“The Era of Artificial Intelligence in Pharmaceutical Industries” IPhO Journal Club Presentation by: Benga Sodiya & Pamela Diokpara

Outline of the Study

1. AI Networks and Tools

2. Classification of AI

3. Artificial Intelligence and Robotics

4. Pharmaceutical Automation

5. AI in advancing Pharmaceutical Drug Development

6. AI in Pharmceutical Marketing

7. AI in Quality Control and Quality Assurance

8. AI in Clinical Field

AI Networks and Tools

2 types:

• Machine Learning (ML):
• Algorithms used to find patterns in data to be further categorized
• Deep learning (DL):
• Part of ML that uses artificial neural networks (ANN)
• ANN’s are electrical impulses that work together to solve problems

Classification of AI

3 categories

Artificial Narrow Intelligence (ANI)

Artificial General Intelligence (AGI)

Artificial Super Intelligence (ASI)

4 types

1: no memory; reactive only

2: uses limited memory to make decisions; however, memory is NOT stored permanently

3: “Theory of Mind”, tries to mimic human thinking, intentions, and desires

4: self awareness and consciousness

Artificial Intelligence and Robotics

Similarities: use a software agent that controls the robot then sensors tell it what to do next in a real world environment.

Pharmaceutical Automation

• Everyone is looking for a way to make things more efficient
• Goal: getting the right medication to the right patient at the right time
• With AI HCPs are able to pick the best medication with the best outcome
• Allows us to learn from real time data
• Right people for clinical trials
• Real time patient feedback
• Data exchanges with partners
• Distributors and caregivers
• Where do you see automation in healthcare?
• Wireless
• Nanotechnology
• Advance storage and memory
• Sensors and analyzers
• Advance software algorithms

AI in Drug Development and Marketing

Drug Development

• AI makes the process of finding appropriate dosage forms of medications easier by QSPR
• Can help overcome the challenges of stability, dissolution, porosity, and other parts of drug design

Pharmaceutical Marketing

• Automating parts of the manufacturing process
• CFD in Reynolds Averaged Navier Stokes
• Effects of agitation and stress levels in different pieces of equipment (like stirring tanks)

AI in the Clinical Field

• The process of conducting a clinical trial is not only long, but expensive
• Only 1/10 of compounds are able to go through a clinical trial.
• The other 9 compounds that fail can come from bad infrastructure, poor technical requirements and poor patient selection
• AI can help minimize these issues with data that is currently available

AI in Quality Control and Assurance

• Humans are needed to maintain quality control testing
• AI can be helpful, but companies have to be careful not to over use AI.
• Implement a “Quality by Design” approach
• cGMP can be used to help companies understand the best ways to implement AI in the QA and QC processes

Background of the Study

AI applications are often associated with the fear of unemployment. However, the industry widely recognizes and praises the advancements in AI technology due ot its significant impact on efficiency. The use of AI reduces the workload of human workers while accomplishing objectives swiftly

Discussion: Methods and Results

The adoption of AI allows for learning from real-time data:

• Identifying the right candidates for clinical trials
• Processing real time patient feedback
• Integrating data exchanges with partners
• Distributors and caregivers

CRITIQUES

The paper doesn't discuss how AI can be a hindrance to Pharmaceutical Industries, outside of the fear of unemployment:

• Job Loss
• Bias and Discrimination
• Privacy Concerns
• Lack of Transparency
• Overreliance

The paper also fails to disscuss the use of AI in other functional areas of pharmacy (medical affairs, informatics, etc.)

Conclusions

Due to the belief held by AI technological approaches that humans are capable of envisioning knowledge, resolving problems, and making decisions, there has been a surge of interest in implementing AI technology for analyzing and interpreting important areas of pharmacy, including drug discovery, dosage form design, polypharmacology, and hospital pharmacy.

The utilization of automated workflows and databases for efficient studies employing AI techniques have proven advantageous.

With the help of AI technologies, the formulation of innovative hypotheses, strategies, predictions, and assessments involving various interconnected elements can be carried out with ease, reduced time consumption, and affordability

But what are our thoughts?

Effects of AI in Medical Affairs

• Medical Affairs, specifically Field Medical Affairs/ Medical Science Liaisons, are professionals who communicate internal information from the company to other healthcare providers, stakeholders and key opinion leaders
• ChatGPT is a very new form of an AI chatbox.
• It has the ability to give real life answers to questions
• Survey was done by Sarah Snyder PharmD on the interest of ChatGPT in medical affairs professionals
• 44% of responses mentioned that they were familiar with the platform.
• Of that 44%, 15% of responders have used it frequently with success.
• Issues with HCPs bypassing the medical affairs department at a company to get information.

What do the experts think?

Lenin once said: “There are decades when nothing happens, and then there are weeks, when decades happen”. Medical affairs has been having some of those weeks of late with the introduction of generative AI (chatGPT, Bard, etc.). In short, we’re in the midst of a transformation and the experimentation, the controversy and the intrigue surrounding this new technology will quickly give way to new platforms, new models and new value recognized industry-wide. This will all happen this year. After that, as Bill Gates has said in a recent article, the age of Artificial Intelligence will be upon us, where most major medical affairs activities, from literature search, to gap analysis, to medical communications and education, to presentation training, insights mining, conference intelligence and medical information will have artificial intelligence juxtaposed as part of the way they are considered and delivered, sometimes without the doer or the user even knowing. This will free up our colleagues’ time to focus on delivering impact to professionals and patients and hopefully, leading to happiness both internally within teams and externally as manual work and routinizable work is delegated to the machines. An exciting future awaits!

Matt Lewis MPA Co-Founder of Inizio Medical Analytics and Innovation (3/23/23)

Remember...

Remember the first letter in AI stands for artificial and we need to keep that in mind when dealing with patients and interactions that happen within the pharmaceutical industry.

References

https://www.linkedin.com/posts/sarahsnyderruns_chatgpt-medicalaffairs-ai-activity-7035962893611266048-x49T?utm_source=share&utm_medium=member_desktop

Praveen Tahilani, Hemant Swami, Gaurav Goyanar, & Shivani Tiwari. (2023). The Era of Artificial Intelligence in Pharmaceutical Industries - A review. Indian Journal of Pharmacy & Drugs Studies, 1–5. Retrieved from https://mansapublishers.com/index.php/ijpds/article/view/3481

Quote from Matt Lewis, (3/23/23)