the community and attending a silver human resource center and a day service center were the subjects. Dialogues were conducted using generic interview items for estimating cognitive function through daily conversation, which were developed through research on estimation methods for cognitive function. From the data obtained from the dialogues, we compared the effects of human-AI interaction on the number of utterances, speaking time, and silence time. This study was conducted at a facility in Japan and included 32 subjects (12 males and 20 females). The results showed significant differences between human and AI dialogue in the number of utterances and silent time. This study suggests the effectiveness of AI in communication with older people and explores the possibility of using AI in social welfare.

Automatic Spontaneous Speech Analysis for the Detection of Cognitive Functional Decline in Older Adults: Multilanguage Cross-Sectional Study

BACKGROUND

The rise in life expectancy is associated with an increase in long-term and gradual cognitive decline. Treatment effectiveness is enhanced at the early stage of the disease. Therefore, there is a need to find low-cost and ecological solutions for mass screening of community-dwelling older adults.

OBJECTIVE

This work aims to exploit automatic analysis of free speech to identify signs of cognitive function decline.

METHODS

A sample of 266 participants older than 65 years were recruited in Italy and Spain and were divided into 3 groups according to their Mini-Mental Status Examination (MMSE) scores. People were asked to tell a story and describe a picture, and voice recordings were used to extract high-level features on different time scales automatically. Based on these features, machine learning algorithms were trained to solve binary and multiclass classification problems by using both mono- and cross-lingual approaches. The algorithms were enriched using Shapley Additive Explanations for model explainability.

RESULTS

In the Italian data set, healthy participants (MMSE score≥27) were automatically discriminated from participants with mildly impaired cognitive function (20≤MMSE score≤26) and from those with moderate to severe impairment of cognitive function (11≤MMSE score≤19) with accuracy of 80% and 86%, respectively. Slightly lower performance was achieved in the Spanish and multilanguage data sets.

CONCLUSIONS

This work proposes a transparent and unobtrusive assessment method, which might be included in a mobile app for large-scale monitoring of cognitive functionality in older adults. Voice is confirmed to be an important biomarker of cognitive decline due to its noninvasive and easily accessible nature.

Harnessing Speech-Derived Digital Biomarkers to Detect and Quantify Cognitive Decline Severity in Older Adults

INTRODUCTION

Current cognitive assessments suffer from floor/ceiling and practice effects, poor psychometric performance in mild cases, and repeated assessment effects. This study explores the use of digital speech analysis as an alternative tool for determining cognitive impairment. The study specifically focuses on identifying the digital speech biomarkers associated with cognitive impairment and its severity.

METHODS

We recruited older adults with varying cognitive health. Their speech data, recorded via a wearable microphone during the reading aloud of a standard passage, were processed to derive digital biomarkers such as timing, pitch, and loudness. Cohen's d effect size highlighted group differences, and correlations were drawn to the Montreal Cognitive Assessment (MoCA). A stepwise approach using a Random Forest model was implemented to distinguish cognitive states using speech data and predict MoCA scores based on highly correlated features.

RESULTS

The study comprised 59 participants, with 36 demonstrating cognitive impairment and 23 serving as cognitively intact controls. Among all assessed parameters, similarity, as determined by Dynamic Time Warping (DTW), exhibited the most substantial positive correlation (rho = 0.529, p < 0.001), while timing parameters, specifically the ratio of extra words, revealed the strongest negative correlation (rho = -0.441, p < 0.001) with MoCA scores. Optimal discriminative performance was achieved with a combination of four speech parameters: total pause time, speech-to-pause ratio, similarity via DTW, and intelligibility via DTW. Precision and balanced accuracy scores were found to be 88.1 ± 1.2% and 76.3 ± 1.3%, respectively.

DISCUSSION

Our research proposes that reading-derived speech data facilitates the differentiation between cognitively impaired individuals and cognitively intact, age-matched older adults. Specifically, parameters based on timing and similarity within speech data provide an effective gauge of cognitive impairment severity. These results suggest speech analysis as a viable digital biomarker for early detection and monitoring of cognitive impairment, offering novel approaches in dementia care.

