Artificial Intelligence is Revolutionizing Life Sciences
Although artificial intelligence (AI) draws a lot of attention in the consumer engagement space, it’s also poised to make a dramatic impact in life sciences. AI for life sciences is becoming particularly relevant due to several trends that are converging and bring new challenges and opportunities for which AI technologies are ideally suited. These trends include precision medicine, improved treatment safety and efficacy evaluations, the increasing complexity of scientific questions, and the explosion of data from wearable and implantable devices.
Meeting the challenges and developing the opportunities all require data analytics on a massive scale. And that means AI. For example, Boston biotech firm FDNA and its recently formed Genomics Collaborative are using AI and deep learning technologies to analyze genotype and phenotype data with the goal of identifying physiological relationships that could lead to new drug targets.
AI for Life Sciences: New Drug Discovery
Life sciences companies face tremendous pressure to improve the speed and productivity of their research efforts. Moreover, in the past couple decades, many already-large pharma companies have grown even larger via mergers and acquisitions.
Of course, this growth produces economies of scale in many areas. But, in terms of IT infrastructure and data sources, it often leaves the resultant organization struggling with duplicate systems and a fragmented data environment. Instead of just one electronic lab notebook (ELN), it has one from each of the acquired companies. And this scenario could be repeated hundreds of times across other systems and the data repositories that support them.
So, researchers tasked with looking into promising compounds for new drugs must first navigate a labyrinth of internal data sources — and their differing access permissions — as well as relevant external sources such as MEDLINE (Medical Literature Analysis and Retrieval System Online).
Moreover, they may have the same problem finding other experts who work with these compounds or related genes, proteins, methods of action, or disease pathways. A search platform powered by AI technologies such as machine learning and natural language processing (NLP) would go a long way toward getting over these initial hurdles.
The same holds true for a research effort aimed at finding off-label uses for FDA-approved drugs. Mountains of potentially relevant data exist, but no organization can afford the time or personnel resources to sift through it manually. Properly trained, machine learning algorithms can do it in seconds. And with an NLP-enabled interface, researchers can ask questions of the data rather than figuring out how to structure a formal query. Think of it as a Socratic dialogue where AI does the heavy lifting.
To learn more about AI for life sciences and how it can accelerate drug discovery, download our solution brief, Attivio Cognitive Search for Drug Discovery.