Advancing biomedical discovery and improving health through mobile sensor Big Data


Software Platform | User Guides | Report a Bug | Forum

Download Software Here


An Open Source Software Suite for Mobile Sensor Data

Learn more about mCerebrum from this webinar with Chief Software Architect Timothy W. Hnat and Lead Software Engineer Syed ‘Monowar’ Hossain.

brainAndtext mCerebrum is a configurable software platform for mobile and wearable sensors. It provides support for reliable data collection from mobile and wearable sensors, and real-time processing of these data for sensor triggered just-in-time adaptive interventions. All software is released under the open source BSD 2-clause license. Dates in parentheses are projected release dates for specific capabilities.Platform Capabilities
Data Acquisition — Data is recorded from a variety of sensor platforms including the AutoSense chestband, Microsoft Band 2, and smartphone sensors. Additionally, information from a participant through an Ecological Momentary Assessment (EMA) can be configured and recorded. Data acquisition can be temporarily disabled through the privacy controls.

  • Sensor Platforms
    • AutoSense Chestband (January 2016) includes a 3-axis accelerometer, a respiratory inductance plethysmography (RIP) sensor, an electrocardiogram (ECG), a galvanic skin response (GSR) sensor, and a skin temperature sensor.  [Source code]
    • Microsoft Band (December 2015) includes optical heart rate, 3-axis accelerometer/gyro, ambient light, skin temperature, UV sensor, capacitive sensor, galvanic skin response, and barometer. [Source Code]
    • Smartphone (December 2015) includes GPS, accelerometer, gyrometer, and ambient light sensors.  [Source Code]
  • Participant Interaction
    • Ecological Momentary Assessment (EMA) (January 2016) provides for the delivery of questions to a participant in the field.  [Source Code]
    • Self-Report (January 2016) allows a participant to record information about activities or behaviors

User Engagement and Control
Participants can be engaged through a number of different mechanisms that are controlled by the Study Manager and include custom intervention applications.

  • Study Manager (January 2016) is responsible for running the field studies and determining when to deliver interventions [Source Code]
  • Interventions through various applications including Mood Surfing [Source Code], Thought Shakeup [Source Code], and other external applications

Sensors to Markers — Algorithms based on published research have been adapted to mCerebrum and generate markers for stress assessment, smoking and puff detection, conversations, and eating behavior detection.

  • Stress – cStress: Towards a Gold Standard for Continuous Stress Assessment in the Mobile Environment (2015)
  • Smoking – puffMarker: A Multi-sensor Approach for Pinpointing the Timing of First Lapse in Smoking Cessation (2015)
  • Conversation – mConverse: inferring conversation episodes from respiratory measurements collected in the field (2011)
  • Eating – A Practical Approach for Recognizing Eating Moments with Wrist-mounted Inertial Sensing (2015)

Data Management — Data is securely stored on the smartphone before being transferred to a cloud service or exported to a local computer. Data is also evaluated for quality to improve data collection from participants.

  • Data storage and transfer: Data is stored locally on the phone’s encrypted storage and a transfer mechanism allows the system to export information to the cloud.   [Source Code]
  • Data quality is actively measured and prompts can be provided to participants to correct incorrectly attached sensors.
  • Real-time data visualization allows most signals within the mCerebrum system to be graphically represented on the phone.

Programming Interfaces

  • DataKit API (December 2015): DataKitAPI for Android contains a set of APIs by which applications can communicate with DataKit on the smartphone. This API module is the only module necessary to augment an Android application with DataKit support. [Source Code]
  • Utilities (December 2015): Utilities contains common routines, which can be used by other Android applications within the mCerebrum ecosystem. [Source Code]

Cerebral Cortex

Enabling Mobile Sensor Data Collection, Analysis, Interpretation, and Intervention

cerebralCortex-logo-MD2K Cerebral Cortex is the big data companion of mCerebrum designed to support population-scale data analysis, visualization, model development, and intervention design for mobile sensor data. It provides the ability to do machine learning model development on population scale data sets and provides interoperable interfaces for aggregation of diverse data sources.  All software is released under the open source BSD 2-clause license. [Source Code]


Data Management

  • Hadoop and HDFS (January 2016)
  • Open mHealth interface
  • Data import and export in common formats (April 2016)

Data Processing and Analytics based on Apache Spark

  • Stress (April 2016)
  • Smoking (April 2016)
  • Conversation
  • Eating

Model Development

  • Scalable machine learning infrastructure
  • Iterative data analysis and model generation

Data Visualization

  • Discovery dashboard
  • Data science visualization (January 2016)
  • Health science visualization
  • Population-scale data


  • Containerized architecture based on Docker (April 2016)