The adoption of Artificial Intelligence and Machine Learning techniques in the pharmaceutical and biotechnology industry has been limited by the pervasive regulatory environment and the need to ensure data privacy on patient information. The lack of general understanding of these techniques and a reluctance to be open to the utilization of the cloud means these techniques need a champion and a clear bottom-line business value to be implemented.
Raj Dasgupta, author of Practical Data Analytics, says, “Unless the pharma machine learning project can clearly articulate what the business stands to benefit in concrete measurable terms, it is hard to get traction with such proposals.”
Where Artificial Intelligence and Machine Learning are Already Working in Pharma
According to a recent Intelligence Unit Report big data innovations reduced the time for recruitment in clinical trials by 37%. AI gives companies more efficient ways to design clinical trials and identify participants by including a much broader set of data including genetic information, social media platforms, patient population and evidence of diseases. These AI-enabled trials become more targeted, quicker and less expensive.
Marketing and Sales are actively using artificial intelligence and machine learning to improve sales field messaging by using natural language processing (another AI technique) on call notes and physician Rx records to understand increases/decreases in prescribing. AI is being used to create novel territory alignment methods and reach no-see physicians. Correlating physician and patient records from disparate datasets and patient-doctor matching provides a level of information previously unavailable to account executives.
Utilizing machine learning for fraud detection in claims data has been actively used for years. Identification of medication non-adherence and even prediction of outbreaks or signal analysis for seizures has had significant impact in patient health and insurance management.
Where Artificial Intelligence and Machine Learning Will Have the Biggest Impact
Processing Scientific Knowledge
The average researcher is dealing with a huge amount of new information every day. Drug Target Review estimates that the bioscience industry is getting 10,000 new publications uploaded on a daily basis. No researcher can process all of the scientific knowledge out there relating to their area of investigation, but artificial intelligence can correlate, assimilate and connect all of this data to help the researcher develop new drug hypotheses.
AI can dramatically reduce the amount of trial and error needed to design a drug candidate once a promising disease target has been identified. BenevolentAI, calling themselves Europe’s largest private AI company, estimates that it can cut the costs associated with drug design by 60% and reduce the development time frame from three years to one. Taking a new drug from lab to market can take a decade and an average expenditure of $2.6 billion. The savings provided by AI are incredibly impactful. Zdnet.com also found that drugs developed with AI tools were 16% more likely to reach market launch.
This is one of the gold mines for AI-based technologies to drive value for the pharmaceutical manufacturer. Because much is already known about the drug in question, firms are using this data and their library of genetic disease models to screen molecules through machine learning technology to derive promising new indications.
The more the pharmaceutical and biotechnology industry improves the standard of care, the more difficult and costly it becomes to improve further. Adopting artificial intelligence and machine learning techniques, particularly in the R&D area, can provide the edge for transformative products and services.