contact us

What we do

Provide Medical
Datasets

medDARE’s image and video datasets are securely stored and transferred according to GDPR and HIPAA regulations. We possess unique data from a network of 60+ clinics and research organizations.

Work with Patient’s
Metadata

We anonymize all datasets we provide, while we also define pathologies, provide medical insights, and work with medical documentation. Our clients get unique data for their research in Healthcare AI.

Training Datasets for Machine Learning

We work with 2D and 3D image datasets and videos, annotating with bounding boxes, keypoints or semantic segmentation. We work with open-sourced tools and in our clients’ software, training their AI algorithms.

medDARE has access to a wide range of fully
anonymized medical images and datasets
for research and machine learning

Datasets

CT SCANS

A computerized tomography (CT) scan uses X-rays and process those on a computer using reconstruction algorithms to produce cross-sectional images of a body.

There are different types, all of which medDARE can provide:

  • Spiral CT
  • Electron beam tomography
  • CT perfusion imaging

Since its invention, the CT has revolutionized diagnostic medicine and is used for preventive medicine and screening throughout the globe. This makes it a perfect candidate for AI research who try for example to diagnose early occurrence of lung cancer or for developments which try to find brain lesions and help radiologists triage faster.

MRIs

MRI scanners use strong magnetic fields, magnetic field gradients, and radio waves to generate images of the organs in the body. MRI read the location of hydrogen atoms which is highly present in the body and therefore essentially map the location of water and fat in the body.

There are different types, all of which medDARE can provide:

  • Spin Echo
  • Gradient Echo
  • Inversion recovery
  • Diffusion weighted
  • Perfusion weighted
  • fMRI
  • MR Angiography

MRI are used a lot in diagnostic medicine but the time and constraints on patient make them less than ideal. They are perfect candidates for AI research to help the detection of abnormalities on prostate scans or to enable physicians to make faster, more accurate diagnostic and treatment decisions.

X-RAYS

X-Rays are the oldest form of medical imaging and are still used extensively to this day.

The sheer amount of X-rays taken around the globe and the efficiency at which they can be taken make them ideal candidate for AI technology like identifying Vertebral Compression Fractures, one of the signs of osteoporosis or to help radiographers triage critical conditions, such as pneumothorax.

Dataset training

BOUNDING BOXES

Bounding boxes are the most common type of datasets training. They are used to segment the area with some particular pathology.

Main advantages of using bounding boxes are high speed of annotation and thus, a lower cost.

medDARE’s team has been working with bounding boxes on multiple projects before, including lung CTs annotation, brain MRIs annotation, X-Ray spine annotation of different vertebrae.

2D SEGMENTATION

2D segmentation is used when there’s a need to contour the exact shape in each slice of a DICOM file. These segments can be extrapolated to different slices.

2D segmentation is more time-consuming than the bounding boxes segmentation but gives more precise results. This type of data training is often used for small pathologies, for example, in brain MRI or for the annotation of vascular system.

LANDMARKS

Landmarks data training – also known as keypoints annotation – works by placing points on an image or slices to label objects on these data.

This type of data training is usually used for facial annotation, facial recognition or annotation of gestures.

Landmarks annotation usually is least time-consuming type of data training.

3D SEGMENTATION

3D segmentation is the most complex type of data training. It requires annotation of objects on all three dimentions of the DICOM file.

This type of annotation is very time consuming, and based on the number of organs or pathologies which needs to be annotated, it can take up to 8-10 hours for annotation of one dataset.

Usually, 3D segmentation is done in 3D Slicer.

Contact Us

    I consent for medDARE B.V. to process my data and agree to the
    terms of the Privacy Policy