Although we're not quite there yet, this is an interesting thought, and well within our technological grasp.
I think the one thing missing from moving this analysis forward is reliably tracking heart attacks.
You'd have to rely on the user to signal that they had a heart attack, so that data analysis could be done leading up to the events. This requires effort by someone that just had a heart attack (not always easy).
Then you also have to trust that your user data is accurate and that your users aren't trying to fake heart attacks by messing with their devices.
I'm still hopeful that these problems will be worked around.
You need a 12 point ECG to even see many arrhythmias. Heart rate monitors aren't useless, but they aren't particularly helpful for a lot of heart conditions. Source: have heart condition.
I think it's relatively clear that the tech works in a controlled setting. It even works surprisingly well in an uncontrolled setting. But as these devices become more mass marketed, sensitivity and selectivity become a much bigger concern. If you detect 99% of all true events, that looks great at first. But how useful is it if only 10% of all detected events are true? Selectivity must be tight before we can issue strong health warnings to masses of people. The challenge of real world noise can be bigger than getting the first principle tech to work.
A recruiter for mc10inc.com reached out to me and told me about their business which is exploring something here. Small wearable electronic stickers that are flexible and waterproof. I suspect that type of tech would even allow the all important 12 point ECG monitoring without the typical discomfort and immobility associated with the current method.
9 comments
[ 2.2 ms ] story [ 32.9 ms ] threadI think the one thing missing from moving this analysis forward is reliably tracking heart attacks.
You'd have to rely on the user to signal that they had a heart attack, so that data analysis could be done leading up to the events. This requires effort by someone that just had a heart attack (not always easy).
Then you also have to trust that your user data is accurate and that your users aren't trying to fake heart attacks by messing with their devices.
I'm still hopeful that these problems will be worked around.
I think it's relatively clear that the tech works in a controlled setting. It even works surprisingly well in an uncontrolled setting. But as these devices become more mass marketed, sensitivity and selectivity become a much bigger concern. If you detect 99% of all true events, that looks great at first. But how useful is it if only 10% of all detected events are true? Selectivity must be tight before we can issue strong health warnings to masses of people. The challenge of real world noise can be bigger than getting the first principle tech to work.