Computer Vision Startup Encord Launches DICOM Image Annotation Tool To Eradicate Manual Data Labelling for Healthcare Industry

• Encord’s platform empowers some of the world’s leading companies building computer vision models for medical imaging including Kings College London, Memorial Sloan Kettering Cancer Center and Stanford Medical Centre

• The first dedicated Digital Imaging and Communications in Medicine (DICOM) tool provides 3D annotation for modalities such as CT, X-ray, and MRI.

• The first DICOM tool for machine learning annotation with full pixel intensity rendering capabilities – 20,000+ pixels compared to just 256 in a regular browser canvas

Encord, the platform for data centric computer vision, has released the first purpose-built 3D annotation tool for healthcare AI that enables users to train and run models to automate medical-imaging annotation in 3D for modalities including CT, X-ray, and MRI.

The tool can render 20,000+ pixel intensities as opposed to existing tools that manage just 256 and expert label review functionality which is key for FDA approval processes.

Encord’s DICOM tool has been developed alongside clinicians and healthcare data scientists to deliver efficient functionality and a seamless user experience when annotating highly-specialised datasets allowing for maximum efficiency and usability for its end-users.

Encord is backed by CRV, Y Combinator, WndrCo and Crane Venture Partners and trusted by the likes of world-leading healthcare institutions including Kings College London where it helped to annotate pre-cancerous polyp videos resulting in increased efficiency by an average of 6.4x, and automated 97% of labels ultimately making the most expensive clinician become 16x more efficient at labelling medical images. It has also worked with Memorial Sloan Kettering Cancer Center and Stanford Medical Centre where it has reduced experiment duration by 80% and processed 3x more images.

Marketing Technology News: Aterian Adds Seasoned Industry Executive To Its Board Of Directors

“Instead, our tool allows healthcare AI firms to unlock the power of automation through deep-learning models. This reduces costs, increases efficiency and accuracy. We’re excited to see the positive impact this will have for our healthcare customers.”

Current methods rely on human interaction to prepare training datasets for AI use. Harnessing the power of automation through deep learning, the DICOM annotation tool replaces manual processes that make AI development expensive, time-consuming and difficult to scale. It allows users to achieve 100% data privacy and security as its platform is deployed to its existing systems rather than moving or sending the data externally.

Not only can Encord’s DICOM tool save time and money but it provides a data pipeline built around the images. Existing DICOM viewers allow annotation but are hard to export and work with and similarly, only a subset of data pipeline companies allow annotation of DICOMs however aren’t seen as a silver bullet.

Encord’s solution combines both the accurate and truthful display of DICOMs, allowing doctors to accurately annotate images, with the data pipeline to then work meaningfully with those annotations.

Marketing Technology News: MarTech Interview with Alexandre de Vigan, Founder & CEO at nfinite

“Existing options rely on outsourcing data to human labellers, including clinicians. Human error arising from this process leads to clinicians wasting time reviewing and correcting labels,” said Ulrik Stig Hansen, Co-Founder and CEO at Encord. “Instead, our tool allows healthcare AI firms to unlock the power of automation through deep-learning models. This reduces costs, increases efficiency and accuracy. We’re 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 features. It provides powerful automation features, such as model inference models and interpolation to help automate the annotation process.