Google Cloud has been on a roll lately in its efforts to expand the use of artificial intelligence in healthcare by forming high-powered partnerships across the industry.
The latest addition to Google’s roster is Varian Medical Systems, which develops a range of cancer-fighting technology, from radiotherapy and radiosurgery hardware to care management and analytics software.
Using the Google Cloud AI Platform’s Neural Architecture Search technology, Varian will input its own repository of imaging data to create and “train” models that can automatically perform organ segmentation on diagnostic images.
Organ segmentation, also called contouring, is one of the first and most important steps in creating a cancer treatment plan. The traditional method sees clinicians spending hours analyzing scans to identify the organs and tissues that will need to be either targeted or protected during radiation therapy. Their findings are then used to calculate the precise dosage of a radiotherapy regimen, and where exactly the treatment should be directed.
Automating the process with AI will not only speed up the treatment timeline, but could also increase access to certain treatments in areas that lack the expertise and resources needed to perform accurate organ segmentation.
The partnership will “turbocharge” Varian’s mission to make personalized, high-quality cancer care widely available, said Corey Zankowski, senior vice president of the company’s office of technology and innovation.
“To that end, we have committed ourselves to Intelligent Cancer Care, which seeks to automate routine or repetitive tasks in the radiation oncology workflow through the use of smart algorithms, machine learning and AI,” Zankowski said.
Varian’s new collaboration comes amid the closing of its acquisition by Siemens Healthineers. Under the terms of the deal, made public last August, Siemens is paying $16.4 billion to purchase Varian, which will keep its name and continue to operate as an independent company under the Siemens umbrella. The transaction, which has since been approved by the necessary shareholders and regulatory agencies in the U.S. and U.K., is expected to be completed by mid-2021.
AI is increasingly becoming the go-to solution in the search for faster and more accurate cancer treatment. Just this week, the FDA cleared its first AI device for the segmentation of brain tumors prior to radiation therapy. Vysioneer’s VBrain uses AI to auto-contour a brain metastasis, meningioma or acoustic neuroma and to detect any additional lesions, all in just a few minutes.
A clinical study published in Neuro-Oncology in March showed that clinicians who used VBrain were able to detect lesions on brain scans with about 12% higher sensitivity than those who performed manual contouring. In all, use of the AI decreased radiotherapy treatment planning time by more than 30%.