The OPTIMAM Mammography Image Database

The OPTIMAM Mammography Image Database (OMI-DB) has been created to support research involving medical imaging. This database contains unprocessed and processed medical images, associated annotations and data and where applicable expert-determined ground truths describing features of interest. The process of collection, annotation and storage is almost fully automated and is extremely adaptable, allowing for quick and easy expansion to disparate imaging sites.

Learn more


Unprocessed Images

Both unprocessed and processed images are stored in the repository.


Regions of interest are annotated by Expert radiologists and additional clinical annotations are paired with each image through local clinical databases.

Multi Vendor & Modality

The database contains images from multiple vendors/modalities providing a comprehensive research dataset.

Fully Representative Collection

Ongoing Image collection is automated allowing for fast secure collection. All screen detected breast cancers are collected, including random sets of normal cases.

Database Stats

Centre A Centre B Centre C Total
# Clients 5248 2468 681 8397
# Studies 11094 7257 4363 22714
# Images 90688 57114 19156 166958

Getting Access

Getting Access

Other researchers can apply to make use of the Images, data and tools we have collected and created.

Application Review

Applications are reviewed by the steering committee.

Ready to Apply? Apply for Access


Until recently, the main focus for this database has been digital mammography, However the OMI-DB is fully capable of supporting heterogeneous modalities and is extendable and expandable to store any associated data. Currently, the majority of associated data is made up of radiological, clinical and pathological annotations extracted from the National Breast Screening System (NBSS). In addition to this data, software and systems have been created to allow expert radiologists to annotate the images with ROI indicated interesting features and additional descriptors of these features. The data from OMI-DB has been used in observer studies and there are several more upcoming studies in progress or planned.

This database comprises processed and unprocessed digital mammograms. Typically these are for women who have had a screen detected breast cancer and contain two views of each breast (medio-lateral oblique and cranio-caudal). The locations of the radiographic features relating to the cancer are marked. Additional clinical information including radiological appearance, pathology, screening history is stored in the database. The cases collected are representative of screen detected cancers since all cases were collected from specific screening centres over set time periods. In addition to these cancer cases a representative sample of normal and benign cases are also available. All of the information in the database is anonymised.

The primary purpose for collecting the data is to conduct the research described in OPTIMAM. This research programme evaluates how various factors affect cancer detection using mammographic images.

Sharing of images and data

Applications to use the images can be made by a web form or email and considered by an internal sub­committee made up of members of the OPTIMAM2 steering committee. All access will be subject to a sharing agreement created in collaboration with CRT. Applications will be subject to scientific critique. For more information see Getting Access

Technical Details

The image database has been published in various forms. Further technial details regarding the collection and usage can be found in the following publications:

  • - Patel, M. N., Looney, P. T., Young, K. C., and Halling-Brown, M. D., "Automated collection of medical images for research from heterogeneous systems: trials and tribulations," Proc. SPIE 9039 Medical Imaging (2014)

Case Studies

The image database and MedXViewer has been utilised in many studies. Details can be found in the following publications:

  • - Alistair Mackenzie, Lucy M. Warren, Matthew G. Wallis, Rosalind M. Given-Wilson, Julie Cooke, David R. Dance, Dev P. Chakraborty, Mark D. Halling-Brown, Padraig T. Looney, Kenneth C. Young, The relationship between cancer detection in mammography and image quality measurements, Physica Medica,
  • - P. T. Looney, K. C. Young and M. D. Halling-Brown. Medxviewer: providing a web-enabled workstation environment for collaborative and remote medical imaging viewing, perception studies and reader training. Radiation Protection Dosimetry (2015), pp. 1–6
  • - L M Warren, L Dummott, MG Wallis, RM Given-Wilson, J Cooke, D R Dance , M Halling-Brown, M Patel and K C Young. Comparison of simulated calcification clusters used in virtual clinical trials and screen-detected calcification clusters. PMB 2015
  • - A Mackenzie, L M Warren, M G Wallis, M D Halling-Brown et al Breast Cancer detection rates using four different types of mammography detector. Eur Radiol. 2016 Mar;26(3):874-83.
  • - Lucy M. Warren, Rosalind M. Given-Wilson, Matthew G. Wallis, Julie Cooke, Mark D. Halling-Brown, Alistair Mackenzie, Dev P. Chakraborty,Hilde Bosmans, David R. Dance, and Kenneth C. Young, The Effect of Image Processing on the Detection of Cancers in Digital Mammography, American Journal of Roentgenology, August, Vol. 203, No. 2 : pp. 387-393 (2014)


MedXViewer (Medical Extensible Viewer) is an application designed to allow workstation-independent, PACS-less viewing and interaction with anonymised medical images (e.g for observer studies). More details

  • - Looney P.T, Young K, Mackenzie A, Halling-Brown M.D, “MedXViewer: an extensible web-enabled software package for medical imaging”. Proc. SPIE 9037 Medical Imaging (2014)



We post notable current or upcoming events (Disabled).


Full list of publications about or involving the OPTIMAM image Database and associated tools.


This work is part of the OPTIMAM2 project funded by Cancer Research UK.

OPTIMAM Data Access Application Form


This is a list of publications about or involving the Optimam image Database and associated tools.

Journal Publications

Conference Proceedings