The MaML Podcast - Medicine & Machine Learning

Judy Gichoya, MD, MS, is an interventional radiologist at Emory University in Atlanta. We begin by talking about Dr Gichoya’s early days in Kenya where she participated in building OPEN MRS, the world’s leading open-source EMR platform. We then talk about her work in using AI to combat bias and social injustices in medicine and the importance of diversifying the datasets we use in AI work today.

03:00 Origins in Kenya, building OpenMRS, path to AI

14:00 Research topics of interest in the Gichoya Lab (Emory) such as bias in AI

21:00 steps we can take to combat bias in datasets

27:00 work on federated learning

36:00 advice to medical students / early-career med students interested in the field

43:00 balancing clinical work and informatics research

50:00 favorite food from hometown!

Daniel Tse, MD, is a product manager at Google Health. In this episode, we talk about his unconventional path to healthcare and AI, and discuss the career angst that many may feel when they wonder that if the path they are on is the right path. I can’t say that we’ve discovered the perfect answer in this interview, but I will say that Dr. Tse is someone who has thought deeply about this question as you can see from his unique career and will probably provide some good words of advice.

Had a real pleasure recording this interview. Hope you all enjoy!

Interviewer: David Wu

Producer: Aaron Schumacher

1:30 Childhood + Budding Interest in Computers

8:00 Path to medicine

12:00 Working at a healthcare startup during medical school

20:00 Working at Google after graduating medical school

30:00 Comparing and contrasting Google with an academic research lab

35:00 MD training + how it relates to working at Google

38:00 What does a product manager do?

45:00 What do you expect is the future of AI + medicine?

49:00 What advice would you give to your 25 yr old self?

52:00 Advice to people who are questioning whether medicine is the right path for them

Jack Moore is a product manager at Qventus, a healthcare start-up based out of Mountain View, California. In this episode, we go into depth about how he and his team are using machine learning to improve the efficiency of hospitals such as the University of Minnesota’s MHealth system. One example we discuss is how Qventus builds and deploys models to improve patient care and reduce ICU bed wait times. At the end, we spend time talking about bias and race in medicine and how we can use AI to combat the propagation of such biases.

Interviewer: David Wu

Producer: Aaron Schumacher

Time Stamps:

Path to healthcare AI and Qventus: 01:00

What does Qventus do 04:15

Story behind Qventus 09:00

What if the AI model and physicians disagree? 15:00

Structure of the Qventus AI models + explainability 21:00

Integration with Epic 30:00

What Qventus use looks like as a provider 36:00

Cool definition of data science 42:00

AI to reduce health inequalities and racial bias in medicine 58:00

Miguel Alvarado is the CTO of Lumiata, a healthcare AI company based in SF. In this episode, we talk about Miguel’s unconventional journey from Microsoft in the 90s all the way up until the present moment where he is helping bring the power of big data and AI to patients, providers, and insurers.

Just like Bruce Wayne, Miguel works during the day, but at night, has an alter ego. we will talk about his passion as an underground house and techno DJ and finally, we will close with a conversation about the importance of being present.

I had a real pleasure conducting this interview with Miguel and I hope you enjoy it.

00:00 Intro

02:00 Beginnings

12:30 What does Lumiata do

23:40 Synthetic Data

30:00 Explainability in models

34:00 what do you think is the future of AI in medicine

40:00 the importance of being present

44:00 the importance of having other passions

51:00 dj’ing, techno, and being fully present

55:00 machine learning and.. music?!

Welcome to the very first episode of the Medicine and Machine Learning Podcast, AKA the MaML Podcast brought to you by the Medicine and Machine Learning Club of the University of Minnesota Medical School. We are a group of students passionate about learning more about the future of medicine and AI. This Podcast will feature interviews from prominent figures in academia, industry, and medicine exploring the cutting edge of healthcare’s newest frontier: medicine & machine learning.

My guest today is Dr. Christopher Weight, the new Director of Urologic Oncology at the Cleveland Clinic. In addition to this role, Dr. Weight is currently helping build a new center for AI and Medicine at the Cleveland Clinic. In this episode, we will be talking about his unique path in medicine and how it led to building one of the largest and most comprehensive kidney cancer imaging datasets we have today.