Simple real-time cloud-connected actigraphy solution for monitoring patient activity using mobile phones

Abstract

Listen

Translate article

Rehabilitation patient activity is not typically monitored in real-time due to lack of suitable actigraphy devices. We present an open-source actigraphy system with integrated real-time cloud data storage based on mobile phones for potential use in patient activity monitoring and quantitation. An iPhone program was created to stream an actigraphy score to a cloud data storage service, Adafruit IO, in real-time. The actigraphy score is calculated as vector magnitude from three-axis accelerometer data at 2.5 second epochs. Actigraphy data was successfully streamed to the cloud storage and visualization platform at up to 30 Hertz, with 60 Hertz possible at a higher cost tier. Data was represented as decimal with two significant digits of precision. No movement at all is normalized to 0. Maximum vector magnitude with moderate walking was 0.39. Maximum vector magnitude with jogging was 1.02 Maximum vector magnitude with the phone in the hand during a throwing motion was 5.47. iOS reported energy draw was low (level 1 out of 20). In summary, we present an open-source design for a simple and inexpensive real-time cloud-connected actigraphy system for rehabilitation patients. Alternative platforms and devices are discussed, as are security and cost implications The widespread use of mobile devices and the iOS platform, with optional integration of global positioning coordinates and other phone data, are advantageous compared to purpose-built existing actigraphy solutions.