As scientists, we are no stranger to skepticism, having been taught to look at everything critically. While doling out skepticism, we also come across it often when working with other healthcare leaders or fielding questions from audiences at conferences or pitching an AI approach to investors. Much of the healthcare industry is still stuck in the 20th century, and hence, it is not so surprising that new technologies such as an AI-driven approach to biopharma may be met with raised eyebrows and thought to be doomed to failure from the outset.
On the one hand, we have people saying that AI could revolutionize biopharma and help us to discover new treatment options, with Deloitte predicting that the AI/biopharma industry will be worth $3.88 billion by 2025. On the other, we have Elon Musk warning that AI could spell the end of civilization as we know it.
With the Rise of AI, What IP Disputes in Healthcare Are Likely to Emerge?
Munck Wilson Mandala Partner Greg Howison shared his perspective on some of the legal ramifications around AI, IP, connected devices and the data they generate, in response to emailed questions.
A certain amount of skepticism is healthy because it can stop us from falling for the hype or for misplacing our hopes and then being disappointed when AI doesn’t live up to our unrealistic expectations.
Hence, we agree that the pharma industry has the right to be skeptical because there’s so much noise and hype out there. It sometimes seems like every healthcare entrepreneur with a computer is being called an AI drug-discovery startup. Historically, AI developers have over-promised and under-delivered, with IBM’s Watson acting as a case in point. It may not be for the lack of trying but the reality is that breaking into something with a totally different approach is just hard. However, while there’s a certain amount of prudence to being skeptical of AI and other new technologies, we can’t afford for that to hold us back. We do not want our skepticism to become a self-fulfilling prophecy that restricts the biopharma industry.
The key, as with most things, is moderation, and we shouldn’t let skepticism take away from the fact that there are plenty of real, talented people who are already using AI to revolutionize the biopharma industry. Better still, the number of such examples is on the rise, and we’ve noticed a huge shift amongst cutting-edge biopharma companies in just the last couple of years.
The applications of AI in Biopharma
AI has a virtually unlimited number of applications in the biopharma industry. That’s because it can take medical data and breathe new life into it, unlocking new treatments and enabling truly personalized healthcare. AI algorithms are essentially prediction machines, and accurate predictions in biopharma equate to better health for everyone.
A Personalized Approach to Medication Nonadherence
At the Abarca Forward conference earlier this year, George Van Antwerp, managing director at Deloitte, discussed how social determinants of health and a personalized member experience can improve medication adherence and health outcomes.
Here are some of the top use cases we’ve got our eyes on.
1. Image processing
One of our favorite use cases of AI in biopharma is for image processing. Let’s take lung cancer as an example here. Millions of patient images are created every year, and the problem with human doctors is that humans can make mistakes. Even with two or three physicians looking at every scan, there can be discrepancies and misses.
The good news is that AI can take on the task of image processing and automatically parse all of the millions of scans to look for anomalies that might have been overlooked. We’re not talking about replacing doctors with algorithms, but rather about humans and machines partnering for better outcomes with the machines carrying out safety checks and managing any oversights.
2. Personalized healthcare
Personalized healthcare is one of the areas of medicine that both of us are very passionate about. Co-author of this post, Dr. Fombu, has talked about the idea of a “Netflix of healthcare” in which every patient’s treatment plan is informed by what’s worked well for other, similar patients using algorithms.
There are still several impediments to achieving true personalized healthcare. Although the technology itself may be ready and waiting, politics and bureaucracy may hinder data sharing, causing a lack of interoperability between different data sources for example. Of course, the other critical issue is the existence of the medicine itself for a particular subset of patients. Personalized healthcare will only broadly work if the drug needed is already discovered, developed and available in the market for the patient. We will touch on this critical point later in the next section.
3. Drug discovery
We find this area of AI and biopharma the most fascinating. Statistics say that bringing a new prescription medication to market can cost as much as $2.8 billion and at least 10 years. Surely there is a lot of room for improvement. Although there are many AI-driven efforts to improve this area, we think a radical change is needed to move the needle. At the end of the day, it is about tackling disease biology. The key is to take AI applications and use them to understand disease heterogeneity in a way that can improve the chances of a drug succeeding in clinical trials. That’s where the rubber meets the road for this industry.
Overall, it is very encouraging to see a huge increase in the number of companies in this AI space, chipping away at this multi-billion dollar problem. Some even argue that drug-discovery costs can be cut by as much as 70% using AI approaches. That works out at a saving of nearly $2 billion. It’s hard to argue against numbers like that. It will be interesting to see what the coming years will bring in this application of AI in the pharma industry.
4. Data management
Last but not least is data management and the application of natural language processing (NLP), a technique that essentially allows an algorithm to take data and to convert it into something more understandable. One of the major challenges that AI algorithms have to face is the fact that if data isn’t in a machine-readable format, it’s useless.
At its most basic level, NLP can take doctors’ notes and digitize them, reducing the amount of time that they need to update patient records. This alone would be a huge help considering that first-year physicians spend three times more time inputting EHRs than they do on patient care.
5. The future of AI in biopharma
Thinking about AI in the biopharma industry isn’t like speculating about some far-off science fiction future. The technologies and approaches that we’ve talked about are already a reality, and the biopharma companies that embrace AI will be the ones that come out on top in the long run.
But there’s much more at stake here than profits and share values. By tapping into the awesome (and ever-increasing) power of AI, biopharma companies can make a real difference to patient outcomes. They won’t just save lives, they’ll also improve people’s overall quality of life, another important but often forgotten metric.
The challenge is finding the needle in the haystack, and that’s where a healthy dose of skepticism can come in handy. Ultimately, we should be excited about AI and its adoption in biopharma, but we should also stay sensible and focus our efforts where they’ll have the greatest impact. The future of biopharma is bright. It falls to us to make it happen.
Photo: metamorworks, Getty Images
Dr. Emmanuel Fombu is an internationally recognized authority on the convergence of digital technologies and healthcare. He is an award-winning and best-selling author, physician, Keynote speaker, investor, entrepreneur, and medical futurist with over 15 years of combined experience in science, research, and healthcare. Dr. Fombu did his clinical training at Emory-Crawford Long Hospital and holds an MBA from both Cornell University’s Johnson School of Business and Queen’s University’s Smith School of Business. He also earned a certification in Artificial Intelligence: Implications for Business Strategy from MIT’s Sloan School of Management and Computer Science Artificial Intelligence Lab. Dr. Fombu has authored multiple research papers and abstracts in renowned peer-reviewed journals. He serves as an external advisory board member at the Massachusetts Institute of Technology’s MIT.nano project. He lives in New York City.
Pek Lum, Co-founder and CEO at Auransa. With more than 20 years of genomics and drug discovery experience, she is the chief architect of the science behind Auransa’s technology. Before founding Auransa, Pek was vice president of Product and chief data scientist for Ayasdi. Pek started off her career at Rosetta Inpharmatics, a pioneer in genomics and microarray technology. Pek received her Ph.D. in yeast genetics at the University of Washington, Seattle. Her work has been published in scientific journals, such as Nature and Cell, and her research has contributed to discoveries in drug development and the understanding of complex diseases. She has previously served as an advisor to the Michael J. Fox Foundation for Parkinson’s Research, Bayes Impact and Resolution Bioscience, and is currently an advisor for Sequence Bio.
