Expanding the use of brief cognitive assessments to detect suspected early-stage cognitive impairment in primary care.

INTRODUCTION: Mild cognitive impairment remains substantially underdiagnosed, especially in disadvantaged populations. Failure to diagnose deprives patients and families of the opportunity to treat reversible causes, make necessary life and lifestyle changes and receive disease-modifying treatments if caused by Alzheimer's disease. Primary care, as the entry point for most, plays a critical role in improving detection rates. METHODS: We convened a Work Group of national experts to develop consensus recommendations for policymakers and third-party payers on ways to increase the use of brief cognitive assessments (BCAs) in primary care. RESULTS: The group recommended three strategies to promote routine use of BCAs: providing primary care clinicians with suitable assessment tools; integrating BCAs into routine workflows; and crafting payment policies to encourage adoption of BCAs. DISSCUSSION: Sweeping changes and actions of multiple stakeholders are necessary to improve detection rates of mild cognitive impairment so that patients and families may benefit from timely interventions.

Detecting dementia among older, ethnically diverse residents of rural subsidized housing.

Rural, ethnically diverse residents face at least twice the risk of Alzheimer's disease than urban residents. Chronic diseases such as diabetes and hypertension which increase dementia risk are more prevalent in rural areas with less access to specialty providers. A home-based approach for increasing dementia detection and treatment rates was tested among rural residents of government-assisted independent living facilities (N = 139; 78% non-White, and 70% with health literacy below 5th grade). Of 28 residents identified at risk during cognitive screening, 25 agreed to further in-depth assessment by adult gerontological nurse practitioners (AGNP). Fifteen of 25 (60%) completing consequent primary provider referrals were diagnosed with dementia and receiving new care (statistically significant; [χ2(1) = 76.67, p < .001, Phi = 0.743]). Home-based dementia management through a community engagement approach can help to meet the Healthy People 2030 goals of earlier detection and treatment and reduce the length of costly institutionalizations.

A program to improve detection of undiagnosed dementia in primary care and its association with healthcare utilization.

OBJECTIVE: Alzheimer's disease and related dementias are common and costly, with increased healthcare utilization for patients with these disorders. The current study describes a novel dementia detection program for veterans and examines whether program-eligible patients have higher healthcare utilization than age-matched comparison patients. DESIGN: Using a telephone-based case-finding approach, the detection program used risk factors available in the electronic medical record (EMR) and telephone-based brief cognitive screening. Holding illness severity constant, dementia detection and healthcare utilization were compared across age-matched groups with and without program risk factors. SETTING: Five Veterans Affairs Healthcare Network Upstate New York primary care clinics. PARTICIPANTS: Veterans aged 70 years and older. MEASUREMENTS: EMR data and the Charlson comorbidity index. RESULTS: Program-eligible patients (n = 5,333) demonstrated significantly greater levels of medical comorbidity relative to comparison patients and were on average more than twice as likely to be admitted to the hospital. They also had nearly double the number of outpatient visits to several services. Similar patterns were seen in those who screened positive on a brief cognitive measure, compared with those who screened negative. CONCLUSIONS: A novel program using EMR data to assist in the detection of newly diagnosed dementia in a clinical setting was found to be useful in identifying older veterans with multiple comorbid medical conditions and increased utilization of hospital and clinic services. Results suggest undetected cognitive impairment and dementia may significantly contribute to healthcare utilization and costs of care in older veterans.

Automatic selection of spoken language biomarkers for dementia detection.

This paper analyzes diverse features extracted from spoken language to select the most discriminative ones for dementia detection. We present a two-step feature selection (FS) approach: Step 1 utilizes filter methods to pre-screen features, and Step 2 uses a novel feature ranking (FR) method, referred to as dual dropout ranking (DDR), to rank the screened features and select spoken language biomarkers. The proposed DDR is based on a dual-net architecture that separates FS and dementia detection into two neural networks (namely, the operator and selector). The operator is trained on features obtained from the selector to reduce classification or regression loss. The selector is optimized to predict the operator's performance based on automatic regularization. Results show that the approach significantly reduces feature dimensionality while identifying small feature subsets that achieve comparable or superior performance compared with the full, default feature set. The Python codes are available at https://github.com/kexquan/dual-dropout-ranking.

Development of an AI-Based Predictive Algorithm for Early Diagnosis of High-Risk Dementia Groups among the Elderly: Utilizing Health Lifelog Data.

