- Joined
- Oct 5, 2013
- Messages
- 370
- Reaction score
- 219
Interesting. Can you elaborate on the role of machine learning within a pharma company? And if or how a pharmacist's knowledge can add to that?
This can be a lengthy topic. You can google "machine learning clinical trials" or "applications of machine learning in pharma and medicine" and read more about it. Pharmcists' clinical knowledge can be very beneficial in clinical data analysis and management, also pharmacoeconomics, just to name a few, despite the fact that most PharmDs currently don't work in those areas since they seriously lack the programming/data analysis skills those positions require the most. So PharmDs now heavily populate in pharmacovigilance, medical/regulatory departments. But again, the more gaps you can bridge, the more valuable you are. From what I have seen so far, very few clinicians can comfortably program, and very few statisticians/programmers really understand why they are doing the job they have been assigned to other than taking the instruction from their superiors. If you look at finance nowdays, Fintech has completely turned the finance world upside down to the point that the CFA exams will have a dedicated Fintech section starting next year. It's just a matter of time for biotech/pharma, at least that's how I see it.
In addition, once you adopt the mindset of a programmer and finish a few projects, you will soon realize just how easy pharmacists' routine work can be automated. Most clinical guidelines is nothing but a series of if/else statements. If you know some SQL and a general purpose programming language like Python or Java and know how to read and interpret clinical guidelines and product monographs, it's not difficult at all to build a clinical recommendation system or medication counselling program that can at least partially substitute pharmacists' clinical roles. If you know ML and consistently keep training such a program, it can eventually replace even the most competent clinical pharmacists, and it never makes the same mistake twice! Don't believe me? Just look at how AlphaGo beat Lee Sedol. I think the only obstacle from wide adoption of such clinical programs in all hospitals and pharmacies across the country is from the legal perspective, but I think this will be inevitable in future since it is just so much cheaper to automate than paying 6 figure annual salaries per pharmacist. The huge cost-saving incentive is just too hard to resist from business standpoint.
Last edited: