Healthcare revolution with Microsoft Azure: An AI wellness check

Generative AI has emerged as a game-changer in various industries, and healthcare is no exception. With its ability to generate new content, models, and insights, generative AI has the potential to revolutionize medical research, diagnosis, treatment, and patient care by allowing healthcare providers to increase productivity with less administrative burden while shifting their focus to patient care. 

Azure OpenAI Service and Epic’s EHR software 

In April 2023, one of the largest electronic health records providers, Epic, used Azure OpenAI Service to integrate large language model tools and AI into its electronic health record software. Epic currently has the largest share of acute care hospitals in the U.S. market: Globally, about 2,130 hospitals use Epic for its medical records software and more than 305 million patients have a current electronic record in Epic. 

One of the integration’s tools, which helps automatically draft message responses, is already being implemented at UC San Diego Health, UW Health in Madison, Wisconsin, and Stanford Health Care.   

“The urgent and critical challenges facing healthcare systems and their providers demand a comprehensive approach combining Azure OpenAI Service with Epic’s industry-leading technology.”

Eric Boyd, Corporate Vice President, AI Platform, Microsoft.

Epic’s exploration of OpenAI’s GPT-4 with Azure OpenAI Service has shown the potential to increase the power and accessibility of self-service reporting. By doing so it’s made it easier for healthcare organizations and their providers to identify operational improvements, including ways to reduce costs and to find answers to questions both locally and within a broader context. And that’s just the beginning. 

AI influencing the future of healthcare

Below we look at more ways that AI is expected to help the healthcare industry pave the way to wellness:

• Accelerating drug discovery: One of the most time-consuming and costly aspects of healthcare is the discovery and development of new drugs. One way to speed up the drug-development process is through computational modeling so that molecules can be prioritized in silico without being physically available, and only the molecules most likely to succeed are synthesized and measured. To enable such a speedup through computational modeling, a machine learning model must be able to precisely predict molecular properties, and whether a proposed drug molecule will be able to affect the protein target associated with the disease. Machine learning is highly effective at recognizing patterns in images and text, where millions of lines of such data are available, thereby accelerating the time it takes to produce drugs that can successfully combat illness.
• Enhancing medical imaging analysis: Medical imaging plays a crucial role in supporting the diagnosis, treatment, and monitoring of various diseases. Project InnerEye from Microsoft Research is developing machine learning and open-source software to empower healthcare organizations and innovators to develop their own solutions that can assist clinicians in planning radiotherapy treatments so that they can spend more time with their patients. Cambridge University Hospitals NHS Foundation Trust is one of the early adopters to use open-source technology from Project InnerEye, creating an Azure-based medical AI tool called OSAIRIS that reduces the amount of time cancer patients wait for radiotherapy treatment. Working alongside OSAIRIS, the specialist can plan radiotherapy treatments about two and half times faster than working alone, ensuring that more patients who need treatment can get it sooner and thus improving the likelihood of better outcomes.
• Supporting pathologists, and their patients, around the globe: PathPresenter, in collaboration with Microsoft, has focused on ensuring seamless interoperability between scanners, image management software, AI models, and hospital infrastructure, to reduce the reporting burden on pathologists and accelerate the adoption of digital workflows by pathologists and institutions worldwide for the benefit of patients and society. Microsoft Azure offers unique solutions in digital pathology that are scalable and can serve digital pathology images several gigabytes in size—even on poor internet connections encountered in remote regions. 
• Equip more clinicians with medical imaging AI: AI has the potential to revolutionize medical imaging by enabling faster and more accurate analysis of imaging data, leading to better patient outcomes. Together, Nuance + Microsoft, and NVIDIA are working to simplify the translation of imaging AI models into existing and trusted clinical applications that can deliver genuine benefits for everyday patient care without requiring providers to change their workflows or their underlying IT systems.
• Improving COVID-19 clinical decision support: Providence, a healthcare system with 51 hospitals and more than 1,085 clinics, is headquartered in Renton, Washington, near the epicenter of the first major US COVID-19 outbreak. Providence used the existing Microsoft Azure Health Bot service and configured it to create an AI-based tool to triage patients and answer their questions specifically about COVID-19 symptoms, freeing providers to attend to the patients who needed it most. The Azure Health Bot with COVID-19 templates that Providence deployed has since been adopted by several thousand healthcare providers, the US Centers for Disease Control and Prevention (CDC), NGOs (non-governmental organizations), and international health authorities.
• Scalable Data Storage: Health First was one of the first healthcare providers to deploy WhereScape with Azure Synapse Analytics. Implementing both familiar and new products helped the network to accelerate its operations. With faster turnaround times, Health First employees could focus on using data to improve patient care and operational decisions. Health First experienced more than a 90 percent improvement in workload processing times. With Azure Synapse Analytics and Power BI, the daily data refresh was about 75 percent faster, down to 3 hours from a 12-hour overall run time, which improved the company’s capability to provide actionable insights for both clinical and operational decisions.

As the use of AI in healthcare continues to evolve, we anticipate a future where healthcare systems are increasingly able to handle more challenging cases and discover solutions to some of the most pressing healthcare issues facing individuals and communities worldwide today. Microsoft’s innovative initiatives in this space highlight the immense potential of AI. We’re looking forward to helping accelerate drug discoveries, enhance medical imaging analysis, and scale to better support operational decisions. 

Our commitment to responsible AI

Microsoft has a layered approach for generative models, guided by Microsoft’s responsible AI principles. In Azure OpenAI Service, an integrated safety system provides protection from undesirable inputs and outputs and monitors for misuse. In addition, Microsoft provides guidance and best practices for customers to responsibly build applications using these models and expects customers to comply with the Azure OpenAI Code of Conduct. With GPT-4, new research advances from OpenAI have enabled an additional layer of protection. Guided by human feedback, safety is built directly into the GPT-4 model, which enables the model to be more effective at handling harmful inputs, thereby reducing the likelihood that the model will generate a harmful response. 

Source: microsoft.com