BACKGROUND/OBJECTIVES: This study aimed to develop a predictive algorithm for the early diagnosis of dementia in the high-risk group of older adults using artificial intelligence technologies. The objective is to create an accessible diagnostic method that does not rely on traditional medical equipment, thereby improving the early detection and management of dementia. METHODS: Lifelog data from wearable devices targeting this high-risk group were collected from the AI Hub platform. Various indicators from these data were analyzed to develop a dementia diagnostic model. Machine learning techniques such as Logistic Regression, Random Forest, LightGBM, and Support Vector Machine were employed. Data augmentation techniques were applied to address data imbalance, thereby enhancing the model performance. RESULTS: Data augmentation significantly improved the model's accuracy in classifying dementia cases. Specifically, in gait data, the SVM model performed with an accuracy of 0.879. In sleep data, a Logistic Regression was performed, yielding an accuracy of 0.818. This indicates that the lifelog data can effectively contribute to the early diagnosis of dementia, providing a practical solution that can be easily integrated into healthcare systems. CONCLUSIONS: This study demonstrates that lifelog data, which are easily collected in daily life, can significantly enhance the accessibility and efficiency of dementia diagnosis, aiding in the effective use of medical resources and potentially delaying disease progression.

An Overview of the ADReSS-M Signal Processing Grand Challenge on Multilingual Alzheimer's Dementia Recognition Through Spontaneous Speech.

The ADReSS-M Signal Processing Grand Challenge was held at the 2023 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023. The challenge targeted difficult automatic prediction problems of great societal and medical relevance, namely, the detection of Alzheimer's Dementia (AD) and the estimation of cognitive test scoress. Participants were invited to create models for the assessment of cognitive function based on spontaneous speech data. Most of these models employed signal processing and machine learning methods. The ADReSS-M challenge was designed to assess the extent to which predictive models built based on speech in one language generalise to another language. The language data compiled and made available for ADReSS-M comprised English, for model training, and Greek, for model testing and validation. To the best of our knowledge no previous shared research task investigated acoustic features of the speech signal or linguistic characteristics in the context of multilingual AD detection. This paper describes the context of the ADReSS-M challenge, its data sets, its predictive tasks, the evaluation methodology we employed, our baseline models and results, and the top five submissions. The paper concludes with a summary discussion of the ADReSS-M results, and our critical assessment of the future outlook in this field.

Detecting Dementia from Face-Related Features with Automated Computational Methods.

Alzheimer's disease (AD) is a type of dementia that is more likely to occur as people age. It currently has no known cure. As the world's population is aging quickly, early screening for AD has become increasingly important. Traditional screening methods such as brain scans or psychiatric tests are stressful and costly. The patients are likely to feel reluctant to such screenings and fail to receive timely intervention. While researchers have been exploring the use of language in dementia detection, less attention has been given to face-related features. The paper focuses on investigating how face-related features can aid in detecting dementia by exploring the PROMPT dataset that contains video data collected from patients with dementia during interviews. In this work, we extracted three types of features from the videos, including face mesh, Histogram of Oriented Gradients (HOG) features, and Action Units (AU). We trained traditional machine learning models and deep learning models on the extracted features and investigated their effectiveness in dementia detection. Our experiments show that the use of HOG features achieved the highest accuracy of 79% in dementia detection, followed by AU features with 71% accuracy, and face mesh features with 66% accuracy. Our results show that face-related features have the potential to be a crucial indicator in automated computational dementia detection.

Speaker-turn aware diarization for speech-based cognitive assessments.

Introduction: Speaker diarization is an essential preprocessing step for diagnosing cognitive impairments from speech-based Montreal cognitive assessments (MoCA). Methods: This paper proposes three enhancements to the conventional speaker diarization methods for such assessments. The enhancements tackle the challenges of diarizing MoCA recordings on two fronts. First, multi-scale channel interdependence speaker embedding is used as the front-end speaker representation for overcoming the acoustic mismatch caused by far-field microphones. Specifically, a squeeze-and-excitation (SE) unit and channel-dependent attention are added to Res2Net blocks for multi-scale feature aggregation. Second, a sequence comparison approach with a holistic view of the whole conversation is applied to measure the similarity of short speech segments in the conversation, which results in a speaker-turn aware scoring matrix for the subsequent clustering step. Third, to further enhance the diarization performance, we propose incorporating a pairwise similarity measure so that the speaker-turn aware scoring matrix contains both local and global information across the segments. Results: Evaluations on an interactive MoCA dataset show that the proposed enhancements lead to a diarization system that outperforms the conventional x-vector/PLDA systems under language-, age-, and microphone-mismatch scenarios. Discussion: The results also show that the proposed enhancements can help hypothesize the speaker-turn timestamps, making the diarization method amendable to datasets without timestamp information.

TOWARDS INTERPRETABILITY OF SPEECH PAUSE IN DEMENTIA DETECTION USING ADVERSARIAL LEARNING.

