The MaML Podcast - Medicine & Machine Learning

Ittai Dayan, MD is the co-founder and CEO of Rhino Health, a distributed computing platform leveraging privacy-preserving federated learning. The platform allows medical researchers and healthcare AI developers to seamlessly access diverse and disparate datasets and use them to create better AI algorithms.

[email protected]

Host: David Wu / Twitter: @davidjhwu

Producer: Aaron Schumacher / Twitter: @a_schu95

Artwork & Video: Saurin Kantesaria

Music: Caligula - Windows96. Used with Artist Permission.

00:56 How did you come to the intersection of medicine and artificial intelligence?

06:15 What type of medicine did you start out studying?

11:35 Could you tell us the story behind Rhino Health?

14:30 What is federated learning?

21:00 Common use cases for Rhino Health?

26:45 Relationship between generalizability and accuracy when using federated learning?

28:15 What were your biggest challenges in creating Rhino Health?

32:40 An example of using Rhino Health?

37:40 How does Rhino Health integrate with EHR’s

38:15 What are your next steps for Rhino Health?

43:10 What do you think the future of AI in healthcare will look like?

48:08 What gives your life meaning and what are your greatest fears?

Beth Beadle, MD, PhD is the Director of Head & Neck Radiation Oncology at Stanford and co-creator of the Radiation Planning Assistant, a fully automated treatment planning assistant. 

00:56 How did you come to the intersection of medicine and AI?

03:20  What is radiation oncology and why does it fit well with AI?

05:54 How has radiation changed in your lifetime?

09:19 What is the Radiation Planning Assistant?

13:10 Describe radiation oncology workflow before the use of the radiation planning assistant

20:45  How does the model compare humans?

23:45 Where was the data source for the radiation Planning Assistant Model?

26:45 What safeguards exist for RPA?

28:00 What has been the response from other physicians, physicists, and patients to this model?

33:30 Timeline for implementation of the radiation planning assistant?

34:45 Background of the Radiation Planning Assistant

38:25 Future of Radiation Planning Assistant implementation

40:55 Adaptive planning in radiation oncology

43:15 Future of the Radiation Planning Assistant in the next 10 years

44:50 Are there any other projects being planned currently?

47:00 How has mentorship shaped your path?

50:25 What is the future of AI in medicine in 10-20 years

51:48 What advice would you tell yourself when you were graduating from medical school

53:20 What brings your life joy and meaning?

54:50 What are your greatest fears?

55:15 Favorite places to travel?

Dr. Tignanelli is the scientific director for the Program for Clinical Artificial Intelligence at the UMN Center for Learning Health Systems Science, the director of UMN Center for Quality Outcomes, and the chair of the Health Information Technology (HIT) Committee for the American College of Surgeons. Dr. Tignanelli's work serving COVID-19 patients during the pandemic and advancing AI research earned him the title of “Health Care Hero” by the Minneapolis/St. Paul Business Journal in 2021.

Follow us on Twitter @TheMaMLPodcast

Guest: Christopher Tignanelli, @cjtign

Host: Madeline Ahern, @maddie_ahern

Producer: Kirsi Oldenburg

Artwork: Saurin Kantesaria

Music: Caligula - Windows96: Used with artist permission

Notes:

01:00 tell us about yourself

04:00 critical care/acute care

06:30 COVID acute care

10:00 computer vision

14:00 rib fracture model

15:00 external vs internal validity

19:00 future of AI in medicine

Dr. Chenyang Xu is currently the President and Co-Chairman of PVmed Technologies, Co-Founding Partner of Silicon Valley Future Academy, Managing Partner of Brightway Future Capital, and an Advisory Board Member for the Johns Hopkins University BME department and formerly Advisory Board Member for the UC Berkeley EECS Department.

He was formerly the Chief Business Officer of RSP Systems and the GM and CTO of Siemens Technology to Business (TTB) at Berkeley where he led the Siemens’s North America’s technology startup partnership and early-stage investment practice out of the Silicon Valley. As former head of Siemens Interventional Imaging Program, he led an R&D team that developed over 10 new computer vision-based medical imaging products (e.g. CartoMerge) and has achieved billion dollar scale new revenue stream.

Follow us on Twitter! @TheMaMLPodcast

Guest: Chenyang Xu

Host: David Wu / Twitter: @davidjhwu

Producer: Aaron Schumacher / Twitter: @a_schu95

Artwork & Video: Saurin Kantesaria

Music: Caligula - Windows96. Used with Artist Permission.

01:00 - tell us about your path

11:00 - the internet + grad school in the 90s

22:00 - “career thinking” advice

30:30 - Siemens VC - looking at 10,000 startups

37:00 - PVMed - AI cancer treatment company in China

43:30 - AI startup ecosystem in China vs. US vs. Europe

56:30 - Potential downsides of implementing AI too quickly?

1:02:10 - Future of AI in medicine in 10-20 yrs?

