Risk Never Sleeps Podcast – Episode #168

Title: How AI Is Unlocking Breakthroughs From Both Massive And Minimal Data
Host: Ed Gaudet
Guest: Dr. Yves Lussier, Chair of Biomedical Informatics and Professor of Medicine, University of Utah
Date: December 17, 2025

Episode Overview

This episode features a thought-provoking conversation with Dr. Yves Lussier, a trailblazer in biomedical informatics and artificial intelligence (AI) as applied to healthcare data. The discussion ranges from the legacy of biomedical informatics at the University of Utah, to the evolving power of AI with both vast and minimal datasets, and the personal journey of Dr. Lussier in transforming academic institutions and healthcare technology.

Main Theme:
How AI can derive clinical insights from both massive and “minimal” data, overcoming traditional limitations, and the important considerations regarding data integrity and patient safety in the digital healthcare era.

Key Discussion Points & Insights

1. Legacy of the University of Utah in Biomedical Informatics (00:43 – 01:46)

• Dr. Lussier recounts the University of Utah’s historic role:

  • One of the world’s oldest and most influential departments in biomedical informatics.
  • Innovations such as the earliest electronic health records, LOINC codes, and HL7 standards originated there.
  • Alumni have established leading departments at Harvard, Columbia, and Taiwan.

• Quote:

  “The goal was really to enhance the oldest department of biomedical informatics in US and likely the world ... the University of Utah had the earliest electronic record, trained the individuals that became the founder of the field in Taiwan.”
  — Dr. Yves Lussier (00:43)

2. Data-Driven Discoveries and The Power—and Peril—of AI at Scale (01:54 – 05:50)

• Historic Case Study: The discovery of the BRCA1 gene (linked to breast and ovarian cancer risk) using large clinical and ancestry datasets, pioneered by biomedical informaticians.

• Opportunities in AI:

  • AI models (like ChatGPT) benefit from massive data, which uncovers “emergent properties.”
  • Yet, large models struggle with “counterfactuals” (rare or incorrect examples), highlighting vulnerability to both error and potential malicious data manipulation.
  • Risks: Data poisoning by adversarial actors, requiring scrutiny of data sources to protect AI model integrity.

• Quote:

  “There are other methods that could at scale create these negative data sets in which they could learn better. But we can also think of enemy countries or organization that would generate erroneous data set that appear to be true.”
  — Dr. Yves Lussier (03:28)

3. From 'Curse' to 'Blessing' of Dimensionality: Making Sense of Minimal Data (05:01 – 07:02)

• Traditionally, AI and statistics struggled with scenarios where variables outnumbered data points (“large P, small n”; e.g., thousands of gene expressions but only a few samples). This was known as the “curse of dimensionality.”

• Dr. Lussier’s work has led to the “blessing of dimensionality”—using new statistical methods to draw clinical insights from very small datasets, even an individual patient with just two samples.

• Application in personalized genomics and “n-of-1” trials:

  • Identify pathways underlying drug resistance or exceptional response quickly and with minimal patient data.
  • Impact: Timely, precise treatment decisions, avoiding unnecessary or ineffective therapies.

• Quote:

  “Our team and some of our colleagues have reduced new statistics and it’s called the blessing of dimensionality.”
  — Dr. Yves Lussier (05:37)
  “With these n-of-1 trials ... overnight you get a solution to their edge case, which can be in a super responder or a resistance to a therapy.”
  — Dr. Yves Lussier (06:00)

4. Observation from the Field: AI Progress and Interdisciplinary Mindset (07:02 – 08:00)

• Dr. Lussier is impressed by keynote speakers at AIMed25 who deliver “substance and brevity.”
• Emphasizes his multidimensional position as physician, engineer, informatician, and entrepreneur.

5. Pioneering Technology: Early AI for Medical Records and Billing (08:00 – 10:30)

• Dr. Lussier founded Purkinje in 1991: the first pen-computer with AI and handwriting recognition for physicians, still in use in thousands of clinics.

• Innovated the move from “system of record” to “system of intelligence”—using concept graphs instead of flat data dictionaries.

• Early innovations improved billing efficiency and accuracy, reducing claim denials and accelerating receivables.

• Quote:

  “Instead of an electronic record based on ... a system of record, it was an electronic record based on a system of intelligence.”
  — Dr. Yves Lussier (08:01)
  “So I’m an expert at billing as well as ... genomic medicine.”
  — Dr. Yves Lussier (10:30)

6. Leadership, Risk, and Institutional Transformation (10:42 – 12:28)

• Dr. Lussier reflects on his career as a “transformative leader,” frequently establishing new units and data warehouses at top institutions (Columbia, University of Chicago, Arizona).

• Greatest risks: Each new leadership role, especially revitalizing established (and sometimes entrenched) departments.

• Quote:

  “The riskiest part was always the next challenge ... every promotion or every change of institution was actually a very high risk.”
  — Dr. Yves Lussier (11:30)

7. Entrepreneurship and Life Advice (12:37 – 14:10)

• Started his first company at 18; had to choose between dropping out for startups or continuing in academia.

• In hindsight, Dr. Lussier would advise his younger self to seek innovation hubs like Silicon Valley or Boston earlier and embrace startup culture.

• Quote:

  “You’ve got to brainwash me on the number of years it takes to train all these multiple disciplines ... It’s just risk it, drop out and risk it all.”
  — Dr. Yves Lussier (13:17)

8. Global Impact and Ongoing Leadership (15:00 – 16:36)

• Over 600 alumni from the Utah program, many transforming international standards.

• Example from Taiwan: Alumni led the development of healthcare interoperability allowing patients and providers full access to medical records—far beyond current US capabilities.

• Quote:

  “You can get access to your electronic entire set of data across any provider ... including everything, which is a hard thing to do in the US.”
  — Dr. Yves Lussier (15:18)

Notable Quotes & Moments

• On the 'curse of dimensionality':
  “With these n-of-1 trials, ... overnight you get a solution to their edge case ... which can be in a super responder or a resistance to a therapy.”
  — Dr. Yves Lussier (06:00)

• On entrepreneurship:
  “I realized I’d make more money by dropping out of engineering or medicine than continuing on this academic trend.”
  — Dr. Yves Lussier (12:51)

• On advice to his younger self:
  “The best advice was drop out and go to these sectors [Silicon Valley or Boston].”
  — Dr. Yves Lussier (13:51)

Important Segments – Timestamps

• History and Legacy of Biomedical Informatics at Utah – 00:43–01:46
• AI Leveraging Large and Small Data, Security Risks – 02:27–05:50
• Personal Genomics, Blessing of Dimensionality – 05:01–07:02
• Entrepreneurship and Early AI in EHRs/Billing – 07:32–10:31
• Leadership Decisions and Institutional Risk – 10:42–12:28
• Advice to 20-Year-Old Self / Startup Culture – 12:37–14:10
• Alumni Impact, Taiwan Interoperability – 15:00–16:36

Resources & Further Reading

• Dr. Yves Lussier’s Wikipedia and LinkedIn profiles
• University of Utah’s Biomedical Informatics Department
• AIMed 25 Insight Series
• Censinet: Risk Never Sleeps Podcast Info

This summary preserves the engaging, insightful tone of Dr. Lussier and the host, providing a complete view of how data-driven innovation—in giant leaps or with just a handful of data points—is transforming patient safety and healthcare systems worldwide.