Summary
In this episode, Anastassia and her guest, Andrea Olsen, discuss the role of AI in drug discovery and building healthcare solutions.
Andrea Olsen is 20 years old today, and she is the founder of Luria, an AI-powered software for personalized healthcare. Based at Caltech University, she studies computational building her business. She founded the Youth Longevity Association for high schoolers interested in a longevity science career. She initiated the Inspire Longevity program at the largest aging pharma industry event, ARDD (Aging Research & Drug Discovery). Her work on identifying genetic targets with AI in brain cancer (glioblastoma multiform) was published in the journal "Aging." It became a foundational case study for Insilico Medicine, one of the best-known AI companies that pioneered AI-based drug discovery. In her free time, Andrea likes to travel and explore new countries. She also has a passion for learning languages and is currently studying her seventh.
Andrea shares her journey into computational neuroscience and her work on glioblastoma research. She discusses the role of AI in analyzing large datasets, the challenges of data quality and bias, and the importance of human involvement in ensuring AI's effectiveness. Andrea also explores the potential of AI to reduce the time to market for medications, the implications of personalized medicine, and the need for better education and adoption of AI in the medical field. She emphasizes the importance of gathering feedback from doctors to create effective AI solutions in healthcare.
Takeaways:
Andrea's journey into neuroscience began during the COVID lockdown.
AI plays a crucial role in analyzing glioblastoma data, one of the most common brain tumors.
The quality of data is essential for accurate AI outputs. While the AI world loves big data, we must look into how to achieve the best outcomes with small data.
Human involvement is necessary to ensure the quality of AI. Whether we review synthetic data to augment real-world medical data or introduce quality gates to decide on the next steps in drug discovery, human supervision remains imperative.
Synthetic data should be used cautiously in drug discovery.
AI can help flag patient risks effectively.
Doctors learn how to trust AI for personalized medicine.
AI can save time for doctors by automating tasks. Even mundane AI applications that help nurses and doctors with their healthcare routines can enable more time for patients and better care.
Gathering feedback from doctors is crucial for the development of ANY AI service and application.
The future of AI in medicine relies on collaboration with healthcare professionals.
Chapters:
00:00 Introduction to AI and Glioblastoma Research
02:36 Andrea's Journey into Computational Neuroscience
03:39 The Role of AI in Analyzing Glioblastoma Data
06:48 Challenges of Data Quality and Bias in AI
10:26 Human in the Loop: Ensuring AI Quality
11:32 Synthetic Data in Drug Discovery
13:04 AI's Potential to Reduce Time to Market for Medications
16:26 Personalized Medicine: The Future of Healthcare
19:37 Education and Adoption of AI in Medicine
22:40 Building a Medical AI Startup
26:12 Feedback and Iteration in AI Development
Hyperlinks:
Andrea's paper in "Aging":
Andrea Olsen (Google Scholar Profile)
Anastassia's links:
