GARCIA adolfo Martin
congresos y reuniones científicas
Natural speech markers of neurodegenerative diseases in Latino populations
Nueva York
Seminario; New York Neuropsychology Group Lecture Series; 2021
Institución organizadora:
New York Neuropsychology Group
Neurodegenerative diseases are growing disproportionately in Latin America, calling for innovative approaches for patient identification and severity estimation. Mainstream diagnostic, prognostic, and monitoring procedures rely on standardized clinical/cognitive tests and neuroimaging. Though irreplaceable, these approaches prove stressful, fatiguing, bound to clinician bias, and unrepresentative of everyday behavior. Also, they are often unaffordable, unfeasible, and non-viable for remote application. Aiming to circumvent these limitations, I will introduce a novel framework rooted in ecological language tasks. First, I will present studies based on automated speech assessments, showing that brief spontaneous monologues offer rich information to identify patients and predict symptom severity. Second, I will describe findings from a naturalistic text paradigm capturing disease-specific comprehension deficits and underlying neurocognitive abnormalities. The evidence obtained so far spans diverse neurodegenerative diseases (Parkinson?s disease, Alzheimer?s disease, frontotemporal dementia), data collection methods (automated acoustic and textual analyses, behavioral measures, structural and functional neuroimaging, high-density electroencephalography), and statistical approaches (inferential statistics, machine learning pipelines). Overall, this approach is typified by minimal stress and fatigue, objective and consistent outcomes, high ecological validity, negligible costs, massive scalability, and direct adaptability for remote implementation. Existing findings and ongoing extensions of the framework represent a promising complement for standard approaches in clinical settings.