Learning algorithms estimate pose and detect motor anomalies in flies exposed to minimal doses of a toxicant

Pesticide exposure, even at low doses, can have detrimental effects on ecosystems. This study aimed at validating the use of machine learning for recognizing motor anomalies, produced by minimal insecticide exposure on a model insect species. The Mediterranean fruit fly, Ceratitis capitata (Diptera: Tephritidae), was exposed to food contaminated with low concentrations of Carlina acaulis essential oil (EO). A deep learning approach enabled fly pose estimation on video recordings in a custom-built arena. Five machine learning algorithms were trained on handcrafted features, extracted from the predicted pose, to distinguish treated individuals. Random Forest and K-Nearest Neighbor algorithms best performed, with an area under the receiver operating characteristic (ROC) curve of 0.75 and 0.73, respectively. Both algorithms achieved an accuracy of 0.71. Results show the machine learning potential for detecting sublethal effects arising from insecticide exposure on fly motor behavior, which could also affect other organisms and environmental health.

Automated image-based analysis unveils acute effects due to sub-lethal pesticide doses exposure. ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY. IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY. ANNUAL INTERNATIONAL CONFERENCE 2023;2023:1-4. [PMID: 38082662 DOI: 10.1109/embc40787.2023.10340800] [Citation(s) in RCA: 0] [Impact Index Per Article: 0]

Pesticides are still abused in modern agriculture. The effects of their exposure to even sub-lethal doses can be detrimental to ecosystem stability and human health. This work aims to validate the use of machine learning techniques for recognizing motor abnormalities and to assess any effect post-exposure to a minimal dosage of these substances on a model organism, gaining insights into potential risks for human health. The test subject was the Mediterranean fruit fly, Ceratitis capitata (Wiedemann) (Diptera: Tephritidae), exposed to food contaminated with the LC30 of Carlina acaulis essential oil. A deep learning approach enabled the pose estimation within an arena. Statistical analysis highlighted the most significant features between treated and untreated groups. Based on this analysis, two learning-based algorithms, Random Forest (RF) and XGBoost were employed. The results were compared through different metrics. RF algorithm generated a model capable of distinguishing treated subjects with an area under the receiver operating characteristic curve of 0.75 and an accuracy of 0.71. Through an image-based analysis, this study revealed acute effects due to minimal pesticide doses. So, even small amounts of these biocides drifted far from distribution areas may negatively affect the environment and humans.

Fair and Privacy-Preserving Alzheimer's Disease Diagnosis Based on Spontaneous Speech Analysis via Federated Learning

As the most common neurodegenerative disease among older adults, Alzheimer's disease (AD) would lead to loss of memory, impaired language and judgment, gait disorders, and other cognitive deficits severe enough to interfere with daily activities and significantly diminish quality of life. Recent research has shown promising results in automatic AD diagnosis via speech, leveraging the advances of deep learning in the audio domain. However, most existing studies rely on a centralized learning framework which requires subjects' voice data to be gathered to a central server, raising severe privacy concerns. To resolve this, in this paper, we propose the first federated-learning-based approach for achieving automatic AD diagnosis via spontaneous speech analysis while ensuring the subjects' data privacy. Extensive experiments under various federated learning settings on the ADReSS challenge dataset show that the proposed model can achieve high accuracy for AD detection while achieving privacy preservation. To ensure fairness of the model performance across clients in federated settings, we further deploy fair aggregation mechanisms, particularly q-FEDAvg and q-FEDSgd, which greatly reduces the algorithmic biases due to the data heterogeneity among the clients. Clinical Relevance -The experiments were conducted on publicly available clinical datasets. No humans or animals were involved.

The relationship of platelet-to-lymphocyte ratio with cognitive decline in T2DM

PURPOSE

We aimed to investigate the role of platelet-to-lymphocyte ratio (PLR) in cognitive decline in patients with type 2 diabetes mellitus (T2DM).

