16 – 16 Solution Diagnosing Cancer V3

If you’ve chosen 0.5, then your chances of accidentally missing a cancer are exactly the same as your chances of calling something a cancer that isn’t. If you call it 0.8, you will be great in reducing the cost of diagnostics, but bad in helping people survive when they actually have skin cancer. 0.2 is the number I would choose from these numbers, because of the asymmetry of costs of the regret. So if you find that a person has been sent to the lab but the tissue was not cancerous, you have a regret, discomfort in getting skin extracted and there’s a cost to the lab tests, but the regret is small. If conversely you overlook a cancer because you’ve been just too cheap to send the person to the lab, then this person might die. And that’s a much, much bigger cost. As a result, 0.5 threshold is not the best threshold, and we really have to set the threshold in a more aggre-

%d 블로거가 이것을 좋아합니다: