Tutorial

How to Annotate 3D Medical Scans on Datature Nexus

https://www.youtube.com/embed/Qtb4FJlh4xs

Annotating 3D medical data is different from labeling 2D images. You are working across hundreds of slices, switching between axial, coronal, and sagittal views, and marking structures that span the full volume. This tutorial shows how Datature Nexus handles 3D medical annotation with tools built for volumetric data.

What This Tutorial Covers

  • Opening a 3D volume in Datature's multi-planar reconstruction (MPR) viewer
  • Navigating axial, coronal, and sagittal planes
  • Drawing segmentation masks across slices
  • Using interpolation to speed up annotation between key slices
  • Managing labels and reviewing annotations in 3D

Why 3D Annotation Needs Specialized Tools

Standard image annotation tools treat each slice as an independent image. That makes it painful to maintain consistency across a volume. Datature's annotator links slices together so your labels stay coherent in 3D space. The MPR viewer lets you cross-reference structures across all three planes before committing a label.

Who This Is For

Radiologists and medical imaging researchers who need to label DICOM or NIfTI volumes for model training. ML engineers preparing ground truth data for organ segmentation, tumor detection, or anatomical landmark models.

Go Deeper

Video Description Lorem Ipsum

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua.Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua.

Resources

More reading...

Building VLMs for Phrase Grounding with Datature Vi
January 14, 2026
Datature Vi

Build a vision-language model for phrase grounding on Datature Vi. Annotate multimodal data, configure a VLM workflow, train, and run inference.

Read
Improving Your Computer Vision Models with Metadata
July 1, 2025
Explained

Improve model accuracy by adding metadata to your training pipeline. Learn how camera settings, timestamps, and sensor data boost CV predictions.

Read
Class Imbalance in Computer Vision, Explained
June 6, 2025
Explained

Learn why class imbalance hurts model performance and how to fix it. Covers oversampling, weighted loss functions, focal loss, and augmentation strategies.

Read
Get Started Now

Get Started using Datature’s computer vision platform now for free.