GetDATA

The medical data marketplace for crowdsourcing datasets and labels.

GetDATA connects AI teams with hospitals, labs, and verified annotators. Post a request with modality, label specs, regions, and budget. Providers upload de-identified data, experts and trained annotators label in-platform. Every contribution is scored for quality and compliance.

  • HIPAA/GDPR-ready
  • Secure submissions
  • Quality scoring (data & labels)
  • Global providers
Live preview
Request
100k Chest X-ray (DICOM) + radiologist labels
Modality: X-ray · Format: DICOM + CSV/JSON labels · Regions: EU/US · Budget: $300k
Submissions
Hospital AQuality 92%
Imaging Lab BQuality 85%
Medical Center CQuality 78%
Indipendent provider DQuality 71%
Pro-rata payout
Split across data & labels

How GetDATA works

Post a request

Specify modality (CT/MRI/X-ray/US), label schema, regions, and budget.

Providers submit

Hospitals and labs upload de-identified datasets.

Labeling

Crowdsourced annotations with quality scoring.

Verify

Compliance & quality checks for both data and labels.

Pay by contribution

Budget splits automatically across providers & annotators.

Medical data types

Radiology imaging

X-ray, CT, MRI, Ultrasound — DICOM with reports & labels

Pathology WSI

Whole-slide images with region masks and cell/tissue labels

Clinical text

De-identified notes, summaries, and NLP entity labels

Ophthalmology

Fundus/OCT with grading and segmentation

Cardiology

ECG/echo images with outcomes and expert labels

Oncology & others

Multimodal cohorts with aligned pathology/radiology

Pricing

Transaction-based. You set the project budget; we charge a percentage only when delivered. Budgets split across data acquisition and labeling.

Starter

8%

Pilots & MVPs in healthcare AI

  • Up to $10k budget
  • Basic verification
  • Email support

Growth

6%

Scaled training runs

  • Up to $250k budget
  • Advanced checks & scoring
  • Priority support

Enterprise

Custom

Regulated, complex studies

  • Unlimited budget
  • DPA/BAA & bespoke workflows
  • Dedicated success

FAQ

How does crowdsourcing work?

After a request is posted, approved contributors can upload de-identified data or take labeling tasks in our tools. All batches are quality-scored.

Who can label data?

Verified radiologists, clinicians, and trained annotators. Low-quality labels are flagged and rejected.

Is the data authentic and compliant?

We verify provenance, require de-identification, and support HIPAA/GDPR, DPAs/BAAs, and enterprise audit hooks.

How do payouts work?

Payouts split proportionally by approved volume and quality score across both data providers and annotators.

What label formats do you support?

Classification, bounding boxes, segmentation masks, measurement tables, and custom JSON schemas.

What timelines can I expect?

Typical fulfillment completes in weeks, depending on modality, cohort criteria, labeling depth and quantity.