CSE 477 Spring 2014

This is the project page for CSE 477 Spring 2014.

Revised case design

I heavily revised the case design from my first mockup. Now the clip is a separate part, and there's one single large clip rather than two small ones. The battery no longer has an external hole, it'll be inserted below the PCB in the main compartment of the case. Also, we wanted to have a switch that would detect when the clip was being worn, so I made some recesses for metal plates on the back surface of the case and on the clip. We hacked up a coin cell battery clip to get a piece of metal with bent out tabs that will touch the other plate when the device is not being worn. Also, there are cutouts for the wires that will connect these metal contacts to GPIO pins on the PCB.

Meeting notes from May 20th

Kickstarter page:

  • Make a “money shot” image

  • Upload video to youtube/vimeo/whatever, link it on kickstarter page

  • Should be updating kickstarter page whenever we have new weekly updates

  • Make second draft of video

    • Introduction (motivation for product)

    • Show “theory of operation”, hardware, and phone app

    • Shots of prototype (sensortag) sensing tilt

    • Shots of iPhone app prototype

    • Shots of UI/data visualization mockups

3D printing case:

  • Get dimensions of board and battery

  • Design case back that holds components (leave space for battery wires?)

  • Design case front, decide on attachment method (hinge, pins, screws)

PCB:

  • Figure out how to panel board without extra cost

Update: UI design, plans

I sent the previous UI prototype to Dori to ask her opinion on it and to ask for examples of the feedback they currently give to patients. She sent back examples and some feedback on the prototype. She also forwarded it to their group at UCSD, who sent us examples of the graphs that were the most popular among their participants.
We got some good feedback on what elements of the UI were liked, and what parts they didn't think were necessary. A summary:
  • The Details screen included the same numbers as the Summary, which was unnecessary. It should focus on showing the breaks in sitting throughout the day as this is "one of the most important concepts for health benefits".
  • They would like activity (walking/moving/etc) separated from standing time in the analysis.
  • They currently use bar charts to show daily stats on how long a patient spent lying, sitting, standing, and moving, either with the bars separated or with them stacked on each other and distinguished by color.
The UCSD group also mentioned they would like a feature where the application asks patients how they are feeling throughout the day and tracks that as well; however, this would be a stretch to implement in the time we have available.
Based on the feedback, last week I drew a revised paper prototype for the Details and Week View screens. For the Week View I decided to use separate bars for lying, sitting, etc., but we could also stack them into a single bar for each day. This might make the UI less visually noisy, but I feel it could make it a little harder to compare the size of each bar between days.

For the Details screen, I filled in the graph with different shades to represent the different categories. In the finished version, this would be colored. The examples Dori sent us use a "stoplight" color scheme: red for sitting, yellow for standing, and green for moving. The examples from UCSD use the same color scheme, plus purple for sleeping time. We would probably adopt the same color scheme for our application.

For the next week, I plan to CAD up a case that will hold our battery and PCB. I've got some experience with Solidworks and with 3D printing so it should go fairly quickly once I figure out how to use the Dimension printer we've got in the lab. I'll need to figure out the precise dimensions of the battery and the PCB with our components attached, so that I can make a case that's compact but still fits everything. I'm planning to make the exterior gently curved, so that it will fit nicely on the hip, which will make the design somewhat more complex. Also, we'll need it to be in two pieces so we can put the hardware inside it, and the way those pieces will attach is still up in the air. We may resort to using one or two small screws to hold it together.