3 ms·
Much of what doctors do could and has been written as expert rule-based systems. You might find the 1980s Mycin program for diagnosing different kinds of infect
by jurassic 12y ago
Much of what doctors do could and has been written as expert rule-based systems. You might find the 1980s Mycin program for diagnosing different kinds of infections interesting; it was said to outperform members of Stanford medical school faculty in the domain that it addresses.
One problem with this style of tech is that your program is only as smart as the rules you give it, and you need seasoned doctors to help put together this knowledge. Even if they had the time, what doctor wants to facilitate the commoditization and automation of his or her profession, the undermining of their voice of authority?
I also question whether regular people have enough common sense and medical vocabulary necessary to safely operate such a system on their own behalf. Seems like it would be a fertile ground for lawsuits.
http://en.wikipedia.org/wiki/Mycin http://en.wikipedia.org/wiki/Mycin
http://en.wikipedia.org/wiki/Expert_system http://en.wikipedia.org/wiki/Expert_system
- j_lev 12y agoThanks everyone for the responses. > what doctor wants to facilitate the commoditization and automation of his or her profession Agree in principle but it just takes one incumbent to go rogue and become rich selling his knowledge, and the entire market is disrupted. It happened in the securities industry - the knowledge of old-school pit traders has been almost completely transferred to algorithms. We're seeing the start of it in drug stores now too. angersock mentions the same concerns below with doctors' egos, but the thing to remember is that initially these systems are brought in to be operated by a trader/pharmacist; complete replacement is several steps down the line.