How AI Reads Blood Cells to Help Diagnose Cancer with Gregory Goldgof, MD, PhD
Tuesday, October 27, 2026
Transcript intelligence
What OpenRounds heard
Source-grounded signals extracted from the episode transcript. Timestamps and supporting excerpts appear when available.
The speaker reports that their lab is automating screening for sepsis, clinical deterioration, and mortality by combining AI analysis of blood smear morphology with patient vitals and labs to generate a calibrated mortality risk score, targeting the ICU inpatient setting.
The speaker reports developing a scanning protocol for liquid specimens before AI implementation, noting the technical complexity of digitizing blood and bone marrow slides at scale.
“then i was like well i really want to bring this all the way to the clinic and to the patient um and to do that you need to have a digital service right you can't have ai if everything is glass and msk was famous for having uh digital service although they no one had a digital heat path service at the time because scanning the the liquid specimens is more complicated and i actually had to develop a protocol for scanning them before i could first i had to develop a protocol for scanning them scan a bunch”
In this episode of AI and Healthcare, oncologists Sanjay Juneja, MD and Doug Flora, MD sit down with Greg Goldgof, MD, PhD, MS, a computer scientist and board-certified pathologist, to talk about DeepHeme, the AI system his lab built to classify the cells in a bone marrow aspirate, the test at the center of diagnosing leukemia and other blood cancers, and what it took to get it validated and into clinical use. Gregory Goldgof, MD, PhD, MS is Director, Artificial Intelligence, Hematopathology Service, Memorial Sloan Kettering Cancer Center. He is a computer scientist and a pathologist board-certified in clinical pathology and clinical informatics. His lab builds AI tools for hematopathology, hematology and cytology.
Show transcript excerpt
In this episode of AI and Healthcare, oncologists Sanjay Juneja, MD and Doug Flora, MD sit down with Greg Goldgof, MD, PhD, MS, a computer scientist and board-certified pathologist, to talk about DeepHeme, the AI system his lab built to classify the cells in a bone marrow aspirate, the test at the center of diagnosing leukemia and other blood cancers, and what it took to get it validated and into clinical use. Gregory Goldgof, MD, PhD, MS is Director, Artificial Intelligence, Hematopathology Service, Memorial Sloan Kettering Cancer Center. He is a computer scientist and a pathologist board-certified in clinical pathology and clinical informatics. His lab builds AI tools for hematopathology, hematology and cytology.
Daily healthcare AI brief
Get the free healthcare AI briefing
One concise briefing built from reporting, research, policy, blogs, videos, and full podcast transcripts.
Free. Get a sign-in link by email to finish subscribing. An account is created if needed. Unsubscribe anytime. Already subscribed? Sign in to save and personalize.