Company able to correctly distinguish metastatic cancer 99% of the time with the help of its deep-learning tool. This rate is greater than human pathologists.

Google trained an algorithm (named Lymph Node Assistant, or LYNA) to spot the features of tumors that have metastasized (that is, spread), which are notoriously difficult to detect. Half a million people die of breast cancer every year and 90% of them are due to metastasis.

According to studies, the 99% rate is superior to the performance of the human pathologists, and the algorithm was also better at finding small metastases on individual slides. Human pathologists can miss these as much as 62% of the time when under time pressure.

Even after performing on a higher accuracy rate the algorithm will be used as a sidekick. The technology will complement human skills and make it easier and quicker to diagnose metastatic tumors. The algorithm halved the time it took to check a slide on average, cutting it to just one minute per slide.

Google is applying AI to health care in a number of different projects. Its subsidiary DeepMind is using AI to look for signs of eye disease at Moorfields Eye Hospital in London, for example. The next step will be to see how LYNA performs in the clinic for real.

Source

Leave a Reply

Your email address will not be published. Required fields are marked *