Dr. Alexandra König, BSc MSc PhD, Institut national de recherche en informatique et en automatique (INRIA); Cobtek (Cognition; Behaviour; Technology) Lab; University Côte d'Azur, France
- Title: Novel Digital Speech Biomarker for Early Detection of Alzheimer's Disease
- Abstract: Future clinical trials targeting Alzheimer's disease (AD) on new disease modifying drugs necessitate a paradigm shift towards early identification of individuals at risk. Emerging evidence indicates that subtle alterations in language and speech characteristics may manifest concurrently with the progression of neurodegenerative disorders like AD. These changes manifest as discernible variations, assessable through semantic nuances, word choices, sentiment, grammar usage (linguistic features), and phonetic/acoustic traits (paralinguistic features). Consequently, automated analysis of speech performance stands as a promising avenue for detecting AD, enabling widespread screening of diverse at-risk populations. The talk will outline applications of novel digital speech biomarkers for measuring cognition (SB-C ) alongside its analytical and clinical validation. The SB-C algorithm demonstrates robustness in detecting Mild Cognitive Impairment (MCI) across various cohorts and languages. In addition, speech and language markers have shown to be useful to assess objectively common neuropsychiatric symptoms in MCI such as depression or apathy. This innovation holds the potential to enhance the efficiency of ongoing trials and augment future primary healthcare practices concerning AD. Identifying subtle cognitive and as well as affective changes through speech analysis signifies a critical advancement in the pursuit of early AD detection, potentially transforming the landscape of AD research and clinical interventions.
Prof. Maria Liakata, EPSRC/UKRI Turing Institute AI fellow, Queen Mary University of London, UK
- Title: Longitudinal language processing for dementia
- Abstract: While the advent of Large Language Modes (LLMs) has brought great promise to the field of AI there are many unresolved challenges especially around appropriate generation, temporal robustness, temporal and other reasoning and privacy concerns especially when working with sensitive content such as mental health data. The programme of work I have been leading consists in three core research directions: (1) data representation and generation (2) methods for personalised longitudinal models and temporal understanding (3) evaluation in real-world settings, with a focus on mental health. I will give an overview of work within my group on these topics and focus on work on longitudinal monitoring for dementia.