Tumor and lesion segmentation datasets

Create precise region labels for tumors and lesions in supported medical images. Keep contours, masks, and expert review connected to the original imaging study.
Try for free
Talk to sales
Tumor & Lesion Segmentation example with source-data annotations and contextual photographs.

Define lesion regions with a consistent protocol

Build supervised segmentation data around the boundaries and classes your model must learn.
Brain MRI image with a purple lesion-region segmentation mask.

Brain lesion regions

Label lesion contours and masks on brain MRI according to your team’s imaging protocol.
Chest CT with a small lung lesion region annotated.

Pulmonary lesion regions

Mark visible lung lesion regions on chest CT with the classes and boundary rules required by the dataset.
Abdominal CT with a focal lesion region annotated.

Abdominal lesion regions

Annotate liver or other abdominal lesion regions while retaining the surrounding anatomy for context.
Chest CT with separate lung and lesion region labels.

Lesion and surrounding anatomy

Label the lesion and neighboring anatomical regions separately when both are required by the segmentation task.

Why AI Teams Choose Unitlab

Bring medical imaging annotation, shared ontologies, expert review, and dataset delivery into one workflow so your team can focus on useful training data.
15X
Faster Medical Imaging Annotation
60%
Free Up AI Engineers’ Time
5X
Lower AI Development Costs

Annotation types for tumor & lesion segmentation

Use image and volume annotation tools to define findings and anatomical regions under a shared labeling protocol.
Brain MRI image with a purple lesion-region segmentation mask.

Lesion polygons

Trace the visible boundary of a lesion with a labeled contour under your task’s annotation criteria.

Lesion masks

Create segmentation regions that describe the lesion class and retain the geometry needed by your training task.

Brain MRI image with a purple lesion-region segmentation mask.

Tumor & Lesion Segmentation FAQs

What is tumor and lesion segmentation annotation?

It marks the regions belonging to a tumor or lesion so a supervised model can learn the boundaries specified by an expert annotation protocol.

Can I create different tumor or lesion classes?

Yes. Define the classes and properties required by your dataset, with clear rules for how annotators distinguish the categories.

Should I use contours or masks?

Choose the geometry required by your model and annotation protocol. Contours express boundaries, while masks express the segmented region.

Can annotators inspect surrounding anatomy?

Yes. The medical viewer keeps labels in source image context, and supported volumes provide the available anatomical views for inspection.

Does the platform automatically determine tumor size or response?

The workflow creates and reviews lesion annotations for your downstream analysis. Measurement and response criteria should be defined and validated by your research team.

How should uncertain lesion edges be handled?

Document the boundary rule, flag uncertain regions with comments or properties, and use expert review to settle the final annotation.

Which medical imaging formats are supported?

Unitlab supports DICOM, NIfTI, and NRRD inputs. DICOM slices are grouped by series into an annotation volume, with available views determined by the imaging data.

How do specialists review medical imaging labels?

Use a review stage to inspect labels against the source images, record comments, and return corrections through the workflow. Your team defines the annotation protocol and expert review criteria.

Can annotated imaging datasets be released for research?

Yes. Versioned releases and supported exports let teams deliver reviewed annotations to downstream pipelines. Select an export that preserves the geometry and metadata your research task requires.

Related resources:Medical annotation platformOrgan & Anatomy SegmentationMedical Image ClassificationAnnotation quality assuranceUnitlab documentation
Need help designing a tumor & lesion segmentation workflow?Talk to Unitlab
Unitlab on InstagramUnitlab on YouTubeUnitlab on FacebookUnitlab on LinkedInUnitlab on X
©2026 Unitlab, Inc. All rights reserved.