I-CubeX Software Usability and Application (ISUA)
| Participants: |
Mahtab Ghamsari-Esfahani Aaron Krajeski |
Bertrand Scherrer Marcelo M. Wanderley (supervisor) |
|
|---|---|---|---|
| Collaborators: | Axel Mulder, Infusion Systems, Ltd. |
|
|
| Funding: | MITACS Accelerate Internships | ||
| Project Type: | Industrial Partnership | ||
| Time Period: | Oct. 2011 - Aug. 2014. (complete.) | ||
Project Description
Block 1 (Mahtab)
Block 2 (Aaron)
Block 3,4 (Bertrand)
This block was devoted to the development of software tools in Max/MSP to be used with the sensors carried by Infusion Systems, Ltd. At the end of this block Max/MSP standalone applications were released for the BioVolt, BioWave, BioBeat, ReachClose and ReachFar sensors.
To do this, an inventory of Infusion Systems sensors and corresponding usage scenarios was first created. This exercise proved central in the determination of missing signal processing abstractions in Max. Those abstractions were then created most of the time stemming from already existing Digital Orchestra Toolbox objects.
In the next phase of the project, the linearization of the response of the ReachClose sensor was considered. To that end a Matlab standalone was created that analyzes the response of the sensor to calibrated stimuli in order to determine various ranges of operations and for each range, determine a mathematical fit that can then be used to improve the reliability of the measurements output by the sensor. A Max/MSP external iCalibrate was developed in order to make it possible to use the response fits found using the Matlab application in real-time, in Max.
Several applications focusing on the biosensors carried by Infusion Systems were developed. For example, the MuscleTrainer application, implements a form of bio-feedback where users are asked to contract a muscle for a predefined period of time and the rest. This application can be adapted for more complex biofeedback scenarios using EMG signals.
The BioBeat application computes the heart rate based on the ECG signal. This program required the development of iFilter, a Max/MSP external allowing to design digital filters. One of the key aspects of iFilter is that it can adapt the coefficients of the filter if the sampling rate used for data acquisition changes. This object is central in the context of biosignal sensor applications as EEG, EOG, ECG and EMG signals have very particular (and different) frequency distribution characteristics.
A prototype for an application detecting the direction of eye movement was also realized and will lead to a released application in the near future.
Other contributions include the development of a reliable mechanism for saving and recalling settings used in Max/MSP standalones with ICubeX software, the systematic adoption of source version control to ease development, and the detection of jabs with the GForce3D-6 sensor.