Speech pause is an effective biomarker in dementia detection. Recent deep learning models have exploited speech pauses to achieve highly accurate dementia detection, but have not exploited the interpretability of speech pauses, i.e., what and how positions and lengths of speech pauses affect the result of dementia detection. In this paper, we will study the positions and lengths of dementia-sensitive pauses using adversarial learning approaches. Specifically, we first utilize an adversarial attack approach by adding the perturbation to the speech pauses of the testing samples, aiming to reduce the confidence levels of the detection model. Then, we apply an adversarial training approach to evaluate the impact of the perturbation in training samples on the detection model. We examine the interpretability from the perspectives of model accuracy, pause context, and pause length. We found that some pauses are more sensitive to dementia than other pauses from the model's perspective, e.g., speech pauses near to the verb "is". Increasing lengths of sensitive pauses or adding sensitive pauses leads the model inference to Alzheimer's Disease (AD), while decreasing the lengths of sensitive pauses or deleting sensitive pauses leads to non-AD.

WavBERT: Exploiting Semantic and Non-semantic Speech using Wav2vec and BERT for Dementia Detection.

In this paper, we exploit semantic and non-semantic information from patient's speech data using Wav2vec and Bidirectional Encoder Representations from Transformers (BERT) for dementia detection. We first propose a basic WavBERT model by extracting semantic information from speech data using Wav2vec, and analyzing the semantic information using BERT for dementia detection. While the basic model discards the non-semantic information, we propose extended WavBERT models that convert the output of Wav2vec to the input to BERT for preserving the non-semantic information in dementia detection. Specifically, we determine the locations and lengths of inter-word pauses using the number of blank tokens from Wav2vec where the threshold for setting the pauses is automatically generated via BERT. We further design a pre-trained embedding conversion network that converts the output embedding of Wav2vec to the input embedding of BERT, enabling the fine-tuning of WavBERT with non-semantic information. Our evaluation results using the ADReSSo dataset showed that the WavBERT models achieved the highest accuracy of 83.1% in the classification task, the lowest Root-Mean-Square Error (RMSE) score of 4.44 in the regression task, and a mean F1 of 70.91% in the progression task. We confirmed the effectiveness of WavBERT models exploiting both semantic and non-semantic speech.

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.

Effects of an Enhanced Training on Primary Care Providers Knowledge, Attitudes, Service and Skills of Dementia Detection: A Cluster Randomized Trial.

Background: Effective training programs for primary care providers (PCPs) to support dementia detection are needed, especially in developing countries. This study aimed to investigate the effect of an enhanced training on the competency and service of PCPs for dementia detection. Methods: We conducted a cluster randomized trial in Beijing, China. Community healthcare centers (CHCs) located in Fengtai or Fangshan District were eligible. The enrolled CHCs in each district were randomly assigned to the standard or the enhanced training group at a 1:1 ratio. PCPs serving older adults in enrolled CHCs were eligible to participate. The standard training group received three-hour didactic lectures, three monthly supervisions, 3 months of online support and dementia screening packages. The enhanced training group additionally received three monthly face-to-face supervisions and 3 months of online support. The participants became aware of their group membership at the end of the standard training. The knowledge, attitudes, service, and skills regarding dementia detection were assessed using questionnaires and submitted dementia detection records, respectively. Results: A total of 23 and 21 CHCs were randomly assigned to the standard and the enhanced training group, respectively, and 58 participants from 20 CHCs assigned to the standard training group and 48 from 16 CHCs assigned to the enhanced training group were included in the final analysis (mean age 37.5 years, and 67.0% women). A significant increase in the knowledge score was found in both groups, but the increase was similar in the two groups (P = 0.262). The attitude score remained stable in both groups, and no between-group difference was found. Compared with the baseline, both groups reported an increase in dementia detection service, especially the enhanced training group (24.1% to 31.0% in the standard training group and 14.6% to 45.8% in the enhanced training group). The completion rate and accuracy of submitted dementia detection records in the enhanced training group were both significantly higher than those in the standard training group (both P < 0.001). Conclusion: The enhanced training had similar effect on the knowledge of PCPs comparing with the standard training, but was better on continuous service and skills of PCPs related to dementia detection. Trial registration: www.ClinicalTrials.gov, identifier: NCT02782000. Registration date: May 2016. The trial was completed in July 2017.

Temporal Integration of Text Transcripts and Acoustic Features for Alzheimer's Diagnosis Based on Spontaneous Speech.

