Predicting Mood Disturbance Severity with Mobile Phone Keystroke Metadata: The BiAffect Digital Phenotyping Study
The latest of three recent academic papers published by the BiAffect team, this has been submitted to the Journal of Medical Internet Research.
Mood disorders are common and associated with significant morbidity and mortality. Better tools are needed for their diagnosis and treatment. Deeper phenotypic understanding of these disorders is integral to development of these tools. This study is the first effort to use passively collected, mobile phone keyboard activity to build deep digital phenotypes.
To demonstrate the feasibility of using passively collected keyboard dynamic metadata to infer mood states.
Within subject design of eight weeks, subjects were provided a mobile phone loaded with a customized keyboard which passively collected keystroke metadata. Subjects were administered the Hamilton Depression Rating Scale (HDRS) and Young Mania Rating Scale (YMRS) weekly. Linear mixed-effects models were created to predict HDRS and YMRS scores (total keystrokes n= 626,641, weekly average 9,791 ± 7,861; total accelerometer readings n= 6,660,890, weekly average 104,076 ± 68,912).
A statistically significant mixed-effects regression model for the prediction of HDRS-17 item scores was created: conditional R²=0.63, P=0.014. A mixed-effects regression model for YMRS scores showed the variance accounted for by random effect was zero, and so a simple linear regression model was created: R²=0.34, P=0.0011. Multiple significant variables were demonstrated for each measure.
Mood states in bipolar disorder are correlated with specific changes in mobile phone usage. The creation of these models provides evidence for the feasibility of using passively collected keyboard metadata to detect and monitor mood disturbances.
Sequential Keystroke Behavioral Biometrics for Mobile User Identification via Multi-view Deep Learning
The second of three recent academic papers published by the BiAffect team, this was presented at the 2017 Joint European Conference on Machine Learning and Knowledge Discovery in Databases.
The first of three recent academic papers published by the BiAffect team, this was presented at the 23rd ACM SIGKDD Conference on Knowledge Discovery and Data Mining.
The increasing use of electronic forms of communication presents new opportunities in the study of mental health, including the ability to investigate the manifestations of psychiatric diseases unobtrusively and in the setting of patients’ daily lives. A pilot study to explore the possible connections between bipolar affective disorder and mobile phone usage was conducted. In this study, participants were provided a mobile phone to use as their primary phone. This phone was loaded with a custom keyboard that collected metadata consisting of keypress entry time and accelerometer movement. Individual character data with the exceptions of the backspace key and space bar were not collected due to privacy concerns. We propose an end-to-end deep architecture based on late fusion, named DeepMood, to model the multi-view metadata for the prediction of mood scores. Experimental results show that 90.31% prediction accuracy on the depression score can be achieved based on session-level mobile phone typing dynamics which is typically less than one minute. It demonstrates the feasibility of using mobile phone metadata to infer mood disturbance and severity.