Centaur.AI collaborated with Microsoft Research and the University of Alicante to create PadChest-GR, the first multimodal, bilingual, sentence-level dataset for grounded radiology reporting. This pioneering dataset aligns clinical text with annotated chest X-ray imagery, enabling AI models to justify diagnostic claims with visual references. This breakthrough enhances transparency and reliability in medical AI by moving beyond simple image classification to localized, interpretable findings, a critical step for safe AI deployment in healthcare.
Emphasized the importance of data curation practices in reducing bias in medical AI, promoting diverse datasets, expert collaboration, and fairness metrics for more equitable outcomes.
Centaur.AI’ latest study tackles human bias in crowdsourced AI training data using cognitive-inspired data engineering. By applying recalibration techniques, they improved medical image classification accuracy significantly. This approach enhances AI reliability in healthcare and beyond, reducing bias and improving efficiency in machine learning model training.
Expert feedback is essential for safe, effective healthcare AI, as emphasized in a Centaur Labs webinar featuring leaders from Google Health, PathAI, and Centaur.
This post explores the importance of DICOM in medical imaging and how Centaur Labs' integration with the OHIF viewer provides precise annotation tools for accurate medical AI development.
Understand why traditional labeling pipelines are hard to scale—and discover how our solution can 10X your pipeline faster, with greater accuracy and efficiency.
Centaur partnered with Ryver.ai to rigorously evaluate the accuracy of their synthetic lung nodule segmentations. Using our expert-led validation framework, we found Ryver’s synthetic annotations performed on par with human experts—highlighting synthetic data’s growing role in medical AI development.
Collaborated with leading researchers to assess biomedical LLMs, advancing AI’s ability to answer medical queries and simplify complex scientific concepts.
A $750,000 grant from the Massachusetts Life Sciences Center will support Brigham & Women’s Hospital researchers in their efforts to transform medical research.
Gamified data labeling enhances model accuracy from 70% to 93% in a case study with Eight Sleep, demonstrating the effectiveness of multimodal annotation.
Partnered with SciBite to accelerate vocabulary curation, cutting the timeline by over two months through expert crowd-labeling, achieving 90.3–95.1% accuracy.
Collaborated with VUNO to annotate brain MRI data, contributing to FDA clearance for VUNO Med®-DeepBrain®, an AI tool designed to assist in early dementia detection.
CEO Erik Duhaime discussed AI safety in healthcare with AI Unleashed, addressing challenges in data, model oversight, and the future of human-AI collaboration.
Centaur Labs’ scaled expert annotation of colonoscopy videos, achieving high throughput and consensus, dramatically enhanced the quality and speed of Satisfai Health’s GI AI development.
Announcing a new DICOM labeling experience and text highlighting features, designed to improve medical image annotation and support better healthcare outcomes.
Centaur.ai teamed Aiberry to annotate a new video dataset for mental health AI, boosting emotion detection and improving depression screening accuracy.
Worked with Volastra Therapeutics to annotate cancer cell images, supporting AI models in quantifying chromosomal instability and advancing cancer research.
Paige collaborates with Centaur.ai to enhance its algorithm, using high-quality data annotations to boost accuracy and performance in breast cancer detection models.
Centaur Labs contributes high-quality data annotations to enhance Consensus’ scientific search algorithm, improving accuracy and boosting research capabilities.
Learn how to automate your data pipeline with Centaur's end-to-end API integrations, streamlining workflows and enhancing efficiency for seamless data management.
Dandelion Health teams up with Centaur.ai to provide AI developers scalable access to high-quality clinical data, driving progress in healthcare technology.
The new AI-powered scientific search engine, Consensus, partners with Centaur.ai to generate high-quality, scalable scientific data labels for research.
From SMS to insurance claims, pathology reports, and scientific studies, this post explores the most common medical text datasets used for NLP in healthcare.
Learn about our partnership with Mayo Clinic spin out Lucem Health, and how clinical AI development teams can access high quality medical data annotations at scale.
Uncover the essence of Centaur Labs, a pioneer in combining human and machine intelligence for superior medical data labeling in the evolving healthcare landscape.
A Centaur Labs study found that disease prevalence and expert feedback significantly influence diagnostic accuracy in dermatology, highlighting the need for contextual data and ongoing guidance to reduce errors and improve clinical decision-making.
Learn more about how Centaur.ai is working with the Brigham and Women's Hospital team to develop multiple AI applications for point-of-care ultrasound.
Access dozens of open-source medical AI image datasets in formats like X-ray, CT, MRI, Ultrasound, Whole Slide Imaging, and more for research and training.
Examine the unique challenges of medical data labeling, why traditional methods fall short, and explore a more accurate, scalable alternative solution.
.webp)
.png)
