Tools and Algorithms - BHQ - Behavioral Quantification
BHQ – Behavioral Quantification - a mobile phone application
A passive data collection of mobile phone usage, e.g. calls, location, accelerometer, Bluetooth, app browsing, etc. The information is anonymized and analyzed individually to 'learn' a patient's routine and detect deviations. Therefore, quantitative analysis of behavior changes, cognition deterioration and mental health relapse can be predicated. The application is already in use in differetn clinical studies. The app is available for installment in mobile phones of volunteers or beta-users.
For more information:
Prof. Uri Nevo - nevouri@tauex.tau.ac.il
Ella Wilcyznski - ellaalon@mail.tau.ac.il
Data collection
The app is operated at the background of the Android operation system of mobile phones, without any user active interaction. All data collected is irreversibly encrypted, such that the data do not include content of use, e.g. the content of a text message is not collected, only the number and time of text messages sent.
The data is collected from multiple sensors available in the device: battery charge, screen light, accelerometer, Bluetooth, calls, gyroscope, location, step counter, etc. It also collects information about the category of application used: social media, news, navigation, browsing, games and more. The data is transmitted to a server collecting the information and time of operation.
Data analysis
The collected data can be analyzed in different manners to accommodate multiple behavior or cognitive questions. It can determine if a subject is asleep or a awake, the quality of sleep, activity inside and outside the house, physical workout, his/her social interactions. The analysis is based on establishing a baseline of behavior for every person after collecting data for 2-3 weeks. Compared to this baseline, it is possible to detect changes in behavior.
The quantitative analysis could be useful in different conditions of mental health deterioration, e.g. reduction of social interactions or less activities outside the house in cases of depression. Cases of patients with reduction of their cognitive state can be detected with impaired/changes in sleep patterns, slower and awkward movement, and less overall use of the mobile phone.