Oge Marques

📚 New Book Announcement: "AI for Radiology" 📚 I hope my latest book will become a valuable resource for medical professionals, computer scientists, and anyone interested in AI's clinical applications. routledge.com/AI-for-Radiolo… #AI #Radiology #MedicalImaging

NVIDIA’s Graduate Fellowship Program is now accepting applications for the 2026–2027 academic year. Selected Ph.D. students receive tuition and stipend coverage up to $60K, plus mentorship and technical support from top NVIDIA researchers during an NVIDIA internship. If you’re advancing work in AI, robotics, computer graphics, autonomous vehicles, healthcare, HPC, or related fields — this is your moment. 📅 Apply by Sept. 15, 2025: nvda.ws/3UsbEpe

We're hiring! We have openings in 9 subspecialties, including Neuroradiology, Breast Imaging, IR, Pediatrics & MSK. Join a leading academic department committed to clinical excellence, research & education. radiology.ucsf.edu/academic-affai…

Our 2022 paper on #ai and #medical #education has reached the 100-citation milestone! scholar.google.com/citations?view…

Our paper on A.I. agents for healthcare just published @natBME @pranavrajpurkar Michael Moritz free access rdcu.be/efW1n

📣 Announced at #GTC25: NVIDIA cuML now accelerates scikit-learn, the most popular machine learning library, up to 50x with zero code changes. Zero code change acceleration is also supported for HDBSCAN and UMAP. Read the announcement blog: nvda.ws/4id2Nlu

You heard it at #SC24, NVIDIA #BioNeMo Framework is now open-source. Researchers worldwide can accelerate the development of life-saving treatments by tapping into the collection of accelerated programming tools for biomolecular research. #drugdiscovery bit.ly/48Z67wQ

Hugging Face and NVIDIA are teaming up to advance robotics. By combining Hugging Face's LeRobot with NVIDIA AI, Omniverse, and #robotics technology, we're enabling researchers & developers to innovate. #CoRL2024 bit.ly/3UK7glT

Learn more about the new NVIDIA AI Blueprint for video search and summarization, which speeds up the development of visual AI agents. It uses #VLM, LLMs, and the latest RAG techniques for long-form video understanding. bit.ly/3YW72L6

Congratulations to all NVIDIA partners and Inception members featured in Time Magazine's Best Inventions of 2024. bit.ly/3YAlk2n

Discover how to build powerful multimodal visual AI agents with NVIDIA NIM. Get hands-on with Jupyter notebooks for applications like streaming video alerts, structured text extraction, multimodal search, and few-shot classification. bit.ly/3CeWP3p

🔬 New research from @UTSWMedCenter creates a 4D deep learning #AI model for identifying #breastcancer metastasis using only MRI and clinical data, for early detection without invasive biopsies.🩺 🔗 Learn more: nvda.ws/3YppB8V bit.ly/3NOdWeM

🙌 Join a series of free, virtual faculty development workshops w/ training in #deeplearning, #generativeAI and #LLMs, data science, & accelerated computing. Register now ➡️ events.nvidia.com/faculty-develo… bit.ly/48agDAQ

See the NVIDIA technical blog on how we are optimizing the 🦙 Llama 3.2 collection of open models with NVIDIA NIM microservices to accelerate flexible #AI experiences: developer.nvidia.com/blog/deploying…

Experience high-efficiency NVIDIA Llama-3.1-Nemotron-51B - a NAS-optimized model achieving 2x throughput while preserving accuracy runs on a single H100 GPU. Technical deep dive ➡️ bit.ly/4gE6ekA

Try Hive’s AI-Generated Image Classification model, the latest powerful image classification tool for identifying and managing AI-generated content, on ai.nvidia.com today. bit.ly/3zD7jII

Leading U.S. medical centers and research institutes are tapping into federated learning to detect cancer. This research collaboration uses NVIDIA FLARE and MONAI for secure, flexible AI model training for tumor segmentation. #NVIDIAAcademicGrant bit.ly/4exjZQl

Generate code with Abacus AI’s Dracarys Large Language Model Dracarys, fine-tuned from Llama 3.1 70B and available from NVIDIA NIM microservice, supports a variety of applications, including data analysis, text summarization, and multi-language support. bit.ly/4e7eLLo

Generate images with consistent characters without fine tuning or training. Consistory maintains subject consistency between text-to-image generations on pretrained models. Explore: build.ngc.nvidia.com/nvidia/consist… bit.ly/4gqsEFZ

📚 Introducing the new Generative AI Teaching Kit for educators developed w/ Professor Sam Raymond from @dartmouth. Enable students to explore the field of #GenAI w/ hands-on experimentation in #LLMs, diffusion models, & more. bit.ly/4e7Ae6o

💫Excited to share our new study: “Uncovering Knowledge Gaps in Radiology Report Generation Models through Knowledge Graphs”. We’ve developed a system ReXKG to extract structured information from radiology reports, building a comprehensive knowledge graph for in-depth model analysis. Our study compares AI-generated vs. human-written radiology reports, evaluating both specialist and generalist models.