Gradient Health

🌎 Which Atlas is right for you?🌍 Having worked with hundreds of AI researchers on dozens of projects, we know that no two use cases are the same. That’s why we’ve created the Atlas suite. Whether you have a straightforward data need but tight budgets and timelines or a highly complex validation process, we have the right solution. Contact our team to find out how Atlas could accelerate your development. #MedicalImaging #AIinMedicine #DataSourcing

2024 is in the books!   What a year it has been! Building on the strong foundation of 2023, we launched our product suite, strengthened our team with top-tier experts, and expanded our capabilities through strategic partnerships and acquisitions.   Here are some of our highlights from 2024 👇   We created a game-changing product suite for AI developers: 🎉 Atlas Product Suite Launch: We unveiled Atlas, redefining data sourcing for medical AI developers. 🎉 We surpassed 11.5 million imaging studies (teaser: we just broke through another milestone TBA soon!)   We bolstered our offering through partnerships and acquisitions: 💡 Acquisition of DataAppraisal: Enhancing access to datasets from across the U.S. 💡 Partnered with Protege: Providing access to millions of de-identified medical images to accelerate AI development. 💡 Partnered with Flywheel: Offering analysis-ready datasets for real-world evidence use cases.   We connected with AI developers and data partners at global conferences: 🎙 Exhibited at RSNA 2024, hosting expert talks on creating unbiased AI and using data for good with Barbara McHugh and Pranav Rajpurkar. 🎙 Thanks to NIH sponsorship, we exhibited at #HLTH. 🎙 We also attended JPR 2024, ECR, HIMSS 2024, and the ACR Annual Meeting.   We launched our first webinar series with a range of guests, hosted by Rod B. Greco: 💻 First Europe, then the world: Featuring Julien Schmidt. 💻 Special holiday webinar: Featuring Brian Casey, Managing Editor at The Imaging Wire. 💻 First Atlas customer webinar: Featuring Remi Quilliet from GLEAMER.   We welcomed several exceptional team members this year: 🤝 Eris Gao (Software Engineer, Atlas Team) 🤝 Cal Nightingale (Software Engineer) 🤝 Ally Han (Business Operations Manager) 🤝 Tainá Moraes (Sales Development Representative) 🤝 Marilda Guimarães (Clinical Data Liaison) 🤝 Benji Meltzer (Special Advisor across product development, operations, and corporate projects)   We shared our story: 👉 Radiology Today featured our CEO Joshua Miller’s article Oncology Imaging: Pursuing Precision. https://lnkd.in/eVcGCAyk 👉 MedCity News published an insightful piece on ethical dilemmas in medical AI. https://lnkd.in/ejZ4SjJV 👉 We were recognized by Bessemer Venture Partners as a promising healthcare marketplace in their updated healthcare AI roadmap. https://lnkd.in/gyetmccN 👉 Joshua shared our journey on several podcasts, including: Planetary Health First Mars Next, Category Visionaries, Skelton Crew, and @VentureSouth! 👉 Our Co-founder Ouwen Huang, PhD, discussed Gradient Health's mission with Charlie Webb on the Smart Health pod.   A final note: 🫶 In light of our strong ties to Brazil, we launched a GoFundMe campaign to support flood victims in Rio Grande do Sul with essential aid.   Thank you to everyone who contributed to making 2024 such a successful year. Happy holidays!

We are incredibly proud to support the launch of ReXrank, an important benchmarking tool developed by Pranav Rajpurkar and co at the Rajpurkar Lab at Harvard Medical School’s Department of Biomedical Informatics. ReXrank is the first public leaderboard to objectively evaluate AI models designed for radiology report generation from chest X-rays, addressing a critical need in healthcare AI. 💡Why ReXrank Matters With nearly 1,000 FDA-cleared AI-enabled medical devices in the USA alone, healthcare providers face a challenge: how to compare AI models objectively. ReXrank is introducing a standardized, transparent framework to benchmark performance across AI models, separating genuine progress from hype. At Gradient, we contributed the ReXGradient dataset, the largest test dataset of its kind with 10,000 diverse chest X-ray studies, ensuring robust evaluation and real-world applicability for these models. 🔬 Key Features of ReXrank 👉 Evaluates AI systems on findings sections and complete reports with impressions. 👉 Leverages multiple datasets and 8 evaluation metrics for comprehensive performance analysis. 👉 Encourages transparency and open debate to drive progress in medical AI. "ReXrank’s large-scale test set provides the comprehensive evaluation framework needed to validate these technologies before clinical deployment and accelerate the development of reliable AI systems,” says Ouwen Huang, PhD, Gradient Health co-founder. Visit rexrank.ai to learn more and view the leaderboard! #Healthcare #Radiology #MedicalAI

