movement through the use of motion-sensing accelerometers and wireless technology.
The first application that is being developed measures the performance of the wearer using an accelerometer while simultaneously providing visual, real-time feedback through a 3D interface that is similar to popular video games (think Nintendo Wii). This application will have the ability to quantify patient movement patterns and qualify their physical adherence to baseline expectations.
If I get a rough idea of price (within 10-20%) that would be enough to get us started. We have a demo application that needs the following items improved:
1. The object that moves in time with the person's body needs to be orientated to the model example on the screen, in size, timing, and space, also sequencing and coordination (ie fluency of movements against the model)
2. The 'syncrate' needs to be computed - there is a space for it on the screen, but it is currently not being changed (just sits at 100%).This number is based on the synchronization of the person's movement and the example. For this first version, all we need is a hardcoded algorithm based on timing (have the person click a button or something like that, and count down from 3 to 1, then start analyzing numbers from the accelerometer compared to some hard-coded values
3. The numbers from the accelerometer need to be stored in a database (Access should work, or SQL Server 2005 Express)
4. These numbers need to be associated either with the person's user name or full name, so they can continue from previous session, and some kind of a scoreboard would be helpful
5. The collection of data from the accelerometer needs to be optimized - we probably don't need to collect every line coming from it, for example. This will take some trial and error.
6. The user (or physical therapist) needs to be able to adjust the angle at which the person should repeat the exercise to - for instance maybe the current example goes from 0 to 45 degrees, they need to be able to adjust it to 40. That may involve, for this version, recording a set of videos - in 1 degree increments, and swapping based on choice. Same with a number of reps, this should be variable.