This week at ESC Congress 2026 (Munich, August 28–31), researchers from the University of Tokyo and Institute of Science Tokyo presented data showing that AI analysis of facial videos can detect undiagnosed hypertension and diabetes with remarkable accuracy — no blood test, no cuff, no clinic visit required. Here's how it works: a high-speed spectroscopic camera records a 30-second video of your face and palms. A machine-learning algorithm extracts pulse-wave dynamics (arterial stiffness markers), skin blood flow patterns, and the spectral characteristics of skin coloring — all of which correlate with blood pressure and blood sugar regulation. The numbers: • Hypertension detection: 95% accuracy from a 30-second recording • Still 90.3% accurate from just a 5-second video • Diabetes detection: similarly high accuracy using facial blood flow patterns This isn't speculative — it's a prospective clinical study. And the implication is significant: a world where cardiovascular risk screening happens through a phone camera at the pharmacy counter, at a wellness clinic, or at a routine eye exam. For preventive cardiology, this is the future of population-level risk stratification — finding the undiagnosed 30-79 year olds with hypertension (there are ~1.4 billion globally) who never come in for a clinic visit because they feel fine. Source: Medical Xpress / ESC Congress, August 26, 2026

https://www.caristo.com/caristo-diagnostics-cari-heart-coronary-inflammation-technology-authorized-by-u-s-fda/ Will be able to assess for inflamed fat around coronary arteries

Looks like Monday mornings are not a good option for Live "Ask Me Anything" No recording for meeting today Please let me know in the comments if Live "Ask Me Anything" are valuable The attendance has been low the last couple meetings If there is a better date and time for the majority of people we will work on using that going forward If you would prefer another format other than Live meetings then please make suggestions Thanks Michael Twyman

A new podcast dropped with Lillie Kane. She's nutritionist. I don't see eye to eye with her on everything from a nutrition slant. You can see that I'm not too thrilled with the red light, yellow light, green light game of picking foods if they are good or bad because, as you all know, it always depends on your context. I was glad that I was able to ultimately bend the conversation towards circadian biology. Discussing what food is really about is a barcode to your light environment and how your body processes the electrons and the protons from the foods. I believe she broke this podcast up into a couple different segments. This is segment one so hope you enjoy.

https://www.perplexity.ai/computer/a/c3d89869-df99-436a-b425-b79e27ecc439 Sneak peak of how I try to keep up to date My weekly cardiovascular updates pulled from multiple sources