I have been self-tracking for a while mostly using the old school pen and paper approach as well as different apps for different purposes. I have run into a few challenges - firstly tracking is too much effort even with apps and whatnot (apps crash, some don’t log all data I need, and so on). The biggest challenge though is that I believe real insight lies in combining data from several sources - for instance, what I eat, how much I sleep and the amount and type of workout could affect my mood and working memory the following day. The other challenge is that every new hypothesis needs a fresh data collection campaign making testing it a serial and hence slow process. The serial process has other chellenges like seasons changing and so on. I’d love to just have data to analyze and test for multiple hypothesis. Keeping all this in mind and the fact that I am tech savy, I have decided to build my own app and tools. I want it to be cloud based, work with limited connectivity (batch uploads), gather information from different sources(more on this below), be as “hands free” as possible and be able to run analytics. In anticipate huge amount of data to be collected. I think I can figure out most stuff except for the hands free part. Here’s the main challenge I am facing - finding devices that have bluetooth interface, which will allow me to route raw data as I want. It’s hard to find an arbitrary device having bluetooth and even the ones that do have them route data to their proprietary apps and produce dashboards making it hard for me to correlate it with other sources.
So here’s what I am looking for: (1) is there a way to hack existing hardware and attach Bluetooth modules to it? Of course, I’ll only be looking at digital devices as analog is too much work or perhaps outright impossible.
(2) Are there any open source frameworks out there I could use to build something like this?
(3) Are there ways to get the raw data and automatically route them to the cloud from devices that do have an app?
Here’s the range of things I wish to measure and aggregate: gps (for running distance and location), handheld counter (for focus and distraction), food scale (for tracking diet), EEG (measuring meditation and sleep quality), smartphone app (self report mood from an options menu). a smartwatch (which buzzes and prompts me to make measurements in the first place)
I know this could be a pie in the sky dream but I anticipate self-measuring for a long time to come and anticipate greatly benefiting from this while amortizing the development effort.
What are some of the hypotheses you’d like to test?
What data do you need, but can you not get with existing apps and tools?
For analysis, there’s definitely “some assembly required”
@ejain, There are a few but I’ll mention this since it has most of the challenges I expect to face
The relationship between the way I stretch (how long and how hard), the amount of water I drink, amount of time spent experiencing stress to the amount of back pain and muscle stiffness I feel during the day.
I want to measure the amount of time I take to hit the peak of my stretch, amount of time I hold there using a finger timer/ counter. The amount of water either using a smartwatch or smartphone app, preferably a watch. I want to measure sleep using a simple smartphone app but am figuring a way to account for the lead time before I doze off as well as the numerous wake ups I have at night. Perhaps an EEG might help but I am not sure. I would want to record the amount of time I am in deep sleep state. Finally, I would want my smartwatch to buzz at random times during the day to remind me to perform simple stretches to gauge the pain level as well as measure my movement in the preceding hour or so. I would want all this to work without assuming wifi connectivity so a good way would be for me to have everything route data to my smartphone via bluetooth and upload the same to cloud. I wish to have a web interface that’ll allow me to run correlations between the pain levels from my stretch to all the other variables I am tracking, ideally via a GUI and a few clicks.
The reason for this level of sophistication is that I have been self measuring in one way or other for the last few years and the amount of barrier involved in making pen and paper measurements requires tremendous self discipline in itself and makes it harder for me to focus on the interventions I am trying to test. This happens to the extent that I either stop the intervention or tracking or both. The other reason is that I have this chronic back pain and the doctors have ruled out any injury or anything major so I believe it’s a simple case of stretching myself. But if tomorrow I come across a theory which connects muscle stiffness to water intake (or a combination of sleep and water intake), then I don’t want to go back and repeat experiments to get the data I didn’t collect in the first place. Of course there is a limit to what and how much I can measure at once and I intend to draw that line on a case by case basis. But I at least want to have as much technology support as possible.
I am fairly confident of finding tools and apps that measure what I mentioned above. I am looking for something that will seamlessly work together and allow me to automate data aggregation into one place.
I think your ROI would be better using existing devices that already have Bluetooth built in.
You can measure water consumption using a bluetooth kitchen scale as a cup holder.
A nine axis Bluetooth IMU would give you body motion for later analysis.
You can also use the sensors on your phone or smart watch (if you wear it all the time) to give you data for analysis.
Regarding the water: I’ve been very impressed with the Hidrate water bottle that tracks intake down to fractions of an ounce. I’m not affiliated with them, just a fan on their device. Unfortunately, doesn’t have an easy way to download the data other than through Apple Health (maybe Google Fit? Haven’t tried it).