Google AI can detect breast cancer more effectively than doctors

Ceria Alfonso
Octubre 17, 2018

Metastatic tumors and cancer cells escaping from their tissues of origin travel through the circulatory system or lymphatic systems and form new tumors in other parts of the body and are hard to detect.

Is there nothing artificial intelligence can't do? Finding them is a time-intensive and hard task for pathologists. Taking tissue from the lymph nodes, or doing a biopsy in medical terms, is a way to detect metastatic tumors, which are cancerous cells traveling from the tissue of origin (the breasts in this case) to a different or secondary site within the patient's body. For comparison, pathologists are on average 81pc accurate at detecting these cells when under time constraints. LYNA also cut the slide review time from two minutes to one minute. "We provide a framework to aid practicing pathologists in assessing such algorithms for adoption into their workflow (akin to how a pathologist assesses immunohistochemistry results)".

This artificial intelligence system called Lymph Node Assistant or LYNA is described in a scientific report titled Artificial Intelligence based on Breast Cancer, published in the American Journal of Surgical Pathology. As the investigators explained, it examines a 299-pixel image (Inception-v3 default image size), then analyzes volumes at the pixel level and then exports them as labels - that is, predictions and classifies them as benign or malignant.

The main issue LYMA is that it can diagnose the disease only at a late stage although scientists are sure it can be adapted to search for other types of tumors.

More news: Bundle de Diablo III: Eternal Collection para Nintendo Switch

When and if LYNA will become ready for practical use, the benefits of the AI would be enormous - not only it would allow doctors more to time care for their patients but it would also lead to more reliable and swift diagnoses that could save hundreds if not thousands of lives.

Google is investing in health care applications for artificial intelligence.

Luckily, Google AI and researchers at the Naval Medical Center San Diego have a solution.

Otros informes por

Discuta este artículo

SIGUE NUESTRO PERIÓDICO