AI and Healthcare

Every oncologist knows the electronic health record is where the cancer story goes to hide. Outside records arrive by fax, get scanned into the media tab under the wrong name, and pile up to hundreds of documents per patient from dozens of institutions, written in coding languages that never quite translate. Screening one chart for a clinical trial takes an hour. Tumor registries run months behind. Drugs get denied because a performance status was in the note but not in the request. No wonder "data" can feel like a four-letter word. This episode is about what changes when AI can finally read all of it.

 

In this episode of AI and Healthcare, oncologists Sanjay Juneja, MD and Doug Flora, MD sit down with Mika Newton, CEO of xCures, to unpack a new AI in Precision Oncology paper on using large language models to turn hundreds of medical records per cancer patient into structured, verifiable healthcare data.

 

In this episode:

• Why a cancer patient's medical records can run to 231 documents from dozens of institutions, and why most of it is still not in discrete fields

• The many "languages" of healthcare data (OMOP, SNOMED, ICD, CPT) and why the crosswalks are never one-to-one

• Learning the structure of a document before extracting from it, and why 95% of the work is organizing rows and columns

• The three buckets: what to automate, where a human belongs in the loop, and when to give up on an unreadable page

• How to vet a data vendor: dual human extraction, model drift, a 17-page buyer's guide, and whether it pays for itself

• Interoperability in healthcare: TEFCA, Individual Access Services, CMS's Kill the Clipboard program, and what changes when patients own their own health data

• The disparity question nobody has solved yet

 

Mika Newton is Chief Executive Officer of xCures, which has applied AI to organizing and extracting structured data from medical records since 2018. He is a founder of this podcast and a member of the editorial board of AI in Precision Oncology, the journal publishing the paper discussed in this episode. Doug Flora, MD is the journal's Editor-in-Chief; the paper went through independent peer review.

 

Read the paper: https://journals.sagepub.com/doi/10.1177/2993091X261492372

xCures Buyers Guide: https://xcures.com/interactive-buyers-guide/

Kill the Clipboard: https://www.cms.gov/health-tech-ecosystem/early-adopters/kill-the-clipboard

 

Subscribe for conversations about artificial intelligence, oncology, clinical medicine, and the future of healthcare.

 

Hosts:

Sanjay Juneja, MD: https://www.linkedin.com/in/sanjayjunejamd

Douglas Flora, MD: https://www.linkedin.com/in/douglasflora

 

Follow TensorBlack on X: https://x.com/tensorblack

Follow TensorBlack on Instagram: https://www.instagram.com/tensorblack

Follow TensorBlack on LinkedIn: https://www.linkedin.com/company/tensorblack/

Follow TensorBlack on Facebook: https://www.facebook.com/tensorblack

 

00:00 AI and Healthcare

00:30 Introduction

01:59 COI

03:01 AI and clinical documentation

04:40 Medicine runs on documents

06:04 OMOP, SNOMED, ICD, CPT: healthcare data languages

08:52 Fax machines and HIPAA

10:04 The media tab problem

11:12 How xCures started

12:03 Document structure before extraction

14:46 231 documents per cancer patient

16:19 Data science is 95% data prep

17:07 Clinical trial screening: an hour per chart

19:14 LLM extraction: enrichment and inference

20:48 AI and healthcare labor

21:58 Human in the loop: the three buckets

24:13 Why the journal published the paper

26:09 AI hallucinations: measure, don't worry

27:17 How to vet a healthcare data vendor

28:05 Dual extraction, model drift, buyer's guide

30:39 Healthcare AI ROI

31:50 EHR integrations vs. patient apps

32:31 Healthcare interoperability, solved in 2023

33:34 TEFCA, Individual Access Services, Kill the Clipboard

35:17 Patient-owned health data

36:00 Health records outside HIPAA

38:22 Digital pathology and health disparities

40:56 Every medical record is a patient

42:47 Data that changes a patient's life

 

#AIinHealthcare #MedicalRecords #CancerCare

Will AI replace doctors? Ezekiel "Zeke" Emanuel, MD, PhD, thinks autonomous AI will soon manage a lot of what physicians do today, starting with chronic disease and the side effects of chemotherapy, and that the better it gets, the more likely a human override is to add errors rather than catch them. His recent JAMA papers made the argument in print. In this episode of AI and Healthcare, oncologist Sanjay Juneja, MD sits down with Emanuel, an oncologist, bioethicist and Penn's Vice Provost for Global Initiatives, to talk through where autonomous AI in healthcare arrives first, where it will fail, and what is left for physicians when it does.

What if medical scans we already collect contain signals that could predict cancer treatment response, reveal disease risk years earlier, or identify biological patterns the human eye cannot see? In this episode of AI & Healthcare, oncologists Dr. Sanjay Juneja and Dr. Doug Flora, speak with AI and biomedical imaging researcher Anant Madabhushi, PhD about the hidden information inside routine CT scans, retinal images, and pathology slides. They explore how AI could help predict treatment response, detect warning signs of disease, uncover meaningful signals in tumor vessels and surrounding tissue, and expand access in regions facing shortages of radiologists and pathologists. The conversation also examines explainability, automation bias, clinical validation, health equity, and the consequences of waiting too long to use potentially valuable technology. Anant Madabhushi, PhD is the Robert W. Woodruff Professor of Biomedical Engineering at Emory University and Georgia Tech and Executive Director of the Emory Empathetic AI for Health Institute. His work focuses on AI, radiomics, computational pathology, and predicting treatment response from routinely collected medical images. Subscribe for conversations about artificial intelligence, oncology, clinical medicine, and the future of healthcare.

Medicine is becoming too complex for any one clinician to keep up with alone. AI may soon change that equation. Dr. Francisco Esteva joins Dr. Sanjay Juneja and Dr. Doug Flora to explain how physicians can use AI to research faster, communicate more clearly with patients, automate repetitive work, and build systems that extend their expertise. Francisco began learning these tools around age 60. Today, he's building AI-powered workflows for patient education, YouTube, research, quality control, and automation. You don't have to become an AI expert. But understanding how to use these tools may soon become an essential clinical skill.

CMS just proposed rules that could decide who gets paid for AI in healthcare, and that's a bigger deal than another “better algorithm” story. Dr. Sanjay Juneja and Dr. Douglas Flora talk with Jordan Johnson (radiation oncology, healthcare finance, compliance and law) about what's actually in the proposal and what happens once it lands. Topics: when does AI actually count as a billable clinical service. Why getting paid also means getting watched more closely. Hospitals leaning toward integrated platforms over one-off point solutions. What to do now, before the rule locks in. Still just a proposal, not law yet, so we should say that somewhere.