Applying artificial intelligence to medical images can benefit doctors and patients, but developing the tools to do so can be challenging. Google announced on Tuesday that it was ready to meet this challenge with its new medical imaging suite.
“Google pioneered the use of artificial intelligence and computer vision in Google Photos, Google Image Search and Google Lens, and we are now making our expertise, tools and imaging technologies available healthcare and life sciences companies,” said Alissa Hsu Lynch, global head. of Google Cloud MedTech Strategy and Solutions, said in a statement.
Gartner Vice President and Distinguished Analyst Jeff Cribbs explained that healthcare providers seeking AI for diagnostic imaging solutions have typically been forced to make one of two choices.
“They can source software from the device manufacturer, image repository provider, or a third party, or they can create their own algorithms with industry-neutral image classification tools,” he told TechNewsWorld.
“With this release,” he continued, “Google takes its low-code AI developer tool and adds substantial healthcare-specific speedup.”
“This Google product provides a platform for AI developers and also facilitates the exchange of images,” added Ginny Torno, administrative director of innovation and clinical, ancillary and research computing systems at Houston Methodist, in Houston.
“It’s not unique to this market, but it can provide interoperability opportunities that a smaller vendor isn’t capable of,” she told TechNewsWorld.
According to Google, Medical Imaging Suite addresses some common issues that organizations face when developing AI and machine learning models. Suite components include:
- Cloud Healthcare API, which enables easy and secure data exchange using an international imaging standard, DICOMweb. The API provides a fully managed, scalable, enterprise-grade development environment with automated DICOM de-identification. Imaging technology partners include NetApp for seamless on-premises data management to the cloud, and Change Healthcare, a cloud-native enterprise imaging PACS used in the clinic by radiologists.
- AI-assisted annotation tools from Nvidia and Monai to automate the highly manual and repetitive task of labeling medical images, as well as native integration with any DICOMweb viewer.
- Access to BigQuery and Looker to view and search petabytes of imagery data to perform advanced analytics and create training datasets without operational overhead.
- Use Vertex AI to accelerate the development of AI pipelines to build scalable machine learning models, with 80% fewer lines of code needed for custom modeling.
- Flexible options for cloud, on-premises, or edge deployment to enable organizations to meet diverse sovereignty, data security, and privacy requirements, while providing centralized management and policy enforcement with Google Distributed Cloud, enabled by Anthos.
Complete technology platform
“One of the key differentiators of Medical Imaging Suite is that we offer a comprehensive suite of technologies that support the AI delivery process from start to finish,” Lynch told TechNewsWorld.
The suite provides everything from imagery data ingestion and storage to AI-assisted annotation tools to flexible model deployment options at the edge or in the cloud, she said. Explain.
“We provide solutions that will make this process easier and more efficient for healthcare organizations,” she said.
Lynch added that the suite takes an open and standardized approach to medical imaging.
“Our integrated Google Cloud services work with a standard DICOM approach, allowing customers to seamlessly leverage Vertex AI for machine learning and BigQuery for data discovery and analysis,” she said.
“By having everything built around this standardized approach, we make it easier for organizations to manage their data and make it useful.”
Image classification solution
The growing use of medical imaging, coupled with labor issues, has made the field ripe for solutions based on artificial intelligence and machine learning.
“As imaging systems become faster, offer higher resolution and capabilities such as functional MRI, it is more difficult for the infrastructure supporting these systems to keep up and, ideally, maintain a length ahead of what is needed,” said Torno.
“Additionally, there are labor shortages in radiology that complicate the personal side of workloads,” she added.
Google Cloud aims to make medical imaging data more accessible, interoperable and useful with its medical imaging suite (Image credit: Google)
She explained that the AI can identify issues found in an image by comparing it to a set of learned images. “He can recommend a diagnosis which then only needs interpretation and confirmation,” she noted.
“It can also cause images to appear at the top of a work queue if a life-threatening situation is detected in an image,” she continued. “AI can also organize workflows by reading images.”
Machine learning is doing for medical imaging what it did for facial recognition and image-based research. “Rather than identifying a dog, Frisbee or chair in a photograph, AI identifies the tumor boundary, bone fracture or lung injury on a diagnostic image,” Cribbs explained.
Tool, not substitute
Michael Arrigo, managing partner of No World Borders, a national network of expert witnesses on health care issues based in Newport Beach, Calif., agreed that AI could help some overworked radiologists, but only if it’s reliable.
“Data needs to be structured in a way that it can be used and consumed by AI,” he told TechNewsWorld. “AI doesn’t work well with highly variable unstructured data in unpredictable formats.”
Torno added that many studies have been done on the accuracy of AI and will continue to be done.
“While there are examples of AI finding things a human hasn’t, or being ‘just as good’ as a human, there are also examples where the AI is missing something. important, or not sure what to interpret because there could be multiple issues with the patient,” she observed.
“AI should be seen as an efficiency tool to speed up image interpretation and help with emergent cases, but not to completely replace the human element,” she said.
Big splash potential
With its resources, Google can have a significant impact on the medical imaging market. “Having a major player like Google in this space could facilitate synergies with other Google products already in place in healthcare organizations, potentially enabling more seamless connectivity to other systems,” Torno noted.
“If Google is focused on this market segment, they have the resources to make a splash,” she continued. “There are already a lot of players in this space. It will be interesting to see how this product can take advantage of other Google features and pipelines and differentiate itself. »
Lynch explained that with the launch of Medical Imaging Suite, Google hopes to help accelerate the development and adoption of AI for imaging by the healthcare industry.
“AI has the potential to help ease the burden on healthcare workers and dramatically improve and even save lives,” she said.
“By offering our imaging tools, products and expertise to healthcare organizations, we believe the market and patients will benefit,” she added.