Every year lung cancer takes more than 1.7 million lives, which is more than breast, prostate and colorectal cancer combined. The major problem is that the majority of cancers cannot be diagnosed at early stages. Google is now working to bring a change with its new AI-based tool, it hopes to make lung cancer prediction more accurate and more accessible.
Google’s AI to Predict Lung Cancer
Normally Radiologists view hundreds of images from a single CT scan in order to screen for lung cancer. Now Google can generate an overall lung cancer malignancy prediction with this new AI model, and identify subtle malignant tissue, or lung nodules, that are usually difficult to see. The AI can also help to reveal the growth rate of suspicious tissue.
Google technical lead Shravya Shetty and product manager Daniel Tse said in a blog post:
“The AI system uses 3D volumetric deep learning to analyze the full anatomy on chest CT scans, as well as patches based on object detection techniques that identify regions with malignant lesions,”
Google has examined 45,856 chest CT screens via its AI. It compared the results against six board-certified radiologists. Interestingly, Google’s AI detected five percent more cancer cases than the radiologists in the first study and also reduced false-positive exams by more than 11 percent.
Google is so satisfied with the initial results but the model needs additional clinical research and testing before it can be deployed. The model will be made available through the Google Cloud Healthcare API.