1:13:00 - “Minority Report”-like AI to predict falls?

1:16:30 - Closing Questions

Steven Lin, M.D. is the service chief of family medicine for Stanford Health Care and the Founder and Executive director for the Stanford Healthcare AI Applied Research Team. 

Follow us on Twitter! @TheMaMLPodcast

Host: Madeline Ahern / Twitter: @maddie_ahern

Producer: Kirsi Oldenburg

Artwork & Video: Saurin Kantesaria

Music: Caligula - Windows96. Used with Artist Permission.

Akilesh Bapu is co-founder and CEO of DeepScribe, an AI-powered medical scribe that passively records and understands a patient’s visit, generates a clinical note, and then seamlessly inputs it into the electronic medical record.

Follow us on Twitter! @TheMaMLPodcast

Guest: Akilesh Bapu / Twitter: @AkileshBapu

Host: David Wu / Twitter: @davidjhwu

Producer: Aaron Schumacher / Twitter: @a_schu95

Artwork & Video: Saurin Kantesaria

Music: Caligula - Windows96. Used with Artist Permission.

1:10 - Pathway towards the intersection of medicine and Machine Learning

9:45 - What specialties are best fitted for DeepScribe implementation

11:55 - How does DeepScribe find relevant information for the clinical encounter

14:45 - The creation process of the DeepScribe hardware.

18:10 - DeepScribe’s ability to discern fragments of the SOAP note

19:45 - DeepScribe and billing concerns

24:25 - Scaling the human scribe capacity behind DeepScribe

29:45 - Implementing predictive models with DeepScribe

32:00 - Voice diagnostics

33:40 - Patient privacy and DeepScribe

38:55 - Vision for deep scribe

40:20 - Google Care Studio EHR mention

42:00 - Physician Burnout

42:35 - Advice for future founders

44:15 - What brings you joy?

45:35 - What are your greatest fears?

47:00 - What gives your life meaning?

50:25 - How do medical students think about the future of healthcare

Dr. Nigam Shah is Professor of Medicine (Biomedical Informatics) at Stanford Medicine and Chief Data Scientist at Stanford Healthcare. Dr Shah’s research focuses on combining machine learning and prior knowledge in medical ontologies to enable use cases of the learning health system.

Follow us on Twitter! @TheMaMLPodcast

Guest: Nigam Shah / Twitter: @DrNigam

Host: David Wu / Twitter: @davidjhwu

Producer: Aaron Schumacher / Twitter: @a_schu95

Artwork & Video: Saurin Kantesaria

Music: Caligula - Windows96. Used with Artist Permission.

TIMESTAMPS

0:45 tell us about your path and how you came to the intersection of MaML

4:15 what is the thesis of your work?

09:10 where is AI on the gartner hype curve

13:30 the equation of medicine - if risk > threshold, take action. Examples: GreenButton Consultation service + advanced care planning

27:00 how has mentorship shaped your path?

32:00 having fun in projects

34:00 “what makes a project fun?”

37:00 closing questions - what do you expect is the future of AI?

43:00 personal questions - what brings you joy?

Dr. Quynh Nguyen is an assistant professor of epidemiology and biostatistics at the University of Maryland School of Public Health. She received her PhD and MSPH in Epidemiology from University of North Carolina at Chapel Hill, Gillings School of Global Public Health. Dr. Nguyen is a social epidemiologist focusing on contextual and economic factors as they relate to health. She joined us to talk about her projects that leverage technology and big data sources to investigate and address health disparities.

Host: Raeesa Kabir

Producer: Kirsi Oldenburg

Artwork: Saurin Kantesaria

Follow us on Twitter: @TheMaMLPodcast

Have a speaker you would like to see on our podcast? Contact us at [email protected]

Dr. Vineeta Agarwala is a general partner at a16z, physician, adjunct clinical professor at Stanford, previously at Google Ventures, Flatiron Health, having received her MD/PhD at Harvard Medical School and the Broad Institute. 

Many thanks to Ashlea Kosikowski from 1AB Media for making this episode happen!

Follow us on twitter: @TheMaMLPodcast

Host: David JH Wu @davidjhwu

Producer: Aaron Schumacher @a_schu95

Design: Saurin Kantesaria

Dr. Harald Kittler is a Professor in the Department of Dermatology at the Medical University of Vienna, in Vienna, Austria. Dr. Kittler is the founder of Dermachallenge, a 2018 startup which uses the principles of gamification to teach and train health professionals. As a dermatologist and dermatopathologist, Dr. Kittler has a unique insight into how future physicians might be trained in his profession. Enjoy some gamification of dermatology by joining the over 5,000 players at Dermachallenge!

Host: Madeline Ahern

Producer: Kirsi Oldenburg

Artwork & Video: Saurin Kantesaria

Music: Caligula - Windows96 (Used with artist permission)

Have a speaker you would like to see on our podcast? Contact us at [email protected]