Where healthcare needs to focus for AI

It’s hard to believe it was only last November when ChatGPT arrived, giving almost everyone everywhere instant access to artificial intelligence. Since that milestone, AI has been a hot topic of conversation for many industries, including healthcare. 

But artificial intelligence is a broad and sometimes misunderstood term. In healthcare, there’s a very real opportunity for AI to make an impact both now and in the future, but what does that look like? We must first look at the types of AI.

The types of AI 

I like to say that the term “AI” is a toolbox and the different types of AI are the tools. Our available tools at the moment are: statistical inference, pattern recognition, machine learning, deep learning, and generative AI. 

To me, the biggest benefit of AI in healthcare is its ability to sort, organize, and analyze data. The healthcare industry is flooded with data. The data is diverse, real-time, real-world, complex, and meaningful, and it has the power to provide doctors with a complete picture for diagnosis and treatment planning. But the challenge is that as an industry, we’re still grappling with how to quickly translate all of that data into insights. THIS is where AI comes in. 

To advance healthcare, and to have the biggest impact on lives across the globe, we need to focus on the types of AI that will most efficiently sort, analyze, and share the existing data—and that is typically deep learning and machine learning AI.

When we say data, we’re talking about genomic information (DNA and RNA samples); clinical information (symptoms, biological information); imaging (X-ray, CT scans); and lab tests (blood draws). Genomic analysis on its own produces a vast amount of data, and while that data plays an essential role in identifying rare diseases and treating cancer with precision medicine, it can take weeks to analyze without AI.

The promise of AI in healthcare

My company works with hospitals, laboratories, and biopharma institutions worldwide to quickly analyze real-world data through AI. Our tools of choice from the AI toolbox are deep learning and machine learning, as those tools are best equipped to tackle diverse, real-world, real-time data. Through AI we provide highly accurate, smartly grouped insights that can be shared with healthcare institutions globally to aid clinicians in making data-driven decisions. And the more data we gather, the smarter our AI becomes, meaning a more globalized and decentralized approach yields stronger results for everyone. 

And that’s not the only use of this technology. AI can also support the biopharma industry by helping identify qualified patients (who have the biomarkers needed for a clinical trial) and thus diversify clinical trials. This use of AI may ultimately speed up the drug development process which means patients have access to more innovative and specialized therapies to treat cancers and rare diseases. 

Our company alone has helped analyze over 1.4 million genomic profiles of patients across 70 countries, and we’re not the only ones doing this work. AI is already helping arm physicians and researchers with analyzed data—much faster than traditionally possible—and this is facilitating the increased use of data-driven medicine.

When we look at AI in healthcare, there’s tremendous opportunity for it to save lives, especially with cancers and rare diseases, but we need to use the right tools at the right time. As an industry, we should be investing in AI technologies that utilize existing data to make precision medicine the gold standard of care.

This way, the data from today’s patients will help inform and treat patients of tomorrow.

Jurgi Camblong is the founder and CEO of SOPHiA GENETICS, a cloud-native software company and a leader in data-driven medicine.