by Patrick Howie, Founder and CEO of MediFind, and author of “The Evolution of Revolutions: How We Create, Shape, and React to Change“
Medical knowledge doubles approximately every 73 days which means healthcare professionals constantly need updated information to keep up with best practices. So how do they stay on top of research when they have many competing priorities?
Doctors deploy a variety of tactics to stay updated on their field such as specializing in a particular area, reading key publications, attending conferences and conducting regular check-ins with their peers. Unfortunately, while these practices are effective, they are incomplete. As a result, and despite their best efforts, it’s nearly impossible for healthcare professionals to stay current on all the advances in their field.
Fortunately, modern technology provides a key tool to address the ever-widening gap between massive amounts of information and the need to know what’s relevant for a given specialty. The solve: artificial intelligence (AI). When deployed properly, AI can offer significant aid to healthcare providers and patients. Here’s how:
Real-time meta-analysis is an ambition shared by many in the healthcare community, and AI tackles the first step in achieving this goal. By creating a real-time view of all the treatments being offered for the 10,000+ known diseases, patients and healthcare experts now have the ability to identify when a new approach is being used to treat a condition and what treatments are most successful – which can lead to life-saving results.
Identifying drug-repurposing opportunities.
By combining a comprehensive understanding of what treatments have been successfully used to treat each of the 10,000 diseases with a knowledge of the underlying similarities across those diseases, we have the ability to quickly identify drug repurposing opportunities.
We’re seeing in real-time how existing treatments can be used to treat patients with the current pandemic, as nearly 200 existing treatments are being studied to fight COVID-19. While this response is unprecedented in scope, we cannot expect that level of activity for other diseases. However, we have the opportunity to use AI to leverage a similar approach to identifying potential treatments for rare diseases, of which 95% have no approved treatment.
More sophisticated clinical trial matching.
Clinical trials have traditionally been a manual process, but as we get smarter about precision medicine, like studying a specific variant of cancer based on mutation type, clinical trials will become more data-intensive than ever.
Using data to match qualified patients with studies will increasingly end the physical boundaries associated with clinical trials – facilitated in part to COVID-19 necessitating more virtual trials.
However, achieving these outcomes requires overcoming a variety of obstacles and challenges – the biggest being the complexity of medical literature. With thousands of existing medical conditions, new ones being continuously identified and the ever-evolving terminology related to diseases, even the most advanced AI models will struggle to truly make sense of the information without the help of medical experts. By having medical experts extract the key information from the research such as the treatment used, the number of patients studied, and most importantly, the outcome, we can use AI to replicate and scale the thinking of a medical expert.
Using AI to help doctors identify the best course of treatment across thousands of disparate diseases is an ambitious approach that requires an evolving process of teaching, reviewing and revising an AI algorithms to “think” like a doctor. The goal is to accelerate the incorporation of new research into treatment decisions as quickly as possible, increasing the speed at which these discoveries move from the lab to the frontlines of medicine – patient care.
Patrick Howie is the CEO and Founder of MediFind – which he started after watching his brother struggle to navigate the healthcare system to treat his rare cancer. Howie is the former head of Global Analytics at Merck and has held senior leadership positions in multiple healthcare startups. He is also the author of “The Evolution of Revolutions: How We Create, Shape, and React to Change“.