1月1日《自然》雜志發(fā)布的一項(xiàng)最新研究顯示，人工智能從常規(guī)乳房X線掃描中識(shí)別乳腺癌的得分已經(jīng)超過(guò)了醫(yī)學(xué)專(zhuān)家。

乳腺癌是女性最高發(fā)的癌癥之一。對(duì)于女性而言，定期做一次乳房X光檢查是很有必要的。但掃描結(jié)果解讀不可避免會(huì)存在一定誤差。為此，谷歌一種人工智能模型出現(xiàn)了！這個(gè)模型可以根據(jù)掃描結(jié)果預(yù)測(cè)乳腺癌，其準(zhǔn)確性與放射科專(zhuān)家相似。

然而一旦涉及到人工智能，類(lèi)似的顧慮就會(huì)被提及：人工智能是否最終會(huì)取代人？

?

圖片來(lái)源：視覺(jué)中國(guó)

A computer programme can identify breast cancer from routine scans with greater accuracy than human experts, researchers said in what they hoped could prove a breakthrough in the fight against the global killer.

研究人員稱(chēng)，一種電腦程序可以通過(guò)常規(guī)掃描準(zhǔn)確識(shí)別乳腺癌，而且它的準(zhǔn)確率比人類(lèi)專(zhuān)家的更高。他們希望該電腦程序可以在與乳腺癌這一全球殺手的斗爭(zhēng)中取得突破。

Breast cancer is one of the most common cancers in women, with more than 2 million new diagnoses last year alone.

乳腺癌是女性最常罹患的癌癥之一，僅去年一年就有200多萬(wàn)名新確診的病例。

Regular screening is vital in detecting the earliest signs of the disease in patients who show no obvious symptoms.

在患者沒(méi)有明顯癥狀的時(shí)候，定期篩查對(duì)于發(fā)現(xiàn)疾病的早期癥狀至關(guān)重要。

In Britain, women over 50 are advised to get a mammogram every three years, the results of which are analysed by two independent experts.

在英國(guó)，50歲以上的婦女被建議每三年做一次乳房X光檢查，而其檢查結(jié)果由兩位單獨(dú)的專(zhuān)家進(jìn)行分析。

But interpreting the scans leaves room for error, and a small percentage of all mammograms either return a false positive - misdiagnosing a healthy patient as having cancer - or false negative - missing the disease as it spreads.

但對(duì)掃描結(jié)果的解讀也有可能出錯(cuò)，而且在所有乳房X光檢查結(jié)果中，有一小部分要么是假陽(yáng)性（將健康病人誤診為患有癌癥），要么是假陰性（在疾病的傳播過(guò)程中，沒(méi)有診斷出疾?。?/div>

Now researchers at Google Health have trained an artificial intelligence model to detect cancer in breast scans from thousands of women in Britain and the United States.

如今谷歌健康中心的研究人員已訓(xùn)練出一個(gè)人工智能模型，它可以通過(guò)乳房掃描來(lái)檢測(cè)英國(guó)和美國(guó)數(shù)千名女性是否罹患癌癥。

The images had already been reviewed by doctors in real life but unlike in a clinical setting, the machine had no patient history to inform its diagnoses.

實(shí)際上，醫(yī)生們已經(jīng)檢查過(guò)這些圖像了，不過(guò)與臨床環(huán)境不同的是，這臺(tái)機(jī)器不知道病人的病史，也沒(méi)有據(jù)此來(lái)進(jìn)行診斷。

The team found that their AI model could predict breast cancer from the scans with a similar accuracy level to expert radiographers.

該研究團(tuán)隊(duì)發(fā)現(xiàn)，他們的人工智能模型可以通過(guò)掃描檢查來(lái)預(yù)測(cè)乳腺癌，而且其準(zhǔn)確度與放射科專(zhuān)家相當(dāng)。

Further, the AI showed a reduction in the proportion of cases where cancer was incorrectly identified - 5.7 percent in the US and 1.2 percent in Britain, respectively.

此外，該人工智能模型還顯示，癌癥被錯(cuò)誤識(shí)別的比例有所下降，其中美國(guó)降低了5.7%，英國(guó)降低了1.2%。

It also reduced the percentage of missed diagnoses by 9.4 percent among US patients and by 2.7 percent in Britain.

美國(guó)和英國(guó)的漏診率也分別降低了9.4%和2.7%。

"The earlier you identify a breast cancer the better it is for the patient," Dominic King, UK lead at Google Health, told AFP.

“越早發(fā)現(xiàn)乳腺癌，對(duì)病人越好，”谷歌健康中心英國(guó)分部負(fù)責(zé)人多米尼克?金告訴法新社。

"We think about this technology in a way that supports and enables an expert, or a patient ultimately, to get the best outcome from whatever diagnostics they've had."

“我們認(rèn)為這項(xiàng)技術(shù)能夠支持并最終使專(zhuān)家或患者從他們得到的診斷中獲得最佳結(jié)果?！?/div>

多知道一點(diǎn)

乳腺癌

乳腺是由皮膚、纖維組織、乳腺腺體和脂肪組成的，乳腺癌是發(fā)生在乳腺腺上皮組織的惡性腫瘤。90%以上的乳腺癌發(fā)生在女性身上。

乳腺并不是維持人體生命活動(dòng)的重要器官，原位乳腺癌并不致命；但由于乳腺癌細(xì)胞喪失了正常細(xì)胞的特性，細(xì)胞之間連接松散，容易脫落。癌細(xì)胞一旦脫落，游離的癌細(xì)胞可以隨血液或淋巴液播散全身，形成轉(zhuǎn)移，危及生命。

全球乳腺癌發(fā)病率自20世紀(jì)70年代末開(kāi)始一直呈上升趨勢(shì)。目前乳腺癌是女性最高發(fā)的癌癥之一，僅2019年便有超過(guò)200萬(wàn)例新診斷。

?