Google Develops AI That Detects Early-Stage Breast Cancer

(Photo : Reuters)

Google has developed an Artificial Intelligence that can detect breast cancer more accurately compared to a human expert or even a doctor. 

Breast cancer is the second leading cause of death of women around the world. The number of women who are affected by this cancer has increased in the past few yew years. And most of the time, early detection of this cancer is not seen significantly due to the lack of resources and medical experts, or equipment.

It is therefore very important to develop a system that will not only detect breast cancer more accurately, but also to detect early signs of this cancer. As we know, early detection is key in beating cancer.

As the year begins, Google announced that it has developed an AI that will detect breast cancer more accurately compared to doctors in this field. The development of this AI is the result of the collaborative work of Google and cancer researchers. The study was published in the scientific journal Nature on Wednesday. 

In an article published by CNN, the study shows that the AI developed by Google can produce more accurate information about breast cancer compared to a human radiologist. The subjects of the study were women coming from the United Kingdom and the United States using tens and thousands of mammograms to test breast cancer.

The study shows that using AI provides fewer false positives which means that cancer is detected when it is not and false negatives where cancer is undetected. The result is better compared to human experts because AI reduced false positives by 5.7 percent for U.S. subjects and 1.2 percent of U.K. subjects. Likewise, it also reduced false negatives by 9.4 percent for U.S. subjects and 2.7 percent for U.K. subjects.

However, there is still more work to do with the AI as it has less information about the patients' medical history and prior mammograms compared to a human expert or doctor. 

According to Professor Ara Darzi "this is one of those transformational discoveries you have in your hand, which could disrupt the way we deliver screening in terms of improving accuracy and productivity."

Darzi is one of the authors of the study and currently works as the Director of the Cancer Research in UK Imperial Centre. He also said that he did not expect such impressive results from the AI.

The American Cancer Society said that despite the large-scale breast cancer screening programs in developed countries, mammogram screening still does not find an estimated one in five breast cancers.

The authors of the study wrote in the publication that "the performance of even the best clinicians leaves room for improvement and AI may be uniquely poised to help with this challenge." The authors added that this could help to address the small numbers of radiologists. 

In a report from the Royal College of Radiologists in 2018, it was found out that 75 percent of the UK Radiological department directors feel that there is a need for radiologists who can give a safe and effective service to patients. Before the end of 2023, it is forecasted that there is a need for more than 2,000 radiologists and if this AI will be adopted it will be a great help to address this shortage.

Darzi said that he does not think that AI can substitute for humans at this stage as there is still a need for a radiologist to read the mammogram. The development of AI that can detect breast cancer sits alongside the AI developed by Google to prevent lung cancer and also prevent serious eye disease.