METHODS

A total number of 261 T2DM patients were enrolled in this study. The T2DM patients were divided into two groups according to the median of PLR (PLR < 96.5, n = 130; PLR ≥ 96.5, n = 131). Cognitive impairment was defined as Mini-mental State Examination score ≤ 26. Student's t test and Chi-square test were used to test the difference between the groups, and logistics regression analysis were performed to verify whether high PLR was an independent factor for cognitive impairment.

RESULTS

T2DM patients with cognitive impairment had significantly higher PLR level when compared with the simple diabetes group (p = 0.003). Incidence of cognitive impairment was higher in the high PLR group, compared to low PLR group (p = 0.040). Multivariate logistic regression analysis suggested that PLR was a risk biomarker of cognitive decline in T2DM patients (odds ratio [OR] = 1.010, 95% CI: 1.001-1.018, p = 0.013).

CONCLUSIONS

We demonstrated that a higher PLR was associated with cognitive decline in T2DM patients. The PLR may help to identify high-risk patients in time and provide clues for further prevention of cognitive dysfunction in T2DM patients.

Acoustic and Language Based Deep Learning Approaches for Alzheimer's Dementia Detection From Spontaneous Speech

Current methods for early diagnosis of Alzheimer's Dementia include structured questionnaires, structured interviews, and various cognitive tests. Language difficulties are a major problem in dementia as linguistic skills break down. Current methods do not provide robust tools to capture the true nature of language deficits in spontaneous speech. Early detection of Alzheimer's Dementia (AD) from spontaneous speech overcomes the limitations of earlier approaches as it is less time consuming, can be done at home, and is relatively inexpensive. In this work, we re-implement the existing NLP methods, which used CNN-LSTM architectures and targeted features from conversational transcripts. Our work sheds light on why the accuracy of these models drops to 72.92% on the ADReSS dataset, whereas, they gave state of the art results on the DementiaBank dataset. Further, we build upon these language input-based recurrent neural networks by devising an end-to-end deep learning-based solution that performs a binary classification of Alzheimer's Dementia from the spontaneous speech of the patients. We utilize the ADReSS dataset for all our implementations and explore the deep learning-based methods of combining acoustic features into a common vector using recurrent units. Our approach of combining acoustic features using the Speech-GRU improves the accuracy by 2% in comparison to acoustic baselines. When further enriched by targeted features, the Speech-GRU performs better than acoustic baselines by 6.25%. We propose a bi-modal approach for AD classification and discuss the merits and opportunities of our approach.

Predicting Mini-Mental Status Examination Scores through Paralinguistic Acoustic Features of Spontaneous Speech

Speech analysis could provide an indicator of cognitive health and help develop clinical tools for automatically detecting and monitoring cognitive health progression. The Mini Mental Status Examination (MMSE) is the most widely used screening tool for cognitive health. But the manual operation of MMSE restricts its screening within primary care facilities. An automatic screening tool has the potential to remedy this situation. This study aims to assess the association between acoustic features of spontaneous speech and assess whether acoustic features can be used to automatically predict MMSE score. We assessed the effectiveness of paralinguistic feature set for MMSE score prediction on a balanced sample of DementiaBank's Pitt spontaneous speech dataset, with patients matched by gender and age. Linear regression analysis shows that fusion of acoustic features, age, sex and years of education provides better results (mean absolute error, MAE = 4.97, and R² = 0.261) than acoustic features alone (MAE = 5.66 and R² =0.125) and age, gender and education level alone (MAE of 5.36 and R² =0.17). This suggests that the acoustic features of spontaneous speech are an important part of an automatic screening tool for cognitive impairment detection.Clinical relevance- We hereby present a method for automatic screening of cognitive health. It is based on acoustic information of speech, a ubiquitous source of data, therefore being cost-efficient, non-invasive and with little infrastructure required.