Will Large Language Models like chatGPT transform healthcare?
They hold tremendous potential to enhance clinician capabilities especially around communication, but challenges remain
https://t.co/YpCi7zRZH0.
🧵 In 2019, DeepMind published a paper in Nature describing models to predict AKI in US Veterans, finding AUCs of 92% for RNN & 89% for GBDT, better than prior work.
Inspired, we built & tested the generalizability of similar models in & out of VA.
Link: https://t.co/NAXT9zJjjf
The opportunity for AI to improve health outcomes is massive. But what’s been missing is rigorous peer-reviewed studies showing AI used in practice. Today we published 3 studies in Nature Medicine that show for the first time ever how AI saves lives.
https://t.co/yHaeFhhFvQ
Does that mean Intermittent Fasting isn't really as great as its cracked up to be?
No, it does have its benefits.
For one, it is much easier to follow than CR. It is also likely to be a more sustainable life-style choice.
2/2
Calorie Restriction (CR) vs Time Restricted Eating (aka Intermittent Fasting)?
Is one better over the other? Not really. Findings from a recent NEJM study. Both eating plans resulted in about 6-8 kg weight loss in 12 months. https://t.co/BewJWxn9e9
1/2
Yet another Nutri Epi study. Apparently this was one of the largest.
Not sure what to make of them any more!
It's time we admitted we just don't have good enough scientific methods to study the effect of food on health at population scale.
https://t.co/luJ2JpYMX4
Promising results from South Asia on community based interventions for Hypertension.
Could lead to up to 25% less cardiovascular events.
Downstream effects of Hypertension cause more than 10 million deaths annually. https://t.co/CxMHj6ipoO
Are genome based predictive risk scores for heart disease better than currently used pooled-cohort clinical risk scores?
NO!
Results from analysis of GWAS data from the UK biobank
https://t.co/PHz251acU7
This review summarizes the evidence for the benefits and possible risks associated with consumption of cow’s milk. The authors describe the relationship of milk consumption to the risks of fracture, obesity, cardiovascular disease, allergies, and various cancers.
The NELSON Trial in summary:
Fancy a low dose CT scan for 1 in a 1000 chance of improving lung Cancer diagnosis?
All the while your chance of dying from any cause remains exactly the same?
https://t.co/mNeUA2nIC8
.@setopati Please for heaven's sake do not publish misleading interpretations of nutriontional epidemiology studies like this.
Studies that are problematic with their methodology to begin with.
Happy to assist if you need help understand these studies.
https://t.co/eDGWmFpZrP
The @Surgeon_General General has just released a report on smoking cessation -- here's what you need to know. Read the full report: https://t.co/3Cq52PUsol #SGCessation
What's a lot more effective at reducing cancer deaths than precision medicine? Population Statistics.
True that!
Our best weapons against cancer are not magic bullets https://t.co/wResIIYGUI
If you have been confused by the recent brohuhaha over plant-based diets and meat-based diets, this JAMA piece offers a blow by blow account.
Don't forget the popcorn!
https://t.co/efvxL6QpUg
Vox has started a wonderful series on national health systems called Everybody Covered.
This week on feature--Taiwan. A great example of a well-functioning single payer system.
https://t.co/FZFxqgcoDP via @voxdotcom