Latest Google AI To Predict Lung Cancer Better Than Humans

Latest Google AI To Predict Lung Cancer Better Than Humans

Alphabets Google is all set to expand its presence in the health sector. 

Google scientists have told that in the novel, AI Artificial Intelligence has been trained to estimate the risk of lung cancer. He claims that AI works better than humans in the diagnosis of lung cancer.

According to the WHO report, more than 1.7 million deaths per year are due globally due to lung cancer. In the year 2018, cases of lung cancer deaths were reported 2.9 million deaths Lung cancer, if diagnosed in the early stages, is better likely to recover.

Technical Lead Shravan Shetty at Google said that the low dose CT screening has reduced the mortality rate, even to this day, doctors have to face challenges to find it in the initial phase, with the increase in financial costs Intensive long procedures have to be faced.

He further said that in 2017 his team started this project to use AI to solve some of these challenges.

 With the use of 3D volumetric modeling with datasets from Northwestern University and other partners, he has made significant progress in designing an AI for the prediction of lung cancer, which will prepare the ground for future diagnostic testing procedures.

The most common challenge faced by radiologists is the use of 2D mages scanned in a single CT scan, which makes it difficult to spot cancer because it can be very minute in nature.

Google scientists have developed AI models to overcome these difficulties. AI generates an overall report of the number of cancer infections in 3D and also helps in detecting infectious lung substances.

Read more: Huawei Removed From SD Association and Wi-Fi Alliance, Google Removes Its Phones From Android Enterprise Device List

 Google's AI can also see the patient's previous record - can compare and detect any discrepancy.

According to the reports published in Nature Medicine, the AI model contains an automatic imaging system which can detect lung cancer at high accuracy and thus increases the chances of first treatment.

Google's team claims that 45.8K de-identified chests CT screening datasets have been used by NIH to design AI models for this lung cancer- National Lung Screening Trial Study and Northwestern University

By comparing a single CT scan, Google's AI has performed better than six radiologists. So far, it is able to detect cases of lung cancer more than 5%, when compared to radiosurgery without radiation.

The Google Cloud Healthcare and Life Sciences team will now help the research team make this new developed AI model accessible through the Cloud Healthcare API.

Earlier, Google was using AI in healthcare, with research being done in the areas of diseases of the eyes.