LONDON–(BUSINESS WIRE)–Encord, the data-centric computer vision platform, has released the first 3D annotation tool purpose-built for medical AI that allows users to train and run models to automate the 3D medical image annotation for modalities such as CT, X-ray, and MRI.
The tool can render intensities of over 20,000 pixels, unlike existing tools that only support 256 pixels and expert label review functionality, which is essential for FDA approval processes.
Encord’s DICOM tool was developed in collaboration with clinicians and healthcare data scientists to provide efficient functionality and a seamless user experience when annotating highly specialized datasets enabling efficiency and ease of use. maximum for its end users.
Encord is backed by CRV, Y Combinator, WndrCo and Crane Venture Partners and trusted by world-renowned healthcare institutions including Kings College London, where he helped annotate videos of precancerous polyps, resulting in increased 6.4x efficiency on average, and automated 97% of labels, making the most expensive clinician 16x more efficient at labeling medical images. He has also worked with Memorial Sloan Kettering Cancer Center and Stanford Medical Center where he reduced experiment time by 80% and processed 3x more images.
Current methods rely on human interaction to prepare training datasets for use by AI. Harnessing the power of automation through deep learning, the DICOM annotation tool replaces manual processes that make AI development costly, time-consuming, and difficult to scale. It allows users to achieve 100% data privacy and security when its platform is deployed on its existing systems rather than moving or sending the data outside.
Not only does Encord’s DICOM tool save time and money, it provides a data pipeline built around the images. Existing DICOM viewers allow annotation, but are difficult to export and use. Similarly, only a subset of data pipeline companies allow DICOM annotation, but they are not considered a silver bullet.
Encord’s solution combines both the accurate and truthful display of DICOMs, allowing physicians to accurately annotate images, with the data pipeline to then work meaningfully with those annotations.
“Existing options rely on outsourcing data to human labelers, including clinicians. The human error resulting from this process leads to clinicians wasting time reviewing and correcting labels,” said Ulrik Stig Hansen, co-founder and CEO of Encord. “Instead, our tool enables healthcare AI companies to unlock the power of automation through deep learning models. This reduces costs, increases efficiency and accuracy. We are excited to see the positive impact this will have for our healthcare customers. »
The DICOM annotation tool will slot into Encord’s existing platform, taking full advantage of the rest of its data pipeline functionality. It provides powerful automation features, such as model inference patterns and interpolation to help automate the annotation process.
Visit https://encord.com/dicom for more information.
Encord is the leading computer vision data platform. Our solution enables machine learning and data mining teams to annotate, manage and evaluate training data. Encord teaches computers to perform practical, real-world tasks more efficiently in all future technologies. Encord eliminates complexity and enables data teams to work more efficiently.
Learn more about Encord at https://encord.com/ or @encord_team on Twitter.