評論
打印
Aside from the almost-instantaneous warning signs--chest pain, shortness of breath, nausea--heart attacks are notoriously hard to predict. Standard models tend to "oversimplify" cardiovascular disease, reducing risk to eight core baseline variables.
除了猝發(fā)的警告信號——胸痛、氣促、惡心以外，心臟病是眾所周知的難以預(yù)測。標(biāo)準(zhǔn)模型似乎讓心血管疾病“過分簡單化”，將風(fēng)險(xiǎn)降至八個(gè)核心的基本變量。
These factors don't automatically foretell a heart attack. Machine learning offers an alternative approach, exploiting big data to minimize human error. Using four computer-learning algorithms, nearly 25,000 fatal or non-fatal cardiovascular events were documented over the study's 10-year period, about 75 percent of which were accurately predicted by the algorithms.
這些因素并不會自動預(yù)測心臟病。機(jī)器學(xué)習(xí)提供了另一種方法，利用大數(shù)據(jù)把人為誤差降到最低。通過四種計(jì)算機(jī)學(xué)習(xí)算法，在10年研究期間，有近25000例致命或非致命心血管疾病被記錄了下來，其中約75%的病例被算法準(zhǔn)確預(yù)測到。
The neural network formula tested highest, beating existing guidelines by 7.6 percent. In fact, the MVP algorithm correctly identified 355 more patients than human doctors, proving its ability to help save lives.
神經(jīng)網(wǎng)絡(luò)公式的預(yù)測率最高，超過現(xiàn)有指導(dǎo)方針7.6%。其實(shí)，MVP算法比人類醫(yī)生更準(zhǔn)確地預(yù)測了355例病情，證明它確實(shí)有能力挽救生命。
譯文屬可可英語原創(chuàng)，未經(jīng)允許，不得轉(zhuǎn)載。
