Post Breast Surgery Mayo Read My Mammogram Results?
A.I. Is Learning to Read Mammograms
Computers that are trained to recognize patterns and interpret images may outperform humans at finding cancer on 10-rays.
Artificial intelligence tin can assist doctors do a better job of finding chest cancer on mammograms, researchers from Google and medical centers in the The states and Uk are reporting in the journal Nature.
The new system for reading mammograms, which are Ten-rays of the breast, is nevertheless beingness studied and is not however bachelor for widespread apply. It is just one of Google's ventures into medicine. Computers tin exist trained to recognize patterns and interpret images, and the visitor has already created algorithms to help detect lung cancers on CT scans, diagnose heart illness in people with diabetes and notice cancer on microscope slides.
"This paper volition help movement things along quite a chip," said Dr. Constance Lehman, director of breast imaging at the Massachusetts General Hospital in Boston, who was not involved in the study. "There are challenges to their methods. But having Google at this level is a very expert thing."
Tested on images where the diagnosis was already known, the new arrangement performed improve than radiologists. On scans from the Us, the system produced a 9.four percent reduction in false negatives, in which a mammogram is mistakenly read as normal and a cancer is missed. Information technology too provided a lowering of 5.vii pct in false positives, where the scan is incorrectly judged abnormal but there is no cancer.
On mammograms performed in Britain, the organisation likewise beat the radiologists, reducing imitation negatives by 2.7 percent and false positives past 1.2 percentage.
Google paid for the written report, and worked with researchers from Northwestern University in Chicago and two British medical centers, Cancer Enquiry Majestic Centre and Royal Surrey County Hospital.
Last year, 268,600 new cases of invasive breast cancer and 41,760 deaths were expected among women in the United States, co-ordinate to the American Cancer Gild. Globally, there are virtually 2 million new cases a year, and more than than half a 1000000 deaths.
Nearly 33 one thousand thousand screening mammograms are performed each yr in the United States. The examination misses about 20 pct of breast cancers, according to the American Cancer Lodge, and false positives are common, resulting in women being called back for more tests, sometimes fifty-fifty biopsies.
Doctors have long wanted to make mammography more than accurate.
"There are many radiologists who are reading mammograms who make mistakes, some well outside the adequate margins of normal homo error," Dr. Lehman said.
To apply bogus intelligence to the task, the authors of the Nature report used mammograms from well-nigh 76,000 women in Uk and 15,000 in the U.s.a., whose diagnoses were already known, to train computers to recognize cancer.
Then, they tested the computers on images from about 25,000 other women in Uk, and 3,000 in the United States, and compared the system's performance with that of the radiologists who had originally read the X-rays. The mammograms had been taken in the by, so the women's outcomes were known, and the researchers could tell whether the initial diagnoses were correct.
"We took mammograms that already happened, showed them to radiologists and asked, 'Cancer or no?' and then showed them to A.I., and asked, 'Cancer, or no?'" said Dr. Mozziyar Etemadi, an author of the study from Northwestern Academy.
This was the examination that found A.I. more accurate than the radiologists.
Unlike humans, computers do not get tired, bored or distracted toward the end of a long day of reading mammograms, Dr. Etemadi said.
In another test, the researchers pitted A.I. confronting 6 radiologists in the The states, presenting 500 mammograms to be interpreted. Over all, A.I. again outperformed the humans.
Merely in some instances, A.I. missed a cancer that all six radiologists found — and vice versa.
"At that place'due south no denying that in some cases our A.I. tool totally gets it wrong and they totally become it right," Dr. Etemadi said. "Purely from that perspective it opens up an entirely new area of enquiry and study. Why is information technology that they missed it? Why is it that we missed information technology?"
Dr. Lehman, who is also developing A.I. for mammograms, said the Nature report was strong, but she had some concerns about the methods, noting that the patients studied might not exist a truthful reflection of the general population. A college proportion had cancer, and the racial makeup was not specified. She also said that "reader" analyses involving a pocket-size number of radiologists — this study used half dozen — were not always reliable.
The next step in the inquiry is to have radiologists try using the tool as function of their routine practice in reading mammograms. New techniques that pass their initial tests with flying colors practise not always perform too out in the real world.
"We have to run across what happens when radiologists have information technology, run across if they do better," Dr. Etemadi said.
Dr. Lehman said: "Nosotros have to be very careful. We want to brand sure this is helping patients."
She said an before technology, computer-aided detection, or CAD, provided a cautionary tale. Canonical in 1998 past the Food and Drug Administration to assistance radiologists read mammograms, information technology came into widespread use. Some infirmary administrators pressured radiologists to use information technology whether they liked information technology or non because patients could be charged actress for it, increasing profits, Dr. Lehman said. Later, several studies, including ane that Dr. Lehman was part of, establish that CAD did not amend the doctors' accurateness and even fabricated them worse.
"We tin learn from the mistakes with CAD and do it improve," Dr. Lehman said, adding that A.I. has become far more powerful, and keeps improving as more than data is fed in. "Using computers to enhance homo performance is long overdue."
She and Dr. Etemadi said that a potentially good use of A.I. would be to sort mammograms and flag those nigh in demand of the radiologist's attention. The system may likewise be able to identify those that are clearly negative, and then they could exist read quickly and patients could promptly be given a clean bill of health.
Although developers of A.I. often say information technology is intended to help radiologists, not supersede them, Dr. Lehman predicted that eventually, computers alone will read at least some mammograms, without aid from humans.
"We're onto something," she said. "These systems are picking upwards things a human might not meet, and nosotros're right at the showtime of it."
[ Like the Science Times folio on Facebook. | Sign up for the Science Times newsletter. ]
Source: https://www.nytimes.com/2020/01/01/health/breast-cancer-mammogram-artificial-intelligence.html
0 Response to "Post Breast Surgery Mayo Read My Mammogram Results?"
Post a Comment