A - 53 A Digital Health Solution for Early Detection of Cognitive Impairment in Primary Care

Archives of Clinical Neuropsychology · 2023

• Psychology
• Cognitive psychology
• Audiology

Abstract Objective To determine which task or combination of tasks provided the most effective way to differentiate cognitively impaired (CI) from cognitively normal (CN) participants in under 5 minutes and to ensure that classification accuracy was equal to or better than a traditional brief cognitive screening task, the Quick Mild Cognitive Impairment (Qmci) screen. Method CN (n = 53) and CI (n = 51) participants completed a risk assessment task, a symbol matching (SM) task, and four speech-language tasks, followed by a second administration of SM to examine utility of practice effects administered on an iPad. Participants also completed the Qmci. Eleven models were tested using Bayesian adaptive regression trees. Results The top three models all included the two SM variables: the one with SM by itself (estimated c = 0.91), one with SM and features from a personal narrative task (c = 0.94), and one with SM and a counting backwards task (c = 0.90). Models with picture description and procedural discourse tasks performed the worst. For comparison, the QMCI-only model yielded c = 0.91. Conclusions A combination of working memory/processing speed and acoustic and linguistic variables from recalling a personal story achieved a high level of classification accuracy, slightly exceeding that of a traditional cognitive screening task. The inclusion of both verbal and nonverbal tasks may be an important feature, allowing for cognitive screening of individuals who are not able to do one type of task or the other. Future work is planned to examine this shortened tool in a pragmatic clinical trial in two primary care clinics.

How people talk about each other: Modeling Generalized Intergroup Bias and Emotion

2023 · 3 citations

• Computer Science
• Computer Science
• Linguistics

Venkata Subrahmanyan Govindarajan, Katherine Atwell, Barea Sinno, Malihe Alikhani, David I. Beaver, Junyi Jessy Li. Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics. 2023.

Multidisciplinary collaboration to develop a digital health solution for early detection of cognitive decline in primary care

Alzheimer s & Dementia · 2022 · 1 citations

• Psychology
• Medicine
• Medical education

Abstract Background Developing digital healthcare solutions that facilitate rapid identification and effective management of patients with Alzheimer disease and related disorders (ADRD) is an area of intense study. A critical juncture for these efforts involves the early detection of cognitive decline. Because primary care providers (PCPs) are the first line of medical care, they are often the first to hear concerns about cognitive decline. However, under‐diagnosis of ADRD in primary care settings is widely recognized, as are the many barriers to routine cognitive screening. While information about brief cognitive screening tools for detecting ADRD is plentiful, PCPs remain uncertain about which patients to assess, which tools to use and how to use them, and how to communicate results. The goal of this project was to design a risk assessment and cognitive screening (RACS) application that specifically addressed the needs and concerns of PCPs to facilitate identification of cognitive decline in primary care settings. Method We employed a multi‐modal assessment approach in designing the RACS app, which first assesses risk for cognitive impairment and then assesses cognitive functioning using a working memory/processing speed task in combination with four speech/language tasks. We assembled a multidisciplinary team with the following expertise to develop and test the app: biostatistics, computational linguistics, computer science, computerized cognitive assessment, gaming/app development, engineering, neurology, neuropsychology, neuroscience, primary care, psychometrics, and speech‐language pathology. Result Programming of app features and pilot testing with people with ADRD was completed in 3 months. Initial development of the connected speech analysis pipeline was completed in 4 months with ongoing testing. Data collection of 50 cognitively normal, 50 mild cognitive impairment, and 50 mild dementia participants is approximately 50% completed within 6 months. Preliminary results based on cognitive performance alone show good ability to discriminate groups. Reduction of speech‐language variables for inclusion in a final cognitive performance score is underway using a variety of machine learning techniques, including the elastic net and random forests. Conclusion The RACS app shows promise as a digital health solution to facilitate early detection of cognitive decline in primary care and may prove useful in other busy clinical settings.

Contrasting Static and Contextualised Embeddings in the use of Semantic Feature Vectors in Neurophysiological Prediction

2021

• Computer Science
• Artificial Intelligence
• Natural Language Processing