Background: Advances in machine learning (ML) technology have opened new avenues for detection and monitoring of cognitive decline. In this study, a multimodal approach to Alzheimer's dementia detection based on the patient's spontaneous speech is presented. This approach was tested on a standard, publicly available Alzheimer's speech dataset for comparability. The data comprise voice samples from 156 participants (1:1 ratio of Alzheimer's to control), matched by age and gender. Materials and Methods: A recently developed Active Data Representation (ADR) technique for voice processing was employed as a framework for fusion of acoustic and textual features at sentence and word level. Temporal aspects of textual features were investigated in conjunction with acoustic features in order to shed light on the temporal interplay between paralinguistic (acoustic) and linguistic (textual) aspects of Alzheimer's speech. Combinations between several configurations of ADR features and more traditional bag-of-n-grams approaches were used in an ensemble of classifiers built and evaluated on a standardised dataset containing recorded speech of scene descriptions and textual transcripts. Results: Employing only semantic bag-of-n-grams features, an accuracy of 89.58% was achieved in distinguishing between Alzheimer's patients and healthy controls. Adding temporal and structural information by combining bag-of-n-grams features with ADR audio/textual features, the accuracy could be improved to 91.67% on the test set. An accuracy of 93.75% was achieved through late fusion of the three best feature configurations, which corresponds to a 4.7% improvement over the best result reported in the literature for this dataset. Conclusion: The proposed combination of ADR audio and textual features is capable of successfully modelling temporal aspects of the data. The machine learning approach toward dementia detection achieves best performance when ADR features are combined with strong semantic bag-of-n-grams features. This combination leads to state-of-the-art performance on the AD classification task.

Classifying Alzheimer's Disease Using Audio and Text-Based Representations of Speech.

Alzheimer's Disease (AD) is a form of dementia that affects the memory, cognition, and motor skills of patients. Extensive research has been done to develop accessible, cost-effective, and non-invasive techniques for the automatic detection of AD. Previous research has shown that speech can be used to distinguish between healthy patients and afflicted patients. In this paper, the ADReSS dataset, a dataset balanced by gender and age, was used to automatically classify AD from spontaneous speech. The performance of five classifiers, as well as a convolutional neural network and long short-term memory network, was compared when trained on audio features (i-vectors and x-vectors) and text features (word vectors, BERT embeddings, LIWC features, and CLAN features). The same audio and text features were used to train five regression models to predict the Mini-Mental State Examination score for each patient, a score that has a maximum value of 30. The top-performing classification models were the support vector machine and random forest classifiers trained on BERT embeddings, which both achieved an accuracy of 85.4% on the test set. The best-performing regression model was the gradient boosting regression model trained on BERT embeddings and CLAN features, which had a root mean squared error of 4.56 on the test set. The performance on both tasks illustrates the feasibility of using speech to classify AD and predict neuropsychological scores.

Towards Computer-Based Automated Screening of Dementia Through Spontaneous Speech.

Dementia, a prevalent disorder of the brain, has negative effects on individuals and society. This paper concerns using Spontaneous Speech (ADReSS) Challenge of Interspeech 2020 to classify Alzheimer's dementia. We used (1) VGGish, a deep, pretrained, Tensorflow model as an audio feature extractor, and Scikit-learn classifiers to detect signs of dementia in speech. Three classifiers (LinearSVM, Perceptron, 1NN) were 59.1% accurate, which was 3% above the best-performing baseline models trained on the acoustic features used in the challenge. We also proposed (2) DemCNN, a new PyTorch raw waveform-based convolutional neural network model that was 63.6% accurate, 7% more accurate then the best-performing baseline linear discriminant analysis model. We discovered that audio transfer learning with a pretrained VGGish feature extractor performs better than the baseline approach using automatically extracted acoustic features. Our DepCNN exhibits good generalization capabilities. Both methods presented in this paper offer progress toward new, innovative, and more effective computer-based screening of dementia through spontaneous speech.

Cognitive assessments for the early diagnosis of dementia after stroke.

The early detection of poststroke dementia (PSD) is important for medical practitioners to customize patient treatment programs based on cognitive consequences and disease severity progression. The aim is to diagnose and detect brain degenerative disorders as early as possible to help stroke survivors obtain early treatment benefits before significant mental impairment occurs. Neuropsychological assessments are widely used to assess cognitive decline following a stroke diagnosis. This study reviews the function of the available neuropsychological assessments in the early detection of PSD, particularly vascular dementia (VaD). The review starts from cognitive impairment and dementia prevalence, followed by PSD types and the cognitive spectrum. Finally, the most usable neuropsychological assessments to detect VaD were identified. This study was performed through a PubMed and ScienceDirect database search spanning the last 10 years with the following keywords: "post-stroke"; "dementia"; "neuro-psychological"; and "assessments". This study focuses on assessing VaD patients on the basis of their stroke risk factors and cognitive function within the first 3 months after stroke onset. The search strategy yielded 535 articles. After application of inclusion and exclusion criteria, only five articles were considered. A manual search was performed and yielded 14 articles. Twelve articles were included in the study design and seven articles were associated with early dementia detection. This review may provide a means to identify the role of neuropsychological assessments as early PSD detection tests.