Question

    When Barzilay had a routine breast X-ray in her early 40s, the image showed a complex group of white spots in her breast tissue. The marks could be normal, or they could be cancerous-even the best doctors often struggle to tell the difference. Over the next two years Barzilay underwent a second X-ray, a MRI and a biopsy and she was finally diagnosed（诊断） with breast cancer in 2014.

Barzilay was treated and made a good recovery. But she remained terrified that the uncertainties of reading an X-ray could delay treatment so she made a career-changing decision.

A computer scientist at Massachusetts Institute of Technology, Barzilay had never studied health before. Her research used machine-learning techniques for natural language processing. But she had been looking for a new line of research and decided to team up with radiologists (放射科医生) to develop machine-learning algorithms (算法) that use computers' superior visual analysis to spot patterns in X-rays that the human eye might miss.

Over the next four years the team taught a computer program to analyze X-rays from about 32,000 women and told it which women had been diagnosed with cancer within five years of the scan.  They then tested the computer's matching abilities in 3,800 more patients. Their resulting algorithm, published last May in Radiology, was significantly more accurate at predicting cancer than practices generally used in clinics. When Barzilay's team ran the program on her own X-ray from 2012-one her doctor had cleared-the algorithm correctly predicted she was at a higher risk of developing breast cancer within five years than 98 percent of patients.

AI applications are entering clinics at a rapid rate, and physicians have met the technology with equal parts excitement about its potential to reduce their workload and fear about losing their jobs to machines. Algorithms also raise questions about how to regulate a machine that is constantly learning and changing and who is to blame if an algorithm gets a diagnosis wrong. Still, many physicians are excited about the promise of AI programs.

1.What advantage do Barzilay's algorithms take of computers to predict cancer?

A.Superior visual analysis.

B.Timely risk warning settings.

C.Natural-language processing techniques.

D.Large storage of radiological knowledge.

2.What does the underlined word "it" in Paragraph 4 refer to?

A.Cancer. B.Research.

C.The team. D.The program.

3.How do physicians react to AI applications in clinics?

A.They doubt the accuracy of the applications.

B.They have mixed feelings about the applications.

C.They promise to make the best of the applications.

D.They expect to rapidly popularize the applications.

4.It can be inferred from the text that

A.Barzilay suffered from breast cancer for two years in her forties

B.at first Barzilay didn't think X-ray was a reliable way of checking cancer

C.with Barzilay's algorithms her cancer could have been diagnosed earlier

D.Barzilay failed to turn to the best doctors to have her breast checked

Answer

1.A

2.D

3.B

4.C

【解析】

本文是一篇说明文，介绍了Barzilay是如何使用人工智能帮助识别X光照片中癌变组织的。但是还是有一些问题需要解决，不过许多医生还是对人工智能项目的前景感到乐观。

1.细节理解题。根据文章第三段中的“to develop machine-learning algorithms（算法）that use computers' superior visual analysis to spot patterns in X-rays that the human eye might miss”可知机器学习算法利用了计算机出色的视觉分析来发现人眼可能会错过的x光片中的图像。故选A项。

2.词义猜测题。根据划线词所在的句子“Over the next four years the team taught a computer program to analyze X-rays from about 32,000 women and told it which women had been diagnosed with cancer within five years of the scan”可知在接下来的四年里，这个团队教一个电脑程序识别来自32，000位女士的X光照片，并告诉it哪些女士在拍照五年内诊断出癌症。由此可见“it”指的就是前文中的“a computer program”。D项（the program）切题。故选D项。

3.推理判断题。根据文章最后一段中的“AI applications are entering clinics at a rapid rate, and physicians have met the technology with equal parts excitement about its potential to reduce their workload and fear about losing their jobs to machines.”可知人工智能应用正以快速的速度进入诊所，医生们面对这项技术时，一方面对其减轻（他们）工作量的潜力感到兴奋，另一方面又担心自己会被机器抢走工作。由此可见，临床医生们对这项技术的应用既感到兴奋，又感到害怕和担忧，他们的感情很复杂。故选B项。

4.推理判断题。根据第一段“When Barzilay had ...t a second X-ray, a MRI and a biopsy and she was finally diagnosed with breast cancer in 2014.”可知Barzilay在40岁出头的时候做了一次常规的乳房x光检查，图像显示她的乳房组织中有一组白点。这些白点可能是正常的，也可能是癌变组织——即使是最好的医生也很难分辨出它们的区别。在接下来的两年内, 她进行了第二次x光检查，MRI和活检, 最终在2014年确诊了乳腺癌。根据文章第四段中“When Barzilay's team ran the program on her ...at a higher risk of developing breast cancer within five years than 98 percent of patients.”可知当Barzilay的团队从2012年开始用她自己的x光进行这个项目时——她的医生发现——算法正确地预测她在五年内比98%的病人患乳腺癌的风险更高。可以推断出，如果当时能够使用人工智能识别技术，Barzilay的病就可以更早确诊。故选C项。