AI-generated digital humans and organs accelerate clinical trials, transforming medical research and treatment development.
How Digital Twins Work
A digital twin heart beats like a real one but lacks blood or a physical body. Adsilico creates these virtual hearts to test cardiovascular devices, like stents and prosthetic valves. Instead of making one model, Adsilico uses AI and large datasets to generate many heart models.
These AI-generated hearts consider differences in age, weight, gender, blood pressure, and ethnicity. Clinical trials often miss these differences, but digital twins fill this gap. Device manufacturers now run more inclusive and diverse trials. “This approach captures diverse anatomies and physiological responses,” says Adsilico’s CEO, Sheena Macpherson. This process creates safer, more inclusive medical devices.
A 2018 investigation revealed 83,000 deaths and 1.7 million injuries were linked to medical devices. Macpherson believes AI-powered digital twins can reduce these numbers. “Safer devices need thorough testing, but clinical trials are expensive,” she says. Virtual testing reduces costs and expands testing scope. “A virtual heart can be tested under different blood pressures or disease progressions,” she adds.
Digital twins enable testing on subgroups underrepresented in trials, like women and marginalized communities. Adsilico’s AI models use cardiovascular data and MRI and CT scans from consenting patients. Detailed anatomical data helps create precise models, allowing accurate testing of devices on diverse anatomies.
Tests involve inserting a virtual device into a digital twin heart within an AI simulation. Thousands of simulations run quickly, unlike human and animal trials, which involve only hundreds of participants. This approach speeds testing and increases analysis.
Cutting Costs and Boosting Success
Drug manufacturers are using digital twins too. Sanofi, a leading pharmaceutical firm, aims to cut testing time by 20% and boost drug success rates. Sanofi uses biological data from people to create AI-simulated patients. These virtual patients diversify trial populations, enhancing the relevance of results.
Sanofi’s AI models simulate drug properties and predict how drugs interact with human biology. The technology mimics real trials, offering early insights into drug performance. “With a 90% drug failure rate, a 10% success boost could save $100 million,” says Matt Truppo, Sanofi’s head of research platforms. He believes AI-powered digital twins will help tackle complex diseases.
Digital twins have limits. Charlie Paterson from PA Consulting warns AI models depend on training data. Many datasets are outdated or lack diversity, which could introduce bias into simulations. To address this, Sanofi supplements its internal data with external sources like health records and biobanks.
Despite these challenges, Adsilico’s Macpherson hopes AI digital twins will end animal testing in clinical trials. “A virtual human heart is more realistic than those of dogs, cows, or pigs,” she says. This shift could lead to safer, more ethical medical research. AI and digital twins are making drug discovery and device testing faster, safer, and more inclusive.
