Google Cloud announced new generative AI advancements for healthcare and life science organizations.

Amanda Pedersen

March 13, 2024

5 Min Read
Concept image showing a doctor using digital health technology and AI healthcare applications
Image credit: pcess609 / iStock via Getty Images

At a Glance

  • Google Cloud launched Vertex AI Search for Healthcare this week at HIMSS24
  • The company introduced a new consumption-priced managed service of its Healthcare Data Engine
  • Google introduced HDE Data Mapper, a new low-code graphical integrated development environment purpose-built for healthcare

Google Cloud unveiled several new solutions at HIMSS24 intended to help healthcare and life sciences organizations enable interoperability, build a better data foundation for their businesses, and deploy generative AI tools to improve patient outcomes.

Vertex AI Search: smarter search for healthcare data

Generative AI (gen AI) solutions can empower healthcare staff and life sciences employees to work more efficiently and effectively, according to Google. The company cites increasing healthcare administrative costs, physician burnout, and the ongoing nursing shortage as reasons such tools are needed in the industry.

Launched this week at HIMSS, Google Cloud’s Vertex AI Search for Healthcare is designed to help developers build better assistive technology for clinicians and other health system workers to alleviate administrative burden. Specifically, it enables medically tuned, gen AI search on a broad spectrum of data, including FHIR data and clinical notes. These search and question-answering capabilities now integrate with MedLM, Healthcare Data Engine (HDE) and Cloud Healthcare FHIR APIs, making it easier for healthcare and life sciences organizations to build the data analytics and AI solutions needed for next era health systems.

Vertex AI Search for Healthcare features include:

  • Medically tuned search: By enabling medically tuned search on electronic health record (EHR) data, scanned documents, and other clinical data, healthcare applications can now surface the most relevant information to clinicians and others engaged in decision-making and processing without the need to click through hundreds of pages of notes or toggle between different screens and applications.

  • Configurable cloud APIs: Healthcare application developers can integrate the medically tuned search directly into the clinician workflow tools using configurable cloud APIs.

  • Question-answering: Clinicians can now use the service to answer questions such as “Does the patient have a history of cancer?” which typically involves not only an understanding of medical terminology related to cancers but also understanding the nuances of language in medical notes and scanned documents.

  • Factuality: Because the results are grounded in the healthcare organizations’ real data, Vertex AI Search for Healthcare can also cite and link to the specific data points that generated the answers, providing transparency to users, increasing confidence in the responses, and helping reduce the risks of hallucinations or inaccurate responses.

  • Data platform integration: Gen AI and search applications require high-quality data to ensure that the outputs of gen AI are useful. HDE provides automated tooling and components to build a unified, high-quality patient data platform quickly and easily. The out-of-the-box integration of Vertex AI Search with HDE enables search and summarization use cases without the need for custom integrations.

  • Deeper medical understanding: While MedLM tools are experts in complex medical topics, Vertex AI Search for Healthcare delivers medically tuned search capabilities grounded in the patient record. When combined, healthcare organizations can find the most relevant answers to complex medical questions both from MedLM as well as within the patient’s medical record. Select customers will have early access to Vertex AI Search for healthcare and life science integration with MedLM.

“Not all generative AI is created equal, and in healthcare, the stakes are particularly high,” said Aashima Gupta, global director for healthcare strategy and solutions, Google Cloud. “Healthcare organizations require enterprise-grade gen AI solutions, grounded in real data. Vertex AI Search for Healthcare is already making a difference for healthcare organizations by helping ensure clinicians have the right information and insights at the right time to inform decisions and improve the overall quality of patient care.”

Healthcare Data Engine: healthcare data platform for the AI era

To help healthcare organizations around the world build an interoperable, high-quality data platform, the foundation to take advantage of gen AI, Google Cloud announced a new consumption-priced managed service of HDE and expanded availability internationally with new features. Key updates include:

Simplified management and streamlined pricing: Customers can now deploy HDE as a consumption priced, pay-as-you-go, managed service, which opens the product to more healthcare organizations, helping them to deploy, build, and manage a near real-time healthcare data platform in the cloud.

HDE will expand beyond North America, and is expected to roll out globally to most cloud regions in Europe and in the Asia Pacific.

Low-code graphical data mapping IDE 

By introducing HDE Data Mapper, a new low-code graphical integrated development environment (IDE), purpose-built by Google Research for healthcare, customers can easily transform their data to build high quality longitudinal patient records in FHIR format to power gen AI applications. 

Foundation for AI and analytics solutions: By integrating HDE with Vertex AI Search for Healthcare, clinicians can look across multiple systems and formats in one search, saving valuable time that can instead be focused on patient engagement and improving the overall patient experience. Integration with MedLM will allow customers to answer complex questions grounded in the patient’s data.

MedLM and more: Google Research innovations coming to Google Cloud customers

Finally, MedLM, a family of foundation models fine-tuned for healthcare industry use cases, will add two new capabilities for Google Cloud customers to explore and test.

First, MedLM for Chest X-ray can help with classification of chest x-rays for operational, screening, and diagnostics use cases. It is a domain-specialized model, launched as an application programming interface (API) that converts chest x-ray images into embeddings. App developers and data scientists can use those embeddings, along with ground truth labels, to train a simple classification model in Vertex AI. 

The second new capability is a task-specific API called Condition Summary that aims to provide a chronological list of patient conditions, along with AI-generated briefs about each condition, with citations from original text.

In the coming months, Google Cloud is planning to bring additional functionality to the MedLM suite to offer even more capabilities, the company noted.

Google Cloud’s customers retain control over their data. In healthcare settings, access and use of patient data is protected through the implementation of Google Cloud’s infrastructure and secure data storage that support HIPAA compliance, along with each customer’s security, privacy controls, and processes.

About the Author(s)

Amanda Pedersen

Amanda Pedersen is a veteran journalist and award-winning columnist with a passion for helping medical device professionals connect the dots between the medtech news of the day and the bigger picture. She has been covering the medtech industry since 2006.

Sign up for the QMED & MD+DI Daily newsletter.

You May Also Like