Google’s Latest AI Program is Best at Detecting Breast Cancer, Study Finds

by Stephen Riddle
Google’s Latest AI Program is Best at Detecting Breast Cancer, Study Finds

The healthcare industry is indeed moving towards advanced technologies; it also includes new approaches such as machine learning. A new clinical study has revealed the contribution of artificial intelligence in detecting breast cancer. Notably, the methodology has offered the best results than proficient doctors. Currently, it is challenging to detect cancer in the primary stages. But in the upcoming time, the machine learning algorithm could play a crucial role in breast cancer diagnosis. A team of American and British researchers has developed this computer algorithm.

As per the research, published in Nature, Google has funded the effort to assist doctors in breast cancer diagnosis. The newly-developed algorithm scans mammograms, or say X-ray pictures. It cuts the number of negative outcomes by 9.4%. At present, breast cancer remains the second most common reason for death in females. Early prediction is the best protection most people have in detecting and curing the fatal illness. Thus, the latest finding seems hopeful to combat the current rate of overlooked cases, which is 20%. Researchers have compared the discovery to a spelling checker for email writing. As per the team, it could reduce the number of overlooked or false-positive cases that could result in life-threatening postponements in treatment. Besides this, the approach will avoid false positives in the course of routine screening. It would save women from needless surgery and a lot of stress.

Experts widely prefer mammograms as a tool to detect cancer; unfortunately, they miss several cases. A scientist at Google and co-author of the study Shravya Shetty says mammograms are highly-efficient, but false positive and false negative figures remain concerning. During the trial, scientists have utilized unnamed mammograms from over 25,000 women across the UK. It also includes data of around 3,000 females in the United States. Shetty noted they have attempted to track the same ideologies that radiologists usually follow. Researchers have trained the program to seek for the symptoms of breast cancer by detecting variations in the breast of women. After that, they have analyzed the computer’s estimates against the factual medical reports of the women. Above all, the AI has offered better results than a radiologist in both the US and UK.

You may also like

Leave a Comment