Accelerate Development of AI Models for Gastroenterology Computer Vision

Encord's platform helps you improve video yield, produce better clearance-ready training data, and accelerate GI model development.

Part of an academic institution? Apply for access here.

Trusted by trailblazing AI teams at leading medical institutions

Stanford Medicine Logo
UHN logo
MSK logo
St Thomas logo
KCL logo
How it works

Build Better Machine Learning Models With Encord

Encord is the leading platform for annotating GI images and videos, monitoring annotator throughput and quality, and managing the expert review process.


Encord offers unlimited video clip length, complex ontologies, and AI-assisted video labeling across all types of diagnostic data, including capsule endoscopy, colonoscopy or endoscopy.

Gastroenterology Segmentation and Annotation for Training Artificial Intelligence Models

Encord is the only computer vision data platform that supports the labeling of videos of any length. For gastroenterology, this means any video can be uploaded, no matter its format or length, ensuring your model has the best datasets for detecting cancerous polyps, ulcers, IBS or any other condition.


And thanks to Encord’s AI-assisted image & video annotation and ability to pre-label data using your own models, you can quickly classify your video clips, label points of interest and tag artefacts.

Ontology of artefact, UCEIS score, and Mayo score for GI

Infinitely Complex Taxonomies

Whether you’re building a model based on UCEIS, MAYO, CANUKA or MMES, Encord’s taxonomy editor can help you get the data you need.


The taxonomies (ontologies) in Encord allow for multiple classifications and sub-classifications - as many as you need to get the data your gastroenterology model needs to deliver accurate diagnoses.

Automated quality control for GI videos

One Annotation Tool For All Labeling

Encord has multiple types of video annotation types within a single platform, including bounding box, polygon, polyline, keypoint, segmentation and classification.


This means your annotators and reviewers only have to use a single piece of software to do their jobs. And your developers don’t have to try and maintain multiple in-house tools to cover each type of annotation task.

KCL Logo
King's College London

KCL used Encord to achieve a 6.4x average increase in labeling efficiency for GI videos.

Problem
Using clinicians to annotate pre-cancerous polyp videos had prohibitively high costs to produce large datasets.
Solution
Deployed Encord's micro-model module to increase clinician labeling efficiency and automate 97% of produced labels.
Results
Highest expense clinician saw 16x labeling efficiency improvement. Cut model development time from 1 year to 2 months.
6.4X
Faster than manual labeling
97%
Automated labels
6X
Faster to AI in production
Encord supports FDA clearance and CE marking processes

Build a Foundation for FDA and CE Regulatory Clearance

Getting the evidence needed to get FDA or CE approval is hard and time consuming. But Encord offers the key features needed to make this process as streamlined as possible.


Put in place granular quality controls, define different sampling rates and set up custom review workflows for each class of label. And on top of that, all data with Encord is fully auditable, allowing you to easily show how you created the dataset your machine is trained on.

Encord improves the quality of your gastroenterology training data

Deliver Cleaner Endoscopy and Colonoscopy Datasets

Encord offers customisable dashboards to help clinical operations teams find the errors, biases and imbalances in the training data their annotation teams have created and reviewed.


Combined with Encord’s ability to easily tag artefacts in the videos (such as mucus, glare or air bubbles), clinical operations teams can be sure they’re providing the best datasets to their machine learning teams.

Expert review quality control for gastroenterology

Efficient Medical Grade Video Annotation

Encord is designed to work with your skilled gastroenterological annotators and reviewers, saving them time and improving their experience.


AI-assisted labeling makes the annotation task less of a chore and means annotators can accurately label more images. Annotation hotkeys helps to create an intuitive and efficient interface for your annotation team.

Accelerate AI model training in gastroenterology