"It decreases the amount of blood flow to the heart, making it difficult to breathe."
AI can predict when you’ll keel over and die clutching at your chest from a heart attack better than doctors can, apparently. A group of researchers scraped together dataset from over 80,000 patients from CALIBER, a large clinical research project studying mortality in England, and used machine learning methods to find common …
making it difficult for oxygen to be circulated around the bloodstream
Nope. The article does not explain it very well, but the arteries which are involved are the coronary arteries which supply blood and oxygen to the muscle of the heart itself, not those which transport blood to the rest of the body.
A heart attack is actually the death of a portion of the muscle tissue of the heart, due to oxygen starvation caused by blocked or constricted coronary arteries. The proper term for a heart attack is a Myocardial Infarction, which literally means heart muscle death.
Note that a Heart Attack is not the same as a Cardiac Arrest.
Difficulty breathing can be associated with heart problems, yes, as in congestive cardiac failure, or atrial fibrillation, where the breathing problems are due to a backlog of fluid in the lungs.
A heart attack does not usually cause breathing difficulty except as a consequence of the associated chest pain, however "shortness of breath" is often reported as a symptom as the body is trying to compensate for the lack of oxygen to the heart. There is no actual impairment of breathing.
The difference is pretty small, and I am not sure that, in the last few years, GP Home Visits are so good an indicator. My brother is currently in hospital after the GP surgery sent an Ambulance, and that felt like one of a range of options they had, from "come to see us" upwards. And can there be a difference between the almost routine and the urgent cases? (I'm thinking of the elderly with limited mobility.)
Did your report over-simplify?
There are so many factors at work including genetic predisposition s for contributing factors.
Personally I consider a major factor for me in keeping my 'heart age' low is the old saying ' You are only as old as the woman you feel'.
My current and first wife are both 18 years younger than me and few people believe I am the age I am.
Over four decades of martial arts helps as does not drinking to excess, when the bottle is empty I rarely open another.
Another major factor is not having a phone number for a doctor, so no home visits for me, those guys are dangerous!
Only this morning I heard on the radio that if you smoke and are socially isolated it's 50/50 which is the biggest health hazard. It's a reasonable assumption there is a correlation between one's social engagement and the youthfulness of one's wife.
And I'm deliberately ignoring other health benefits that might come from having a youthful wife, such as a desire to keep oneself fit for her, joining her on country walks, or being taken out dancing, let alone more domestic physical activity.
There used to be large whisky bottles attached to optics in pubs. I don't know what size they were but Jeroboam seems like a good starting point.
Wine can come in much larger bottles but if you can't pour it easily it's probably too big (Methuselah - 6 litres, and that's not the largest).
> What value does AI add to this? It just seems to me an expensive and long winded way of working something out that we already know so you can add "AI" to the name.
Try reading the article again.
There's a data set of patients who have had different factors who have gone to have heart attacks. The correlation isnt clear to the medics since there are many, possibly inter-dependant, variables - i.e it's not obvious, as you think it is. The computer software, using techniques that gave come to be known as AI or Machine Learning, generates various models that fit the data. These models are then tested against more data sets.
I read the article, it picked out 586 variables though the only one mentioned is home visits which I grant is one that probably should be taken into account however what else could there be? It's not like the NHS has genetic screening and how would you quantify the relationship between other conditions? Lets say you have 5000 people who had heart attacks and 62% had a wart on their hand at some point in their life would the AI consider that a factor? Then what if you have 500,000 people with warts that didn't have a heart attack. How would this be measured using machine learning as you are only using the heart attack patients to build up your model then applying that to everyone? Do they consider these things?
The system is looking at a lot of data, some of which it may never have occurred to researchers to examine. You mention warts... well, perhaps it's possible that people who have verrucas are those more likely to have gone to public swimming baths... it's not implausible that these people are fitter and so have smaller risk of heart diseases. But then, the rich might have their own swimming pool, less chance of verrucas, and due to good diet, health education etc might have even lower chance of heart disease. Complex, inter-dependant factors.
Verrucas might be in the data set, private swimming pool ownership - or number of hiking holidays taken - are unlikely to be.
Anyway, you were dubious of the method based upon its conclusion.
...They perform a "Stress Test" (which isn't, as you may suppose, renegotiating NAFTA with a semi-sentient simeon), but is being connected to an EKG and walking on a treadmill until one's heart rate reaches 140 bpm. Takes maybe 30 minutes. Probably costs $150 all in.
Apparently, there's also a connection to Chaos Math in terms of how the heart regulates the beating of the various chambers. IIRC, it was mentioned in James Gleick's book "Chaos - making a new science".
Biting the hand that feeds IT © 1998–2019