Mindstrong’s science begins with the smartphone. With over three billion smartphones globally and more than 75% of American adults owning a smartphone (92% between ages 18 – 29), this powerful mobile computer has become ubiquitous. Digital phenotyping collects data from the smartphone to provide measures of emotion, cognition, and behavior. In three rigorous clinical studies with over 200 person-years of data, Mindstrong has used powerful machine learning analytics to show that specific digital features correlate with cognitive function, clinical symptoms, and measures of brain activity.
Digital phenotyping uses three kinds of signals from a smartphone. Sensors measure activity, location, and social meta-data (e.g. number of messages out vs messages in). Human-computer interactions capture keyboard performance including typing and clicking. Voice and speech data analyzed with natural language processing can yield insights into emotion and cognitive coherence. Together these signals contribute to the picture of mood, cognition, and behavior we call the digital phenotype.
Individual Biomarkers Load on Distinct Cognitive Domains †
To identify the digital phenotyping features that could be clinically useful, research volunteers completed extensive neuropsychological testing, clinical assessments of mood and cognition, and, in some cases, neuroimaging with fMRI. The results revealed a set of digital biomarkers from human-computer interaction (keyboard performance) that correlate highly with select cognitive measures and brain connectivity.
Individual Biomarkers Load on Distinct Neural Circuits †
In those research subjects with depression who volunteered for fMRI brain scans (n = 100), specific digital biomarkers from human-computer interaction that correlated with neuropsychological measures of cognitive control and reward also correlated highly with activity in those brain areas implicated in cognitive control and reward.
While single markers correlated highly with cognitive and neural features, combining markers yielded even higher correlations. In one study with 100 patients with depression, combining biomarkers yielded high correlations with the PHQ-9, a gold standard measure of depression.
Biomarker Loading on Depression Constructs †
† Mindstrong biomarker platform was used by the National Institutes of Health grant UH2HL132368 “Engaging self-regulation targets to understand the mechanisms of behavior change and improve mood and weight outcomes”.
Mindstrong Health is participating in several trials selected to test the value of digital phenotyping in clinical populations.
Mindstrong has developed and patented a biomarker panel that measures brain function from interaction patterns captured passively and continuously from human-computer interfaces found in ubiquitous mobile technology.