Video Journal of Oncology (VJOncology)’s Post

Virchow2 has been ranked #1 on the Eva leaderboard, a leading evaluation framework for oncology foundation models by kaiko.ai. Eva rigorously assessed 13 foundation models on 6 WSI-classification tasks using a fixed protocol that prioritizes fairness and comparability over state-of-the-art performance. Virchow2 achieved state-of-the-art performance across classification and segmentation tasks for various tissue types, including breast, colorectal, kidney, lung, and prostate. What sets Virchow2 apart? ☑️ Trained on 3M whole slide images of H&E and IHC samples ☑️ Learning across 5x, 10x, 20x, and 40x magnifications ☑️ Available as a commercial or an open-weights model via Hugging Face and Azure AI Studio Explore Virchow2 and learn why it’s setting a new standard in the field: 📈 Leaderboard: https://lnkd.in/ghUY6zyR 💻 Hugging Face: https://lnkd.in/gTG_j3bw ☁️ Azure AI Foundry: aka.ms/Virchow2

🌟 Bridging the gap between technology and pathology: insights from the Congress of the Italian Group of Pathology for Head and Neck #Cancer  Our Product Manager Francesco Martino recently participated as a discussant at the Congress of the Italian Group of Pathology for Head and Neck Cancer, where the conversation centered around two transformative topics: Telepathology and AI in #Pathology. Here are some key takeaways from the event: 🔍 Legislation vs. Reality: There’s a clear gap between legislation and real-world applications of #telepathology solutions. As always, the law is struggling to keep pace with technological advancements. 🤖 AI needs explainability: #pathologists need more transparency and validation in AI algorithms to build trust. The consensus? AI models should not only be validated through metrics but also through methods like Grad-CAM, which help clarify the decision-making process. 📋 Pathologists are always in control: even with #AI assistance, the pathologist remains responsible for the final diagnosis. That’s why having medical-certified algorithms is non-negotiable. 🗝️ Integration is key: Laboratory Information Systems (#LIS) are central to pathology workflows, and integrating telepathology solutions into the LIS is crucial to provide comprehensive insights in one unified platform. #GIPATecCongress #HeadAndNeckCancer #InnovationHealthcare

Artificial intelligence (AI) is reshaping pathology, revolutionising how diseases are detected and diagnosed. Jonathan Bridges, Chief Innovation Officer at Exponential-e, shares insights in the Journal of mHealth on how AI is empowering pathology departments with enhanced speed and accuracy in identifying conditions like breast, bowel, and cervical cancer. While these AI advancements are promising, Jonathan also addresses the critical security and governance challenges that must be overcome to safely integrate AI in medical settings. Read more about this shift in diagnostics and how it could change patient care: https://lnkd.in/eDU6Hy24 #HealthcareInnovation #AIDiagnostics #PathologyAI #DigitalTransformation #HealthcareTechnology #mHealth

AutoRadAI, an AI pipeline, demonstrates remarkable efficacy in detecting extracapsular extension (ECE) in prostate cancer (PCa) using multiparametric MRI. With a dataset of 1001 patients, including 510 pathology-confirmed ECE cases, the model achieved an AUC of 0.92 for image quality and 0.88 for ECE detection. See link to study in the comments below 👇 By integrating deep learning models like ProSliceFinder and ExCapNet, AutoRadAI significantly outperforms traditional radiology reports, achieving an accuracy (ACC) of 0.82 compared to radiologists' 0.61. This innovation not only enhances diagnostic accuracy but also supports nuanced clinical decision-making, potentially improving patient outcomes and streamlining treatment planning. The tool’s user-friendly web application facilitates rapid ECE identification, making it an indispensable asset in clinical settings. We look forward to see how AutoRadAI can transform clinical practice. #AI #DeepLearning #Medicallmaging #ProstateCancer #MultimodalAl #Innovation 👉 Follow xCures Read our LinkedIn Newsletter: https://lnkd.in/dnNJV2ti https://xcures.com/ 👀

[#Lunit #Oncology] If you are attending the Translational Digital Pathology & AI Summit next week, do not miss the chance to learn more about #AI in #DigitalPathology! Mark your calendar for Wednesday, April 17, 2pm for our presentation: "Analyzing Pathology via AI Models to Augment Drug Development" You can also meet with our team onsite and learn how AI can help achieve your goals. Ben Goodwin Deboleena Sarkar, Ph.D. Michael Hreczuck Click on the link to register and learn more about the event 🔗 https://lnkd.in/gRgdWhAb #translational #digitalpathology #drugdevelopment #pathologyai #event

AI has transformed many industries in a short period of time. How is it transforming medical research? Tomorrow, join Stanley Bishop as he explains what scientific foundation models are and how they can change the research arena on generic diseases and oncology. Register here: https://buff.ly/3yxGEwD #LiveEvents #AI #GenerativeAI #AIAndHealth #AIResearch

🧬 𝐀𝐈-𝐃𝐫𝐢𝐯𝐞𝐧 𝐏𝐫𝐨𝐠𝐧𝐨𝐬𝐭𝐢𝐜𝐬 𝐢𝐧 𝐎𝐧𝐜𝐨𝐥𝐨𝐠𝐲 🌟 🔍 Precision at Its Best: AI leverages massive datasets to predict cancer progression with unprecedented accuracy, tailoring treatment plans like never before! 🎯 ⏱️ Time-Saver: Advanced algorithms minimize diagnostic timelines, allowing clinicians to make quicker, data-driven decisions for optimized patient care. ⏳ 🤝 Collaborative Intelligence: Oncologists and AI systems working hand-in-hand create a symbiotic relationship, enhancing understanding of complex cancer pathways. 👫🤖 📈 Continuous Learning: Machine learning models evolve with every new data input, constantly refining their prognostic capabilities and staying ahead in oncology research. 📊 For cutting-edge insights and comprehensive literature reviews in biomedicine, explore https://www.sciqst.com, your powerful tool for navigating the vast world of research! 🌐✨ #AIInOncology #CancerResearch #MedicalInnovation #FutureOfMedicine

📣 Partnership Announcement 📣 We're partnering with Protege, the platform for AI training data, to provide developers with access to millions of de-identified medical images, accelerating the creation of AI models in healthcare.    Why is this important? 👉 Our comprehensive imaging datasets will now be available through Protege, enabling the creation of multimodal, longitudinal datasets.  👉 This is especially valuable in oncology, radiology, and chronic disease management, where visualizing the evolution of a condition alongside clinical and contextual data can greatly improve diagnostic precision and treatment planning.    Read more about the partnership: https://lnkd.in/e3HjjSZ8 #PartnershipAnnouncement #MedicalImaging #TrainingData

This week, AI technology has taken a leap forward in healthcare. Researchers have developed AI software that can interpret mammograms and detect breast cancer 30 times faster than human doctors with 99% accuracy, reducing the need for unnecessary biopsies. This is a game-changer for early detection and treatment. Learn more about the potential of AI in healthcare here. #AI #Healthcare #Innovation"