My research is broadly in the area of psychophysiology - I believe
that engineering and technology now has come to a point, where
wearable and mobile monitoring (together with advanced data analysis)
could provide an insight into the cognitive and emotional processing.
This is especially important in the context of mental disorders, where
diagnostics currently heavily depend on clinical judgement.
My PhD work covers three main areas, starting from the definition of
physiological features, correlated to selected symptoms of mental
disorders, including the development of instrumentation for
acquisition of multichannel psychophysiological information and
oncluding with creation of a framework for automated symptoms scoring.
My background is mostly in the area of embedded software engineering and
I worked on almost all modern hardware platforms - from 8 to 64 bit. But
it is such a good feeling to create something with your hands, that I've
drifted a bit towards the dark side - hardware design ;)
The big concept of this device is that modern pulse oximeters are insanely
expensive and absolutely not flexible. But if one could connect a pulseox
to a computer or a mobile phone - we would get a hospital grade monitoring
system for very little money.
This device is pretty simple (however still unfinished) - it includes Freescale
MCF51MM ColdFire MCU with integrated USB (device), TriAmp and OpAmps as well as
16-bit ADC and connects to Nellcor-compatible probe. When you connect it to a
computer (or mobile phone) - it pretends to be a serial
port and simply streams the Red-IR-Off ADC readings in CSV format. Easy.
Mental signs tracker
This device is still at a very early stage and when finished, will become a
wearable probe for acquiring pulse oximetry, electrodermal activity, temperature,
acceleration and streaming it to a mobile phone using Bluetooth Low Energy.
The basic principles are the same as with the PulseOx, however the device includes
ARM® Cortex™-M4 based Kinetis K50 (basically the same set of peripherals as
MCF51MM, but smaller package size).
The board you can see is a preliminary design for hardware/software
development, but soon, soon it will become a real thing ;)
You may have noticed, that my hardware projects proudly include "mobile" in
the names? In practice it means that there shall be a mobile application to
get the data from device and process it in a smart way. So far there is just
one and not completely related ;)
Actopsy is a little app for collecting your activity data from a mobile
phone. It is part of my PhD work at the University of Oxford and is used
to crowdsource the data for my research, so I could analyse it and extract
features, describing the behaviour patterns and regularity.
I'll share all the interesting results with the next versions of the app -
so don't miss the updates! Install the app, give it a few days (or better
a week) to collect your data and check your weekly activity profile. And
tell your friends to install it also!
By the way, Actopsy is on Google Play ;)
Activity analysis toolkit
Quite important area of my PhD work is data analysis ;) The Activity Analysis
Toolkit is a continuously developed project, where I accumulate all the Matlab
algorithms I implement for analysis and visualization of my data.