Accelerating Academic Medical Research with an AI-Driven Data Strategy

Academic medical centers (AMCs) are a critical keystone of healthcare systems worldwide. They serve as major hubs of medical research, pioneering new treatments that advance and set the standard of care throughout medicine. They also educate and train the next generation of healthcare professionals, ensuring that the medical field continues to advance. In the U.S. alone, there are more than 230 active AMCs, and a significant number are part of a health system. Patients who receive care at AMCs are more likely to receive the most up-to-date therapies and treatments.

But AMCs are encountering significant obstacles. According to a recent opinion piece from AMC leaders at Emory University and Mass General Brigham, “Academic medical centers in the U.S. are struggling with a mounting, long-term financial crisis that … will negatively affect discovery, innovation, education, and these centers’ ability to provide quality care.” They cite several contributing factors to ongoing deficits at AMCs, including reduced private insurance reimbursement, gaps in government payments for care and increased labor costs. 

At the same time, there’s no shortage of opportunities for AMCs to grow as the healthcare industry expands. New biomedical advancements, such as the clinical validation of CRISPR technology and the rise of biomaterials, are providing AMCs with new tools and methodologies to improve health outcomes. Also, the federal government is increasing healthcare funding, including $300 million in new funding across the National Institutes of Health (NIH) and an increase of $1.3 billion in funding for health centers. 

To overcome the challenges and capitalize on the opportunities, AMCs must find ways to accelerate medical research, secure healthcare funding in a highly competitive landscape and create new revenue streams. To accomplish these goals, AMCs need a modern data strategy that includes a robust, cloud-based data foundation and capabilities such as generative AI, data and AI applications, easy access to third-party data, and seamless data sharing and collaboration. 

The ability to align research data with universally accepted data standards, such as the OHDSI OMOP Common Data Model, is a strategic imperative for AMCs. By standardizing data, they can easily share information with other organizations, enhancing collaborative research efforts and enabling them to participate in larger and more competitive funding opportunities. They can power collaboration internally between the clinical practice and research departments, as well as externally with research partners. Interoperability and transparency, which can lead to broader scientific and clinical advancements, are also key criteria for funding agencies.

However, on-premises data platforms and legacy technologies create data silos and cumbersome data-sharing processes, making it difficult for AMCs to standardize and collaborate on data. These platforms lack the scalability needed to manage the extensive data volumes and multiple concurrent research workloads essential for AMCs. They also cannot easily collect, process or share multimodal health data, which encompasses a wide variety of data types — including clinical notes, protein sequences, chemical compound information, medical imaging and patient data. Each of these data types can require specialized software packages, hardware environments and data processing techniques. The lack of a simplified, scalable technology platform leads to delays in research, wasted staff time and lost funding opportunities. 

Critically, legacy systems can also create serious data privacy and security issues. AMCs must ensure data remains private and secure across the entire research lifecycle to comply with regulations such as HIPAA. But outdated software may not be compatible with modern security protocols or meet regulatory standards for data protection, leaving AMCs vulnerable to data breaches and noncompliance penalties. The complexity of legacy systems also makes it challenging to consistently monitor and manage security as well as preserve privacy. 

AMCs with a modern data strategy can meet these challenges and seize the opportunities. The foundation of this strategy includes a modern, fully managed data platform that enables interoperability, collaboration, scalability, data privacy and security. 

Modern data platforms can integrate various data sources, allowing AMCs to collect multimodal health data, such as clinical, genomic and imaging data, into a single, governed repository. They provide tools to align the data with open community data standards. With centralized data in standardized formats, researchers can more easily share and access data, facilitating collaboration with internal partners, such as hospitals, as well as with other institutions. 

Modern cloud data platforms can also scale to support diverse data volumes and multiple, highly computational research workloads in parallel, without having to compete for resources and cause delays or wastage.

Importantly, a fully managed cloud-based data platform includes data governance, enabling AMCs to manage their data according to applicable standards and regulations, which is crucial to maintaining healthcare data privacy and security. 

AMCs should ensure their modern data strategy includes a few additional capabilities, including: 

Generative AI: Gen AI technology uses large language models (LLMs) and large vision models (LVMs) that are trained on vast amounts of text and image data sets. AMCs can use gen AI to analyze large data sets, including clinical images and medical literature, and reveal data patterns. Generative AI can also create new content, including text, images, video and audio. This allows researchers to focus on more complex tasks and accelerates the pace of discovery. 

Data and AI applications: Revolutionary data and AI applications that use gen AI are poised to dramatically enhance the research capabilities of AMCs, unleashing unprecedented potential for scientific discovery. These apps enable the integration of genomic information and clinical data; the exploration and analysis of the microbiome; and data collection and analysis from biomedical sensors.  

Data marketplace: A modern data marketplace can facilitate collaboration with outside research entities that can help drive research outcomes. It features live, ready-to-use data that can be easily integrated without relying on APIs. These data sets encompass a wide range of fields, such as social determinants of health, insurance claims and environmental data. Additionally, a data marketplace can provide AMCs with the capability to share their data securely and legally, adhering to data privacy regulations. 

To learn more about employing a modern data strategy that sets your AMC up for success, join us on August 7 for our Accelerating Clinical Research: Harnessing Multi-Modal Data, AI and Secure Collaboration webinar or download our 4 Ways Snowflake Accelerates and Augments Healthcare Research ebook.