Solving The Data Problem For AI In Drug Discovery
While artificial intelligence has proven its value in drug discovery, for most companies, the power of their AI systems is only as strong as the data those systems are trained on. However, stakeholders – from individual companies to consortiums and service vendors -- are finding creative approaches to overcome the so-called data problem and strengthen their AI models.
You may also be interested in...
Eric Topol says there is not enough regulatory teeth to require companies to share data with the medical community. Experts at a workshop on the application of artificial intelligence and machine learning for precision medicine also want data and algorithms to be accessible.
AI is beginning to transform biologics discovery. The power of algorithms used in biologics discovery has increased over the last decade, and today between 50 and 60 AI-enabled biologics are in different stages of discovery, preclinical and clinical development. We expect the number of AI-enabled biologics to continue to grow rapidly, driven by advances in AI technology and algorithms, growing computing power, increasing availability of data, and evolving discovery workflows. We show that the volume of data used for training algorithms in biologics discovery is increasing exponentially over time, a trend reminiscent of Moore’s Law in computer technology.
Janssen’s Troy Sarich outlines why it’s hard to emulate randomized controlled trials with real-world evidence studies, emphasizing that the two are “not in competition.” He also highlights the huge strides made by AI-driven technology firms to provide “research-ready” structured data and new game-changing advances in the area of health sensors.