Namkee Oh, MD

Minimally invasive liver surgery | Transplantation surgeon | Surgical data science |

[Namkee Oh, MD @NamkeeOh]

The OR is sensor-rich, yet data remain unsynchronized, unstructured, or discarded.

Our review argues that physical AI in surgery needs a Surgical Data Factory: a closed loop to capture, structure, and use operative data.

Surgeons must help build it.

https://t.co/t69AAUB59v

[Phosphen @phosphenq]

He taught machines to learn. He won a Turing Award. Now he says we're building AI all wrong.

"We want a path towards intelligence that isn't limited by human abilities."

In 71 minutes at MIT, Richard Sutton reveals what comes after the chatbot.

Self-taught + play + abstraction + superintelligence

Worth more than a year of AI roadmaps on your timeline

[Alexi Gladstone @AlexiGlad]

Progress in AI is driven by approaches that make weaker assumptions, which allows for better scaling

But representation learning has relied on strong assumptions like augmentations, masking, cropping, etc... until now!

🎬 Introducing Temporal Difference in Vision (TDV), a new paradigm for representation learning built on a single assumption: causality

TL;DR:
- We introduce TDV, the first approach to learn good representations without any augmentations, masking, cropping, or pixel-based reconstruction
- TDV matches SOTA recipes like DINO and iBOT on dense spatial tasks
- We show that as data scales, weaker assumptions work better

🧵Thread:

[Namkee Oh, MD @NamkeeOh]

기술은 의학의 발전을 가속화 시킬 수 있습니다. 특히, 수술은 기술의 도움으로 개선될 여지가 큽니다.
이에 대한 제 생각을 정리해보았습니다.

https://t.co/0SmxFm38jQ

[Namkee Oh, MD @NamkeeOh]

I'm pleased to share our recent publication in the Journal of Medical Internet Research (JMIR): "Large Language Model–Assisted Surgical Consent Forms in Non-English Language: Content Analysis and Readability Evaluation"
https://t.co/sem0BTKDAl

[Moo Jin Kim @moo_jin_kim]

Introducing OFT—an Optimized Fine-Tuning recipe for VLAs!

Fine-tuning OpenVLA w/ OFT, we see:
-25-50x faster inference ⚡️
-SOTA 97.1% avg SR in LIBERO 💪
-high-freq control w/ 7B model on real bimanual robot
-outperforms π₀, RDT-1B, DiT Policy, MDT, Diffusion Policy, ACT

🧵👇

[Namkee Oh, MD @NamkeeOh]

@chester_roh 깊은 통찰을 공��해주심에 늘 감사합니다!

[Andrew J. Hung, MD, FACS @AjhungMD]

Our latest paper on "Human AI collaboration for unsupervised categorization of live surgical feedback" has arrived at Nature Digital Medicine @npjDigitalMed

Led by our postdoc @RKocielnik @caltech this paper explores how live surgical feedback in the OR can be evaluated by AI and humans to best predict when meaningful change will occur by the training surgeon.

Open access here:
https://t.co/TLGLp3o7dh

Funded by @theNCI
Collaboration with @AnimaAnandkumar

[Namkee Oh, MD @NamkeeOh]

It was an honor to present at CLINICCAI during my first-ever MICCAI conference. I shared our work on developing an AI-based navigation system for enhancing liver surgery safety.
#MICCAI2024 #CLINICCAI #AIinSurgery

[Namkee Oh, MD @NamkeeOh]

This journey has been a proof of the power of dedication and continuous learning. We hope that our research will help advance the field of surgical AI and ultimately improve patient outcomes.

[Namkee Oh, MD @NamkeeOh]

I'm happy to announce our research team's latest paper, 'Real-time segmentation of biliary structure in pure laparoscopic donor hepatectomy', has been published in Scientific Reports (https://t.co/kheKmUIcdQ).

[Namkee Oh, MD @NamkeeOh]

Data preprocessing was not easy, and annotation took countless hours. The subsequent training of the segmentation model also involved many trials and errors.

[Cyril Zakka, MD @cyrilzakka]

Super excited to introduce SAM2 Studio! 🚀🤖

I've been getting a lot of questions lately on supporting AI inference tailored for patient data and sensitive workflows.

We optimized SAM2 to run completely on-device in real time for all of your medical segmentation workflows - including surgical video segmentation, radiographs, pathology slides and more!

[Namkee Oh, MD @NamkeeOh]

수술과 인공지능.
인공지능이 환자에게 어떻게 도움을 줄 수 있을까.
수술하는 사람으로서 생각과 실제 연구사례를 소개드립니다.

https://t.co/8Dd7sLzqy3

[Hong-Yu Zhou @HongYuZhou14]

We announced MedVersa, a generalist AI that excels in multifaceted medical image interpretation! 🚀🩺

👉https://t.co/AnibKW0rdO

MedVersa has two promising features🧐:

1. Learning from vision and language supervision. This maximizes the flexibility of the framework. Imagine combining the SAM and CLIP into a unified model and the mutual benefits it will bring!🤪

2. Leveraging existing modules/tools. MedVersa functions the LLM as a TRAINABLE orchestrator. This design features a notable level of extensibility, allowing integration with advanced modules/tools.🤩

A great journey with @subathraadithan and @jn_acosta
, led by @EricTopol and @pranavrajpurkar

Stay tuned for more updates!👀

#artificiallyinteligence

[Eric Topol @EricTopol]

Our comprehensive review of #AI in surgery is just out @NatureMedicine
https://t.co/zDuBUUJn4K
an inspiration and joy to work with super-talented @chrisvarghese98 @ogrady_greg @ewenharrison on it
Free access https://t.co/mEVDkPpQT3

[npj Digital Medicine @npjDigitalMed]

A crowdsourcing approach to obtain high-quality tissue annotations, bypassing the need for costly expert input is presented in this article. With real-time deployment, this multimodal #AI model promises safer and more efficient #colorectal surgery.

https://t.co/WIyibDWIi5

[Namkee Oh, MD @NamkeeOh]

3D auto-segmentation of biliary structure of living liver... : International Journal of Surgery https://t.co/C8R7dm6Hfg

[Namkee Oh, MD @NamkeeOh]

My colleagues and I believe deep learning can improve clinical outcomes for patients. These 3D automated segmentations provide a better understanding of biliary anatomy, and meticulous bile duct division can lead to better outcomes. @IJSurgery

[Namkee Oh, MD @NamkeeOh]

These 3D automated segmentations provide a better understanding of biliary anatomy, and meticulous bile duct division can lead to better outcomes.