Digital Biomarkers Correlate with Traditional Neurocognitive Assessments
Digital phenotyping can use a range of different signals from a smartphone. Mindstrong has focused on human-computer interactions such as taps, scrolls, clicks and other activities on the smartphone touchscreen. Our science is based on these measures alone, which are completely passive and do not rely on personal information (unlike location or speech data).
Digital biomarkers and cognitive traits. Volunteers (n=27) compared on neurocognitive tests and digital biomarkers. Correlations across multiple cognitive trait measures = 0.7-0.8 (approximately the test-retest variance of the traditional tests). Source: Dagum, Digital Medicine, 2018.
Biomarker Loading on Depression Constructs †
To identify the digital phenotyping features that could be clinically useful, Mindstrong used powerful machine learning methods to show that specific digital features correlate with cognitive function, clinical symptoms, and measures of brain activity in a range of clinical studies.
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-smartphone interactions that correlate highly with select cognitive measures, mood state, and brain connectivity. These interactions were comprised of taps, swipes and other touchscreen activities which were completely content-free and not reliant on capturing personal information.
Biomarkers Load on Mood Constructs
|PAN_P||PANAS Positive||Positive Affect|
|PAN_N||PANAS Negative||Negative Affect|
|PHQ 9||Patient Health Questionnaire||Depression|
|SHAPS||Snaith-Hamilton Pleasure Scale||Anhedonia|
|CGIS||Clinician Global Impression – Severity||Global pyschiatric functioning|
|HAMA||Hamiliton Anxiety Scale||Anxiety|
|HAMD||Hamiliton Depression Scale||Depression|
FIGURE. Correlations between the top Mindstrong Biomarkers and Clinical measures. The p-values from all correlations are corrected to limit the false discovery rate to 1%. Correlations with a p-value below this threshold are shown in gray.
This publication describes the study used to identify digital biomarkers that predict performance on gold standard neurocognitive tests.
RELATED CONTENT: Behind the Paper. This accompanying article describes the story behind Mindstrong’s technology, and the journey Paul Dagum, Founder and CEO, took to develop the pioneering concept and work.
This poster reports preliminary results from a Phase 2A trial of BTRX-246040, a potent and selective antagonist of the human nociception receptor (NOPR), as a treatment for anhedonia in patients with major depressive disorder. Mindstrong’s digital phenotyping is one of the outcome measures being used to assess efficacy.
This Commentary describes the changing ecosystem as small and large tech companies become involved in biomedical research and healthcare.