AI together now

Generative AI – not without interoperability

We’ve seen the topic of generative AI overshadow almost every conversation surrounding innovation and development across life sciences in the past couple of years.

Generative AI holds immense potential in revolutionising drug discovery, disease diagnosis, treatment optimisation and personalised medicine.

But its maximum potential will not be fully realised without genuine interoperable systems and the seamless flow of data.

Generative AI has the capacity to analyse vast data sets, simulate biological processes and generate novel hypotheses, meaning that the technology is poised to unlock new frontiers in biomedical research and healthcare delivery.

Its promise is for more effective clinical trials, greater preclinical collaboration, post-market surveillance and for a more comprehensive understanding of patient populations and their responses to different treatments.

But this will only be truly achievable with greater interoperability, and this is something the industry is coming to understand.

Back in May 2021 the Office of the National Coordinator for Health Information Technology (ONC) launched Health Interoperability Outcomes 2030, engaging stakeholders to outline achievable and measurable ‘Interoperability Outcome Statements’ for the year 2030.

With over 700 submissions from voices across the industry, the study offered a unique perspective.

From empowering individuals in their own homes with access to their electronic health records (EHR), to fostering collaboration among healthcare providers for seamless care coordination, interoperability serves as the cornerstone of future healthcare delivery.

The subsequent ‘trusted exchange framework’ and ‘common agreement’ pledged to establish a universal governance, policy and technical floor for nationwide interoperability, simplifying connectivity for organisations to securely exchange information.

It’s clear that the industry is committed to harnessing the full potential of advances in technology, but also understands that it has to be underpinned by interoperability.

Greater interoperability will have a significant impact on clinical trial operations, breaking down data silos known to hinder collaboration among researchers, clinicians, pharmaceutical companies and healthcare institutions.

For instance, throughout the site location identification and patient recruitment process, a reliance on guesswork and suboptimal recruitment strategies will be eliminated.

Instead data and advanced analytics can be deployed to proactively identify the best locations and target evidence-based catchment areas.

This prevents costly delays that have been known to cost organisations nearly $8 million per day. For some organisations, a delay of just one day at these eye-watering costs is the difference between delivering a life-saving therapy to market or shutting down the business.

To enable better forecast trial enrolment and planning, data from historical studies can be used to forecast how best to deliver enrolment plans against pre-agreed milestone deadlines.

Similarly, when it comes to reviewing the data obtained from trials, advanced analytics platforms such as SAS Viya can better manage data, develop models and deploy insights, rapidly freeing up hours of staff time.

Despite its undeniable importance and a clear commitment to change, achieving interoperability in the complex landscape of life sciences remains a formidable challenge.

While there’s no denying the size of the challenge, requiring not only technological solutions but also standardised protocols, governance frameworks and stakeholder collaboration, the benefits will be undeniable.

They are also the key to unlocking the full potential of generative AI.

Kayt Leonard is Global Health and Life Sciences Advisor at SAS. Go to sas.com