Using A.I. to analyze mammogram screenings as reported by The New York Times

Reviewed by: Kennedy Barkhouse and Celina Tang

Average Score: 5 / 5

In Summary: A.I. is learning to read mammograms to help doctors find breast cancer. Radiologists and A.I. were shown mammograms and determined if the patient had cancer or not, A.I. was able to decrease the percent of false positives and false negatives. The New York Times article summarizes the scientific article and doesn’t shy away from discussing the shortcomings of A.I. The article avoids the use of jargon and presents the information without bias. 

Original News Story

Research Article

Image by: Maximillian Soltysiak

Kennedy Barkhouse

              ​​The phenomenon of A.I. screening mammograms for malignancies, one of its many new medical applications, has been revered by many doctors in the field. The technology in this study has decreased the number of false negatives as well as false positives, beating radiologists reading the same mammograms with known outcomes. While A.I. overall outperformed humans, sometimes it missed cancer that all the radiologists found. Alternatively, sometimes A.I. correctly identified cancer that none of the radiologists did.

              Grady does an excellent job of consulting with those directly involved in the research, as well as experts with differing opinions. She has comprehensively reviewed the original Nature published article, expressing doubts as well as hope for the new technology from aforementioned experts in the field. She has done her research and simply reported the facts-no clickbait title. She remained neutral in the language she used, and correctly reported the percentages of false positives and negatives found in the journal article this news story was based on.

              As an author, she is easy to find; her twitter handle is listed as well as a glowing biography illustrating her long and award-winning career. In terms of graphics, the only critique would be the exclusion of the image of the malignancy the radiologists found but A.I failed to, and instead only including the mammogram in which A.I. outperformed the radiologists in locating the malignancy.
​
              Overall, Grady correctly and properly illustrates the findings of the article without the use of jargon. Grady carefully reported the findings accurately, in a well written piece–unsurprising given her extensive science background and reporting style.

Celina Tang

              The New York Times article delves into Artificial Intelligence and its newfound medical implications. The author discusses the future of A.I. and its ability to detect malignancies. Specifically, A.I. can now read mammograms and recognize patterns associated with cancers. This new system has outperformed numerous radiologists, reducing the number of false negatives and false positives.

               The author flawlessly incorporates the key aspects of the original journal, all the while providing credible perspectives from within and outside the study that have varying opinions, all the while remaining neutral. Additionally, readers have access to the author’s Twitter, past articles written, as well as an extensive biography that only amplifies the author’s credibility and experience with scientific reporting.

              ​Although the visual presentation of the article is not perfect, as the number of advertisements and social media prompts are slightly overwhelming, the author successfully reports the findings by the researchers, all the while discussing its shortcomings and the future of A.I.in a way that is free of jargon and appropriate to the target audience.                          

Index for Review

The views expressed by the reviewers for this article are not endorsed or shared by SciFeye. The interpretation of the review of the news story using the SciFeye Index was done independently by two SciFeye reviewers. We encourage you to conduct your own evaluation of the accuracy and quality of the news story using the Index.