Hacker News Reader

Comment by bsder | Hacker News Reader

Part of the issue is that the doctor is on the hook if the tool fouls up.

Consequently, having an extremely skeptical viewpoint on tools is perfectly rational.

I can also tell you from discussions with doctors that one of the problems is that the intersection of GUI programmer, competent engineer, and medical domain knowledge is either a null set or a single person. (For example: EEG analysis seems to be a natural fit for ML/AI--enormous amounts of data with events only sporadically scattered in it--yet there is nobody capable of handling the intersection of talents required.)

View on HN

It reminds me of Moneyball; in it, the author points out that the statisticians were frustrated by their inability to get traction within the MLB, but their pitch essentially boiled down to "you guys don't listen to statistics, you should listen to these new ones we just developed," which left unsaid that the statisticians made up new quantifiers because the old ones were ineffective. The MLB had the lived experience of those statistics being ineffective, so they knew Bill James et al were right about that, but the idea that the answer was more numbers that didn't make a lot of intuitive sense was a hard sell.

I would also add that my perception from working at a major east coast hospital has actually been that hospital IT clamps down on new tools more than anyone because of HIPAA requirements, etc, that the doctors ignore/don't care about as much as they should. It's a complicated, layered system.

> intersection of GUI programmer, competent engineer, and medical domain knowledge is either a null set or a single person

Ah interesting.

My neighbor's son is a doctor who needed better software. He trained himself to become a GUI programmer. The resulting company is very succesful. (As reported by his mother, so I'm allowing for some parental bias.)

There is a bunch of ML/EEG devices on the market, it is a commodity technology by now, just google "fda machine learning eeg".

Reply on news.ycombinator.com