THANK YOU FOR SUBSCRIBING
At the end of the 20th century the words “artificial intelligence (AI)” elicited strongly negative connotations associated with the unrealized utopia of AI assisted galactic exploration and humanoid robots taking over mundane daily tasks. We are in a different world now. Over the last few years, artificial intelligence not only put itself on the map in high tech but is also being talked about at the top of the automotive, financial and retail industries. It now begins to gain traction in healthcare and biopharmaceutical research. Luckily, the proliferation of personal assistants, household robotic appliances, and popularization of self-driving cars have shifted our mindset and attracted funding into AI applications in a variety of areas. This new dynamic makes me believe artificial intelligence will stay with us moving forward.
The pharmaceutical industry is well known for its conservatism and regulated environment - understandably so, given the risk for severe adverse intervention effects or a lethal treatment outcome. However even in the pharma world, I see a steady rise of interest in cutting edge technologies including AI based applications and products. A good example of this is an AI based medical device developed for accurate detection of diabetic retinopathy and recently approved by the FDA. Machine learning applications in healthcare image analysis are exploding with a sizable number of start-ups funded with future products in radiology, pathology and surgery where the AI based analytics can expedite patient care and make detection of disease relevant pathology more accurate.
"The proliferation of personal assistants, household robotic appliances, and popularization of self-driving cars have shifted our mindset and attracted funding into AI applications"
What about the pharma R&D fundamentals? A handful of companies have been using AI methodologies in target discovery by analyzing vast amounts of data. Some of these groups have approaches relying on publically available data such as scientific papers and shared molecular and assay data, others developed algorithms and platforms to identify potential therapeutic targets from “scratch” by working with complex in-vitro disease models and using AI to actualize and productize high-throughput molecular and phenotypic data. These are challenging scientific problems and the solutions are still largely in the stage of validation. However, if successful, these pioneering organizations will have an opportunity to disrupt the lengthy and expert driven pharma discovery process. Downstream in the R&D pipeline, significant progress is apparent in drug design. AI approaches are making great strides in medicinal and structural chemistry where pattern recognition software can learn known docking mechanisms and predict novel interactions between small and large molecules. These approaches create a tremendous opportunity to speed up the drug discovery process by making it more comprehensive and exhaustive in a highly automated fashion, and hence, more efficient.
Precision medicine is another rapidly developing area well positioned to benefit from the expansion of the AI enabled tools. Analyzing myriads of data points including molecular and digital health information and integrating this disparate information into predictive signatures is a challenging problem. AI based software allows us to analyze an extremely large feature space and pinpoint specific risk factors associated with or driving patient outcomes. AI effectively is a technology link connecting real world patient data, clinical development observations and multi-modal molecular data into a patient management information system. The near future holds an immense promise for the AI application in precision medicine and next generation diagnostics.
Looking back at the World Medical Innovation Forum organized by Partners HealthCare a few weeks ago, it is apparent the AI disruption has begun across healthcare. Rigorous scientific and clinical validation work is ongoing in research and development. These efforts will define the future of the AI adoption in the pharmaceutical industry in the near term.