I wrote a website that helped my wife's grandfather manage his diabetes. It was essentially captured that moment in time his blood sugar, blood pressure, and weight. This was 10 years ago, and his doctor loved it because he could generate a report and bring it to his visits.
What you're doing here is far beyond what I did and it's stuff like this that really make me white-pilled on the future. I know you said that the model is worse than you, but just think about how far stuff like this has come and what it might look like in the future.
Yeah all models are weirdly much worse at predicting hypos than hypers. I'd say the models are better than me when it comes to predicting hypers. Could be a data leak in my brain, but I'm not sure. There are neurons in the brain that sense falling blood sugar [1], so it is likely that I have access to some info that these models don't, albeit unconsciously.
I wrote a website that helped my wife's grandfather manage his diabetes. It was essentially captured that moment in time his blood sugar, blood pressure, and weight. This was 10 years ago, and his doctor loved it because he could generate a report and bring it to his visits.
What you're doing here is far beyond what I did and it's stuff like this that really make me white-pilled on the future. I know you said that the model is worse than you, but just think about how far stuff like this has come and what it might look like in the future.
Yeah all models are weirdly much worse at predicting hypos than hypers. I'd say the models are better than me when it comes to predicting hypers. Could be a data leak in my brain, but I'm not sure. There are neurons in the brain that sense falling blood sugar [1], so it is likely that I have access to some info that these models don't, albeit unconsciously.
[1] https://pmc.ncbi.nlm.nih.gov/articles/PMC10450823/