Hacker News new | past | comments | ask | show | jobs | submit login

Thanks for linking. If I'm reading that correctly, it's pretty bad in comparison to an average human radiologist at ~6% false positive rate [1]. There's probably a bias factor in there, however, where humans are hesitant to predict potential cancer due to the cost/time involved in follow-up screening.

[1] https://www.ncbi.nlm.nih.gov/pubmed/21643887




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: