02-01-2020 4:37 am Published by Nederland.ai Leave your thoughts

Google Health has developed a system that can more accurately identify breast cancer than radiologists, in the last sign that artificial intelligence could improve the early detection of disease in images.

In a document published in the scientific journal Nature, Google Health experts, the DeepMind unit of Alphabet, and the UK and US universities showed the AI model both reduced false positives, in which patients “Needs are wrongly told they have cancer, as well as false negatives, where the disease is present but has not been diagnosed.

Screening mammography is known to be imperfect because, according to the American Cancer Society, about one in five breast cancers is not detected. More than half of all women get a false positive every 10 years, leading to anxiety and unnecessary treatment, which was estimated in a 2015 study in the journal Health Affairs to be over $ 4 billion a year.

Dominic King, the British leader for Google Health, said the results were “really exciting” and showed how AI can be used to help screen for cancer at earlier stages when the disease is more difficult to accurately detect. The algorithm was trained and tested on unidentified images from nearly 120,000 mammographies in the US and the UK.

DeepMind recently transferred control of its health division to parent company Google. Mustafa Suleyman, who supervised the health team, leaves DeepMind for a new job that investigates the possibilities and effects of applied artificial intelligence at Google.

Dr. King, who was previously a breast cancer surgeon, said that Google partially started the study because senior radiologists did not think that the British cancer screening services were sustainable. In 2018, the Royal College of Radiologists estimated that the country would need more than 1,000 additional full-time diagnostic radiologists to meet demand.

He hopes that the AI could one day receive legal approval to be a supportive tool for clinicians – but not to replace them. “It could be a second opinion, here and there, giving a push or recommendation to spend more time viewing this scan, or to highlight examples where cases are missed,” he said.

Large technology companies are increasingly interested in using their expertise in artificial intelligence in healthcare, in particular the use of computer vision algorithms to recognize patterns in images that visual signs should investigate in areas such as pathology, ophthalmology and dermatology. .

Google has already created the Lymph Node Assistant, which is 99 percent accurate in detecting late breast cancer cells that have spread throughout the body. DeepMind is preparing for the commercial launch of a device that can diagnose complex eye diseases just as accurately as specialists.

Big Tech also faces competition from start-ups such as Kheiron Medical, which trains a similar algorithm to identify signs of breast cancer. Based on training data from hospitals in Hungary, it beats the average performance of a radiologist.

The Google Health paper was written in collaboration with the Cancer Research UK Imperial Center, Northwestern University in Illinois and the Royal Surrey County Hospital. It turned out that the AI model reduced the number of false positives by 5.7 percent in the US and 1.2 percent in the UK, where each scan is checked by two radiologists. It reduced false negatives by 9.4 percent in the US and 2.7 percent in the UK.

Professor Ara Darzi, a paper author and director of the cancer center in Imperial, said it was “very encouraging.”

“There will of course be some challenges to address before AI can be implemented in mammography screening programs around the world, but the potential for improving healthcare and helping patients is huge,” he said .

Tags: , , , , , ,

Leave a Reply

Your email address will not be published.

thirteen + fifteen =

The maximum upload file size: 256 MB. You can upload: image, audio, video, document, spreadsheet, interactive, text, archive, code, other. Links to YouTube, Facebook, Twitter and other services inserted in the comment text will be automatically embedded. Drop file here