> When you're talking about how close it is to "parity", it matters a lot. I guess the question there is how valuable 'close to parity' is. If you can treat mo

> When you're talking about how close it is to "parity", it matters a lot. I guess the question there is how valuable 'close to parity' is. If you can treat more patients, or treat them faster, or even much cheaper, then close to parity is a big deal. If examining patients and avoiding destructive false positives are the limiting factors, it's not clear that parity-level decisions are a significant benefit.