91–100 of 197 posts
Earlier quoted context omitted.
EKGs should be extremely easy for AI to identify every disease with a range of probabilities and even some humans can’t identify from EKGs. Do we have the labelled dataset for this?
I wonder if all symptoms would show on an EKG though. Would reduced coronary blood flow alter electrical signals ?
Earlier quoted context omitted.
EKGs should be extremely easy for AI to identify every disease with a range of probabilities and even some humans can’t identify from EKGs. Do we have the labelled dataset for this?
I wonder if all symptoms would show on an EKG though. Would reduced coronary blood flow alter electrical signals ?
Attach an EKG, have the patient ride a bike with significant exertion.
If there is a lack of blood flow, it will show up in the EKG as alteration of the electrical signals through the heart.
Earlier quoted context omitted.
I'm not sure quip generalizations like that are useful. If it was as simple as cheese bad, we would see the dutch and the like as outliers in statistics, but that's not the case.
https://www.nature.com/articles/s41467-020-19589-w ”we evaluated the full spectrum of nutrient intake and identified a significant positive correlation between ImP [imidazole propionate] and saturated fat intake (driven by high cheese intake)” Perhaps this depends on the type of cheese consumed. Of course, it’s a correlation; ImP could be modulating eating habits and making people prefer eating cheese.
There is an enormous scandal to come behind the vilification of cholesterol and the simplification of its level's interpretation to come, which led to the current epidemic of obesity and metabolic/inflammatory/autoimmune diseases. Cholesterol levels vary hugely based on the genetics, epigenetics and lifestyle of an individual, and there is a very large amount of individuals that are within normal range with much higher ldl levels. For example, the cholesterol levels of centennials are usually extremely high.
Things to look out for: - high very small LDL(you need a proper analysis of your LDL levels with a histogram of the size distribution, which is very rarely done and more expensive) - high Triglyceride/HDL ratio(in US units, anything above 2 is not a very good sign) - high hbA1C (metabolic issues) - high lp(a) and/or lp(b) - high hs-CRP (general inflammation, but can be caused by infection if you are sick)
Usually those are all related and high when affecting a normally healthy individual).
Earlier quoted context omitted.
>he said it takes 10-20 years for meaningful advances in medicine to show up in general use Could it be that it takes that long to determine whether those advances are actually worthwhile? I can’t count the number of HN posts I’ve seen touting breakthroughs in medical research that ultimately didn’t pan out.
No, this is the time it takes for the proven treatment advances to reach rank-and-file doctors (and insurance policies). I've read the numbers 17 to 22 years I think. There are studies on this, but I don't have references handy.
Look at the adoption of CAR-T therapies. It took 3-4 years before they were regularly used in the US.
One close friend died of a heart attack at 42 and another found a 95% blockage after his CAC scan came back north of 900 at age 40. I'd get it if it's available, the ability to catch certain catastrophic conditions is invaluable.
Before getting a CAC scan, I'd probably do these tests first: * ApoB - about 20% of people with normal cholesterol results will have abnormal ApoB, and be at risk of heart disease. * Lp(a) - the strongest hereditary risk factor for heart disease. * hs-CRP - inflammation roughly doubles your risk of heart disease * HbA1c - insulin resistance is a risk factor for just about everything. * eGFR - estimates the volume of…
Unfortunately there is no approved oral medicine to lower Lp(a) that I am aware of. (I mean given a normal LDL.) Statins don't lower it afaik. An oral medicine named muvalaplin is being tested for it.
https://my.clevelandclinic.org/health/drugs/22550-pcsk9-inhi...
Earlier quoted context omitted.
https://www.nature.com/articles/s41467-020-19589-w ”we evaluated the full spectrum of nutrient intake and identified a significant positive correlation between ImP [imidazole propionate] and saturated fat intake (driven by high cheese intake)” Perhaps this depends on the type of cheese consumed. Of course, it’s a correlation; ImP could be modulating eating habits and making people prefer eating cheese.
This is much more likely related to gut microbia metabolism than it is to histidine. Unlike the dogma wants it to be believed, high protein and fat intake does not necessarily lead to cardiovascular issues. It does when combined with other lifestyle(sedentary, low intensity), metabolic(diabetes/prediabetes, inflammation), nutritional(highly refined carbohydrates, nitrated compounds, oxalate/other anti nutrients and r…
Yes and no: the study linked to in the El Pais article suggests that certain bacteria are converting histidine to imidazole proprionate.
More histidine -> more food for those bacteria -> bacterial overgrowth, dysbiosis and increased ImP levels -> eventually atherosclerosis.
There are probably some microbe(s) capable of outcompeting the ImP-producing ones even in the presence of increased histidine, which would serve as a mitigating factor.
Earlier quoted context omitted.
Fascinating how these tests are something that is an option in America with people getting them. In the UK we have the NHS and, although private healthcare is available, the NHS are gatekeepers with long waiting lists to see the doctor. In UK culture you are just not going to 'waste NHS time' by asking for the tests that could inform you on what lifestyle choices you might need to make in order to head off chronic no…
>Do healthcare providers actively upsell testing in the USA? In my experience, normal doctors do not, but there are a lot of private businesses that make their living selling testing. Also, consider that despite a lot of people knowing their levels in the US through testing availability, health outcomes are not better. So, we know more, but don't do anything about it. I don't know what's worse.
Earlier quoted context omitted.
"It should be very easy for an AI to look at an x-ray, CT, ultrasound or MRI image and tell what disease a human got, even some that humans don't know of. Are there any labeled datasets " -Some Software dude, every month since 1971
No the examples you give are extremely difficult compared to the 2D graphs of an EKG. The Apple Watch is clearly doing some fairly accurate inference with a single lead around arrhythmias, for example. I’m really sorry for being enthusiastic about machine learning, I just finished doing an intensive ML bootcamp and it was fun getting results. I also actually have some heart issues so I’d love to see if I could get a…
Then maybe learn a bit more about the domain before posting silly things.
Earlier quoted context omitted.
Fascinating how these tests are something that is an option in America with people getting them. In the UK we have the NHS and, although private healthcare is available, the NHS are gatekeepers with long waiting lists to see the doctor. In UK culture you are just not going to 'waste NHS time' by asking for the tests that could inform you on what lifestyle choices you might need to make in order to head off chronic no…
>Do healthcare providers actively upsell testing in the USA? In my experience, normal doctors do not, but there are a lot of private businesses that make their living selling testing. Also, consider that despite a lot of people knowing their levels in the US through testing availability, health outcomes are not better. So, we know more, but don't do anything about it. I don't know what's worse.
Out of the approx. 250 million adult americans, a large cross section do manage their health.
While average outcome might be better in the UK, it's useless to lump the 60% (150 million) of americans who are not obese in with the 40% (100 million) who are obese. And while this is easily the most major, it is just one measure of health.