HealthNewsReview

Yes, yes, real AI died in the AI Winter because the settlers didn't have enough parenthesis to last until spring, it was all very tragic, and now we're all just stirring the pile of linear algebra until the results look good. [1] But that doesn't make IBM's marketing copy any better. And IBM loves making grandiose claims that don't work out in practice. [2] [1] https://xkcd.com/1838/ [2] https://www.healthnewsreview.org/2017/02/md-anderson-cancer-centers-ib…

> "Healthcare providers will be using our platform to gain access to next-gen diagnostic algorithms" And what is the nature of these "next-gen diagnostic algorithms"? Are they the research results of medical experts published to medical journals using limited data sets on limited populations that promise to magically cure each and every malady of any patient population, or rather, is a more realistic example of a "next-gen diagnostic algorithm" something not useless, but less grandiose, such as algorithm that tries to address the likelihood of a narrow problem like sepsis occurring in the ICU? When there are competing algorithms/research papers around the same topic, how do you decide which algorithm is best? And how "actionable" is the algorithm's prediction? All this reminds me a little of IBM Watson marketing and MD Anderson Cancer Center, where the hype machine generated high expectations, and the results did not match the hype. "IBM pitched its Watson supercomputer as a revolution in cancer care. It’s nowhere close": https://www.statnews.com/2017/09/05/watson-ibm-cancer/ "MD Anderson Cancer Center’s IBM Watson project fails, and so did the journalism related to it": https://www.healthnewsreview.org/2017/02/md-anderson-cancer-centers-ib… Google's DeepMind Health also seems worth mentioning in this context:…

Why would that be? In countries with universal health care (which would be pretty high up on the regulation scale) costs for medical procedures are significantly lower. [0] Americans have to understand that they are drastically overpaying for almost everything medical including drugs. There is no way around it. [0]: https://www.healthnewsreview.org/2012/03/comparing-health-care-costs-i…

I chose one at random, it didn't explain an underlying mechanism why vitamin D might have such a large range of health benefits like you claim. It is quite easy for an organisation to p-hack. It's also quite easy for an organisation to ignore correlation vs causation. So I was wanting to know in summary (a blog is fine) what the underlying mechanism is to which vitamin D could solve so many problems. [edit] I'll add an article critical of your way of researching - https://www.healthnewsreview.org/2017/03/vitamin-d-cancer-how-can-2-ne…

Here is a link explaining that a lot of meditation studies are funded by those starting the cult (e.g. Transcendental Meditation, which involves mumbling random gibberish taught to you by a "master"): https://www.healthnewsreview.org/2016/06/conflicts-of-interest-abound-…

Radicava was approved in Japan in 2015 and by the FDA in 2017 to treat ALS though presumably most of the research for it occurred long before then. It costs $145000 per year in the US and $35000 per year in Japan. It slows down the progression of the disease in a specific subset of patients (early stages of the disease with rapid progression of symptoms) which comprises approximately 7% of people with ALS. Trials among a general ALS population failed to show any efficacy. https://www.healthnewsreview.org/2017/09/facts-about-the-new-als-drug-…

Yes, it would be helpful for journalists to take a look at the concept of absolute vs relative risk: https://www.healthnewsreview.org/toolkit/tips-for-understanding-studie…

Well, this news that came out last year is somewhat relevant. MD Anderson Cancer Center's IBM Watson project failed, with $62 million paid to IBM and PwC for essentially no results. However, I don't blame IBM, I blame people who dove in without clearly thinking it through. https://www.healthnewsreview.org/2017/02/md-anderson-cancer-centers-ib…

and one that hasn't been terribly successful in a lot of areas! It's often sold as almost a software/business consulting effort, which requires a ton of money and time to get up and running MD Anderson Cancer Center wasted $62 million on it: https://www.healthnewsreview.org/2017/02/md-anderson-cancer-centers-ib…

The NFL spends massive amounts of money deflecting responsibility: https://www.healthnewsreview.org/2018/01/nfl-cdc-tbi162490/

You way underestimate IBM sales staff. The real problem is that Watson doesn't work, and health care providers spent millions on it which could have gone to real, actual research. https://www.forbes.com/sites/matthewherper/2017/02/19/md-anderson-benc… https://www.forbes.com/sites/jasonbloomberg/2017/07/02/is-ibm-watson-a… https://www.healthnewsreview.org/2017/02/md-anderson-cancer-centers-ib…

Most evidence does not suggest this. 2015: "[...] the NIH made the bold announcement that intensive intervention with a combo of three drugs to reach a new target systolic blood pressure of under 120 mm Hg reduced the rate of heart attacks, heart failure and stroke by 30% and cardiovascular deaths by 25%, " http://www.healthnewsreview.org/2015/09/nih-sprint-study-sparks-questi… http://www.cochrane.org/CD000028/HTN_blood-pressure-lowering-drugs-red… http://eurheartj.oxfordjournals.org/content/early/2012/04/16/eurheartj… http://www.escardio.org/The-ESC/Press-Office/Press-releases/Last-5-yea… There's more evidence available.

Actually he mentions it: >>University officials have repeatedly claimed to have been investigated and exonerated by various legal and regulatory bodies, but those claims have fallen apart. And has these links: http://www.healthnewsreview.org/2013/05/questions-journalists-should-a… http://www.scribd.com/doc/147683397/Attorney-General-Office-Letter-Ref…

"The bigger picture is that almost all hard R&D takes place under the umbrella of some sort of protection from copying." There is a huge difference between protection from copying, and protection from building similar things that achieve the same goal. Patents are not simply "protection from copying". Drug patents often cover things like "using the same body mechanism to achieve a desired result". (Now, you can argue this is necessary to protect them, but let's be clear - it's not straight copying if i make a pill with a completely different chemical composition that ends up triggering the same mechanisms. However, it likely runs afoul of their carefully drafted patents. I'm also aware there are chemical compound patents that would not cover me doing this, but usually, when the drug companies can, they get both) "Intel spends billions a year on semiconductor R&D, and protects it with trade secrets" and design mask protection (see https://en.wikipedia.org/wiki/Integrated_circuit_layout_design_protect… "Pharma companies spend billions on drug R&D and protect them with patents. " They are also heavily subsidized, unlike your intel example. Tax payers, for example, pay about 50% of pharma's R&D costs (and pharma includes a lot more than actual R&D in R&D cost). http://www.healthnewsreview.org/2012/02/the-costly-myths-about-pharmac… and http://www.thefiscaltimes.com/Columns/2012/02/16/What-Those-New-Super-… "Hard R&D that isn't…