Back To What Is New
Article

[SiliconANGLE] Google Cloud's Medical Imaging Suite to made AI-based diagnosis more accessible

10.04.2022

Google Cloud is bringing its expertise in vision-based artificial intelligence to the healthcare industry with the launch of its new Medical Imaging Suite today.

Vision AI has the potential to make a massive impact in healthcare. Medical imaging is one of the most critical tools used by hospitals to diagnose patients, and each year billions of images are made by clinicians to help them understand why people are sick. Google said medical images are so important that they account for about 90% of all healthcare data.

The problem doctors have is that medical images are complex, as it takes expertise to look at one and understand if the patient is inflicted with some kind of disease. Until now, that has always meant human expertise. As a result, radiologists and other healthcare professionals in many hospitals face a huge workload.

Available now, Google Cloud’s Medical Imaging Suite is designed to change this, using AI algorithms to scan medical images and provide faster and more accurate diagnoses. In this way, it thinks it can help increase productivity for healthcare workers while improving care access and patient outcomes.

The offering is designed to address the most common pain points that have prevented healthcare organizations from implementing AI before. The Imaging Lab provides access to ready-made AI-based annotation tools created by Nvidia Corp. and Monai that can perform repetitive, manual tasks such as labeling medical images.

Also onboard is Google’s BigQuery and Looker services, which can be used to search and view petabytes of imaging data. User can then perform advanced analytics on this data, or even use it to create training datasets for new AI models.

The suite’s Imaging AI Pipelines module provides access to Vertex AI, giving users a way to rapidly create AI pipelines and build scalable machine learning models with minimal custom coding.

“Google pioneered the use of AI and computer vision in Google Photos, Google Image Search, and Google Lens, and now we’re making our imaging expertise, tools, and technologies available for healthcare and life sciences enterprises,” said Alissa Hsu Lynch, global lead of Google Cloud’s MedTech Strategy and Solutions. “Our Medical Imaging Suite shows what’s possible when tech and healthcare companies come together.”

Google Cloud’s Medical Imaging Suite also goes to great lengths to ensure the privacy and security of image data, solving one of the main headaches for healthcare providers: compliance. The suite leverages Google’s Cloud Healthcare application programming interface to provide secure data exchange over the DICOMweb standard for secure image transfers. The Cloud Healthcare API is a fully managed, enterprise-grade development environment that offers automated DICOM de-identification.

Lastly, Google said there are multiple deployment options available for the Medical Imaging Suite, including cloud, on-premises or at the edge. This flexibility will ensure that healthcare customers are able to meet a diverse set of data security, privacy and sovereignty requirements, Google said, with centralized management and policy enforcement provided by Google Distributed Cloud.

Constellation Research Inc. analyst Holger Mueller said healthcare is one of the industries that can benefit most from the adoption of cloud and AI, especially in the area of image processing, which can leverage both technologies.

“It is clear that software will ultimately be better than a human when it comes to image analysis, as it can look for more patterns, has access to more data, doesn’t have a bad day, doesn’t need to sleep and never goes on vacation,” Mueller explained. “Security and privacy are a key concern though, so it’s good to see Google’s new offering address both concerns with its advanced security capabilities. The next thing is to see how uptake in the industry will be. The world needs better next-generation applications for better patient outcomes.”

Google says it’s paving the way for hundreds of hospitals that have been held back by security and compliance concerns, to finally adopt AI for medical imaging. They include New Jersey-based Hackensack Meridian Health, which has deployed Google’s tools to de-identify petabytes of images. Its plan is to create a vast pool of anonymous training data that will be used to create new AI models for predicting metastasis in prostate cancer patients — a life-threatening condition that disproportionately affects Black men in the U.S.

“We are working toward building AI capabilities that will support image-based clinical diagnosis across a range of imaging, and be an integral part of our clinical workflow” said Sameer Sethi, HMM’s senior vice president and chief data and analytics officer. “Google Cloud’s imaging capabilities, including standardized storage and de-identification, are helping us unlock the value of our imaging data so clinicians and researchers are equipped with digitized decision support that fits into their clinical workflow.”

By Mike Wheatley

Previous
There is no more Previous post
Back to news page
Next
There is no more Next post
Back to news page