Sorry, best I can do is an app that tells you if the person you are talking with on the phone is fat. Great for 900 numbers and such. It can even chime in and say "Fat person detected" on the call in a robot voice.
"Scientists at Klick Applied Sciences in Canada, working with faculty at Ontario Tech University in Canada, trained the AI using 267 voice recordings from people living in India.
Roughly 72 per cent of the participants (79 women and 113 men) had already been diagnosed as nondiabetic. The other participants (18 women and 57 men) had been diagnosed with type 2 diabetes.
All participants recorded a phrase six times per day for two weeks, resulting in a total of 18,000 recordings. The scientists then pinpointed 14 acoustic differences between those with and without type 2 diabetes."
True, but accuracy tells us something, and an accuracy of 0.75±0.22 tells us that what we know is very close to “we have reasonable confidence that it is at least very slightly better than a coin flip”.
If 75% have diabetes (or don't have diabetes -- either way), then 75% accuracy is also the accuracy of a naive coin flip.
That is to say, don't compare to the accuracy of an even coin flip; compare to the accuracy of a coin flip that flips according to the proportion of the people with the attribute you care about.
It's literally just identifying overweight people and then leaning into the correlation between weight and obesity. If you're obese, chances are you either have diabetes or will soon.
Off topic, but not sure if this level of audio classification task can be ever done with few-shot prompting a multi-modal model. Don't think GPT4V will be there yet, but food for thought.
Note that the “optimal model” includes age and BMI.
The predictive model seems to be logistic regression and Naive Bayes. There’s no fancy AI here. Just some basic feature summary.
I don’t have time to go into it too detailed but I’m noticing they’re saying some vocal features are P <0.001 for the matched datasets even though the values across the two classes seem awfully similar. I’m not sure what’s going on there.
I _quickly_ checked the study [1] and didn't see anything obvious.
Given that our voices reveal so much about the state of a human (think of mood, emotions), I think it's at least possible that voices could reveal something about metabolism. Just we cannot hear it because hearing diabetes is not as useful as hearing something about the mood of another person.
I didn't see anything in the study (that I also skimmed tbh) that is clearly wrong, which makes me dismiss the results.
It's just because I don't think this is evidence enough to change my priors. i.e. My gut tells me this is wrong and I don't buy it.
The sample is 267 people, tiny enough that I'd expect an analysis to be done with a linear model with few features.
They used 14 features, but had a pipeline to select "model (out of 3), feature set, and threshold for prediction".
There's _so_ many degrees of freedom there that _by default_ I assume there's leakage.
I'd love to see this paper replicated! It would be amazing if this were true.
But if I had to bet on this, I wouldn't give more than 20% chance of being true.
Readit News
Fat affects vocal chords, and this can be detected in the vocal patterns.
This whole "AI Tool" could be replaced with a simple question:
"Are you fat?"
But people often aren't honest with themselves about this topic.
Roughly 72 per cent of the participants (79 women and 113 men) had already been diagnosed as nondiabetic. The other participants (18 women and 57 men) had been diagnosed with type 2 diabetes.
All participants recorded a phrase six times per day for two weeks, resulting in a total of 18,000 recordings. The scientists then pinpointed 14 acoustic differences between those with and without type 2 diabetes."
For a clinician, sensitivity and specificity are much more useful. It's too bad they didn't publish these.
That is to say, don't compare to the accuracy of an even coin flip; compare to the accuracy of a coin flip that flips according to the proportion of the people with the attribute you care about.
"Arrest the witch!!"
AI to Diagnose Diabetes Based on Seconds-long Snippet of Voice:
"Uh. Interesting ...!"
Note that the “optimal model” includes age and BMI.
The predictive model seems to be logistic regression and Naive Bayes. There’s no fancy AI here. Just some basic feature summary.
I don’t have time to go into it too detailed but I’m noticing they’re saying some vocal features are P <0.001 for the matched datasets even though the values across the two classes seem awfully similar. I’m not sure what’s going on there.
I _quickly_ checked the study [1] and didn't see anything obvious.
Given that our voices reveal so much about the state of a human (think of mood, emotions), I think it's at least possible that voices could reveal something about metabolism. Just we cannot hear it because hearing diabetes is not as useful as hearing something about the mood of another person.
[1] https://www.mcpdigitalhealth.org/article/S2949-7612(23)00073...
It's just because I don't think this is evidence enough to change my priors. i.e. My gut tells me this is wrong and I don't buy it.
The sample is 267 people, tiny enough that I'd expect an analysis to be done with a linear model with few features. They used 14 features, but had a pipeline to select "model (out of 3), feature set, and threshold for prediction".
There's _so_ many degrees of freedom there that _by default_ I assume there's leakage.
I'd love to see this paper replicated! It would be amazing if this were true. But if I had to bet on this, I wouldn't give more than 20% chance of being true.