Chad Vanderbilt, MD (@VanderbiltChad) profile picture

Investigator @MSKCancerCenter. #Pathologist. Research focuses on computational methods in diagnostics. Tweets focus on AI research in pathology.

Joined February 2022
Introducing our weekly course! Large Vision Models: From Foundations to Applications is live. 🚨 The first lecture is Large Vision Models: Contrastive, Triplet & InfoNCE Loss A new lecture is posted every Friday morning, New York time. The introduction is up now, and notes go out with each video. lvm.artificialintelligencepa… We start with self-supervised learning theory and implementation (i.e. techniques used to train a pathology foundation model) taught by Siddharth Singi. Next are the pipelines we built in-house to fine-tune those encoders on real tasks taught by myself and Neeraj Kumar, Ph.D.. Then we cover modern LLM pipelines, including Jev, and how we use them to extract useful information from clinical data taught by Ali Kamali, PhD. We finish with clinical applications, and what it takes to carry a model into practice including lectures by Dr. Greg Goldgof, MD, PhD, MS on applications in hematopathology. No background in machine learning or pathology is required. The series is the path from a foundation model to a system that can be used on patients. A special thank you to Siddharth Singi for spearheading this initiative. We are dedicated to making this knowledge widely known to supercharge the future of AI applications in medical setting. We are looking forward to seeing you on Friday mornings.
1
5
47
🚨 AI at Scale for Clinical Impact (ASCI) Workshop Accepted Paper Announcement: 36 accepted papers. Ten talks. Twenty-six posters. The full program for ASCI, a NeurIPS 2026 workshop, is now live at asci.artificialintelligencep… Archived manuscripts will be posted in the Proceedings of Machine Learning Research on the day of the conference. Congratulations to everyone whose work was accepted! The work in this field is progressing so quickly. Thank you to the authors who submitted and to the program committee. The quality of submissions was impressive and all of the spotlights are pushing the state-of-the-art in this field. We are looking forward to the talks and posters in Sydney. Keynotes and Invited Speakers: Thomas Fuchs, Hoifung Poon, Piotr Bojanowski, Jorge Reis-Filho, Jana Lipkova, Ruijiang Li, Jakob Nikolas Kather, Bo Wang, Nigam Shah, Danielle Bitterman Organizers: Chad Vanderbilt, Neeraj Kumar, Ph.D., Gabriele Campanella, Ruchika Verma, Ph.D., Jia Wu, Rukhmini Roy Muhammad Waqas Sheikh, PhD, H.R. Tizhoosh, Saghir Alfasly, Wataru Uegami, Kostas Triaridis, Joel Saltz, Katherine Hoadley, PhD, Ali Kamali, Siddharth Singi, Dr. Greg Goldgof, MD, PhD, MS
1
1
76
To all of the healthcare AI researchers out there, we are hosting the AI at Scale for Clinical Impact (ASCI) workshop at NeurIPS 2026. asci.artificialintelligencep…
Replying to @NeurIPSConf
@NeurIPSConf 2026 workshop call for papers - AI at Scale for Clinical Impact: Cancer Pathology Foundation Models - at @ICCSyd , Australia (Dec 11–12). asci.artificialintelligencep… In-depth coverage of AI for pathology and cancer care! @Ruchika__Verma @VanderbiltChad @ThomasFuchsAI
268
Chad Vanderbilt, MD retweeted
Strong discussions at our Team Endometrium Research Retreat yesterday, centered on Integrating and Advancing AI-Based Research. Appreciative of everyone who contributed with thoughtful presentations, meaningful questions, and a collaborative atmosphere throughout!
4
1
10
663
Chad Vanderbilt, MD retweeted
In a silent trial involving 197 lung cancer patients over the span of four months, a pathology foundation model was able to identify presence of EGFR mutations with high accuracy, potentially preventing further genetic testing in 43% of cases. nature.com/articles/s41591-0…
8
1
31
5,001
Chad Vanderbilt, MD retweeted
Nature paper: nature.com/articles/s41591-0… GigaPath is available on HuggingFace and GitHub github.com/prov-gigapath/pro… and easily deployable to the cloud via Foundry: ai.azure.com/catalog/models/… Foundry model catalog includes leading open source and proprietary health and life science AI models like GigaPath that can be connected to healthcare specialized agents: microsoft.com/en-us/industry…
1
1
114
【EGFRを病理スライドからAIで予測】 nature medicine ✓AIモデル(EAGLE)によりEGFR変異をHEスライドから高精度(AUC 0.890)で予測できた ✓迅速(44分)に結果が得られ、貴重な組織を温存できる ✓本質的にデジタルなので遠隔地への展開も可能 doi.org/10.1038/s41591-025-0…
1
1
103
Chad Vanderbilt, MD retweeted
Wow - the authors map NPV/PPV thresholds that let hospitals drop 25 - 43 % of rapid EGFR tests without losing accuracy (Fig4). AI is changing digital pathology forever.
1
2
161
Chad Vanderbilt, MD retweeted
In a silent trial involving 197 lung cancer patients over the span of four months, a pathology foundation model was able to identify presence of EGFR mutations with high accuracy, potentially preventing further genetic testing in 43% of cases. nature.com/articles/s41591-0…
1
6
18
4,468
Study reported in @NatureMedicine nature.com/articles/s41591-0…, showed that AI can accurately predict genetic mutations from routine pathology slides. mountsinai.org/about/newsroo… @IcahnMountSinai
3
1
234
Manuscript is published in Nature Medicine today. Details: linkedin.com/posts/activity-…
27
Chad Vanderbilt, MD retweeted
🚨 #NeurIPS2025 Competition: Self-supervised Learning for Cancer Pathology Models 🧬 Train models on 300M+ tiles across 39 cancers. Evaluated by MSKCC, Mayo, MD Anderson, Sinai & more. 🔗 deeplearningpathology.org All @NeurIPSConf comps: blog.neurips.cc/2025/06/27/n… @VanderbiltChad
1
1
1
470