For some years I've been thinking about how to use modern machine learning and artificial intelligence in animal training. Most "automated" uses of animal training involve lab animals pressing some buttons, but it should be possible to teach a mouse/rat to visit a particular place in a cage, or hold a certain pose, sniff out something, or a parrot to hum a tune.
But it would also require an algorithm to score "progress" toward a multi dimensional training goal and encode that into binary reward/no reward signals.
I'd say hyperthyroidism is difficult to overlook. At least I hope nobody is started on long term stimulant treatment without routinely checking simple blood parameters.
ADHD has severe consequences on children: less happiness, less self control, more aggressiveness, higher risk of substance abuse and in general much lower grades.
Unless a new treatment can demonstrate that it can reduce these consequences as well as medication does in the majority of cases, let's stick with the treatment that works, shall we?
Today the population limiting factor for Humans seems to be women's education. No kidding, this seems to be the number one factor in birth-rates. The more education women are getting the less babies are born. And I am generally in favor of both, more education and less procreation.
I don't believe they do, because the death rates just aren't that high. Population-limiting factors for Humans have historically been mostly food supply and childhood diseases. I've heard that Malaria wasn't so much of a problem until people where driven into malaria-infested areas due to population growth.
How is a doctor supposed to diagnose a Virus like Zika without advanced lab tests? The symptoms are non-specific and often non-obvious. In fact, I doubt Zika virus was regularly tested for.
I'd exclude breast cancer because it's detectable with non-medical means before it becomes fatal. I also wasn't speaking so much about medical incidence statistics, but more about perceptions.
If you love what you do, you'll never work an hour in your life. So maybe academics shouldn't work at all, since the compensation isn't worth doing something you don't love ;-)
One point of MOOCs is that they are not text books. That does not mean text books are bad or that MOOCs are all you'll ever want.
Some MOOCs create video content totally "from scratch", but usually they build on the lecturers' ability to hold a lecture in the topic, often honed through years of repeated lecturing on the same topic.
I would not subscribe to the statement about what "the evolution says". Evolution does not produce the "best" individuals, it just drives population towards a local maximum in their particular environment, from their particular starting point. And that may mean a long life span or a short life span.
There is no cancer which affects "the old" predominantly. The chances stay pretty much the same (with some exceptions). Also, at some age beyond 60, cancer stops being the main medical cause of death, being replaced by infections.
I also don't subscribe to the theory that individuals have to procreate to boost the survival of their genes. In reality it's a bit more complex than that, and even very old people can and do contribute to the survival of their children or their close relatives.
I'd also like to recommend the novel series "The Expanse". While not entirely accurate, at least the start of the series deals with a lot of the near-future limits in space travel: lack of gravity and the consequences thereof for humans, acceleration, radiation etc.
And in the new TV-Adaptation there are fascinating little details like sparrows flying/hovering in 0.3G inside a tunnel inside an Asteroid (Ceres) with an artificially induced spin.
Well, preventing Cancer has been a priority for evolution since the first multicellular organisms arose. It's just not that easy.
Developing a drug is an economically difficult thing: Spend billions for the chance to win billions. The only alternatives to big pharma in developing these drugs are big government and big charity (Gates' Foundation et al).
The obvious solution is probably the best: Use an asynchronous framework (like node.js or asyncio) for the REST interface and wait for the completion of the task to return a response.
That process wouldn't handle all the details of the task, but it can wait for notification of the task via other means (celery, rabbitmq, redis, whatever...).
Personally I think REST is the best paradigm for most microservices. There is one exception: Data streams are inherently stateful and some kind of websocket/message passing is a lot more useful in that case.
But it would also require an algorithm to score "progress" toward a multi dimensional training goal and encode that into binary reward/no reward signals.