💬 Key Feature: Access Our Team 💬    We know how important it is to get quick answers to your questions as you navigate the Atlas platform. Fast responses mean you can access your data without delay, helping you stay productive and focused on what matters.    That’s why we’ve built direct access to our dedicated support team right into Atlas. Just click on the icon in the bottom-right corner of your screen, and our chat feature will open, ready to connect you with our experts.   From here, you can contact our knowledgeable team, who are equipped to handle your questions and provide clear, efficient responses. Whether it’s a simple inquiry or a more technical question, our team is here to support you every step of the way.   We’re committed to providing the fast, effective support you need.    Get a demo of our platform to see all it can do: gradienthealth.io    #MedicalImaging #AIDeveloper #AIinHealthcare 

In case you missed it... Last week we were privileged to host Brian Casey for a Holiday Special, looking back at 2024 and predicting medical AI's role in 2025. Grab a hot cocoa and enjoy ☕️ #Webinar #MedicalAI #MedicalImaging

Join us for a special holiday webinar hosted by Rod B. Greco and featuring the GOAT, Brian Casey, Managing Editor at The Imaging Wire! Together, we’ll reflect on a whirlwind year of medical imaging AI in 2024, gain Brian’s unique perspective on RSNA 2024, and look ahead to 2025 as Brian shares some bold predictions for the future. Whether you’re a developer, researcher, or radiologist with an interest in the medical imaging AI space, this is your chance to hear from one of the most respected and knowledgeable voices in the field. #MedicalImaging #Radiology #AIinMedicine

It’s happening today! Join us for an unmissable webinar as Rod B. Greco and The Imaging Wire’s Brian Casey unpack 2024 in the world of medical imaging and share predictions for 2025. This is a must-watch for AI developers, researchers, and radiologists looking to stay ahead. ⏰ 10am PST | 1pm EST | 7pm CET 📍 Live on LinkedIn #MedicalImaging #Radiology #AIinMedicine

When it comes to developing effective medical AI, real-world data is essential. Here’s why: ✅ It accurately represents real patients: Real data captures the nuances and diversity of actual patient cases. This helps to create AI models that work across demographics and avoids oversimplification. ✅ It's essential for rigorous training and testing: Real data provides a solid foundation for testing models against real patient outcomes, helping to verify accuracy and reliability. ✅ It's crucial for regulatory success: Regulatory bodies require evidence that AI models are trained and tested on representative data, ensuring that the technology is safe, effective, and fit for real-world application. ✅ It better captures rare cases: One limitation of options such as synthetic data is the lack of unpredictable "outliers." These outliers—those rare, complex, or challenging cases—are essential to capture for ensuring a model is fully applicable to its intended audience. ✅ It mitigates bias risks: Large, representative datasets help reduce bias by covering a range of real cases, which synthetic data can struggle to represent accurately, especially when accounting for subtle patient differences. However, despite these advantages, some medical AI developers turn to synthetic data as an alternative, but why? Real data can be expensive, time-consuming to obtain, and there’s always a risk of confidentiality breaches when handling patient information. We created Atlas to overcome these three hurdles by offering high-quality, cost-effective, and quick access to millions of rigorously de-identified datasets for training and validating medical AI devices. 👉 Learn how Atlas can support your AI training needs: gradienthealth.io #MedicalImaging #MedicalAI #TrainingData 

🩻 New Feature Announcement: SNOMED CT Tags 🩻 One piece of feedback we've frequently received from our Atlas customers is the need to streamline searches while maximizing the clinical relevance of their data. That's why we've introduced a powerful new feature: alignment with SNOMED CT. SNOMED CT is the world's most comprehensive and precise clinical health terminology system. By integrating it into our platform, we're enabling customers to find clinically relevant data faster and with greater accuracy. How does it work? 👉 The advanced search functionality in our Atlas platform now includes SNOMED CT tags. 👉 As the user types, the platform has an autocomplete tool that suggests SNOMED CT tags, allowing them to locate relevant terms quickly. 👉 When a tag is selected, all related child tags in the SNOMED CT hierarchy are included, broadening search results and ensuring comprehensive data retrieval. Why is this important? 🎯 Faster and more precise searches: Customers can find relevant DICOM images and medical records faster by using clinically validated SNOMED CT tags. 🏥 Streamlined and enhanced clinical relevance: The autocomplete list and hierarchical tagging make it easier to capture all relevant terms without needing deep knowledge of the SNOMED CT structure. 🧑💻 Flexibility: Users can switch between the SNOMED CT-enhanced search and the standard text-based search, allowing them to adapt the search approach to different needs. This feature is live for existing customers, enhancing search precision and speed. New customers will also benefit from this improved search capability. Get a demo of our platform to see its full capabilities: gradienthealth.io #MedicalImaging #AIDeveloper #AIinHealthcare #RSNA2024

It's day two of RSNA 2024! Yesterday was incredible! The booth was buzzing, and Rod B. Greco and Barbara McHugh's talk about how Atlas helped Naitive create a safe and unbiased AI tool went incredibly well. If you want to get your hands on Atlas for yourself, come over and meet us at booth #5746 in the AI Showcase for an interactive demo.    #RSNA2024 #MedicalImaging #Radiology