Timezone: »

 
Emily Fox
Emily Fox

Fri Dec 13 01:30 PM -- 02:00 PM (PST) @
in Machine Learning for Health (ML4H): What makes machine learning in medicine different? »

Models of Cognition: From Predicting Cognitive Impairment to the Brain Networks underlying Complex Cognitive Processes Talk

The ubiquity of smartphone usage in many people’s lives make it a rich source of information about a person’s mental and cognitive state. In this talk, we first consider how such data sources can be used to provide insights into an individual’s potential cognitive impairment.  Based on a study enriched with subjects diagnosed with mild cognitive impairment or Alzheimer's disease, we develop structured models of users’ smartphone interactions to reveal differences in phone usage patterns between people with and without cognitive impairment. In particular, we focus on inferring specific types of phone usage sessions that are predictive of cognitive impairment. Our model achieves state-of-the-art results when discriminating between healthy and symptomatic subjects, and its interpretability enables novel insights into which aspects of phone usage strongly relate with cognitive health in our dataset.

We then turn to a scientific analysis of brain functioning underlying cognitive behaviors. Recent neuroimaging modalities, such as magnetoencephalography (MEG), provide rich descriptions of brain activity over time enabling new studies of the neural underpinnings of complex cognitive processes. We focus on inferring the functional connectivity of auditory attention using MEG recordings.  We explore notions of undirected, contemporaneous interactions using sparse and interpretable deep generative models, as well as time-varying directed interactions using Bayesian dynamical models.

Author Information

Emily Fox (University of Washington, Apple)

More from the Same Authors