Google’s AI system reportedly capable of detecting breast cancer

By Mateen Dalal

	Radiologists miss nearly 20% of breast cancer spotting in mammograms
	Google’s AI technology is developed with Alphabet Inc.’s DeepMind AI


 

As per a latest study, a novel AI technology developed by American multinational technology firm- Google could provide enhanced results in spotting breast cancer, as compared to radiologists.

Reportedly, the research witnessed the AI-powered computer compare the technology with medical experts as they both screened mammograms.

 

Apparently, the study stated that the AI solution was as accurate as the humans at spotting incidences of breast cancer and it prevented the occurrence of false-positive cases better. Healthcare professionals are aiming at deploying similar technology with an aim to enhance the rate of breast cancer detection, which affects close to one in eight women along with other health issues.

 

For the record, the comparison was facilitated by researchers from the U.S. and U.K. and was published in the journal named Nature. This suggests that AI could help bring about drastic positive changes in the healthcare industry.

 

According to the American Cancer Society, radiologists miss close to 20% of breast cancers in mammograms. Moreover, half of all the women who get the screenings done over a 10-year period report a false positive result.

 

The outcomes of the study developed with the DeepMind AI system of Alphabet Inc. which amalgamated with Google Health in September showcases a key potential in early detection of breast cancer.

 

The team that included researchers at NHS and Imperial College London trained the technology to identify cases of breast cancer on more than tens of thousands of mammograms. Additionally, they compared the performance of the system with actual results from a set of nearly 25,856 mammograms in the U.K. and nearly 3,097 from the United States.

 

As per the study, the AI mechanism could identify cancers with a similar degree of accuracy as leading radiologists and reduced the total number of false-positive results by close to 5.7% in the US-based group and by 1.2% in the British group. Furthermore, it also cut the number of false negatives where tests were marked as normal by 2.7% in the British group and 9.4% in the U.S.

 

 

Source Credits: https://www.independent.co.uk/life-style/gadgets-and-tech/news/google-ai-breast-cancer-artificial-intelligence-deepmind-healthcare-a9267081.html