AstraZeneca Challenge: Automated Detection and Quantitation of Bone Marrow Cells

$25,000 USD

Challenge overview

Many oncology therapeutics have dose limiting toxicities due to adverse effects in the bone marrow. Safety evaluation in preclinical species typically relies on histopathology evaluation, which can miss subtle shifts in hematopoietic cell populations due to high cell density and tissue complexity. AstraZeneca is seeking an automated tool for the detection and quantitation of hematopoietic cells in bone marrow tissue sections.

This is a Reduction-to-Practice Challenge that requires written documentation, output from the proposed algorithm, and submission of the source code and/or an executable for validation, if requested by the Seeker.

Hematopoietic stem cells in the bone marrow give rise to other cells such as red blood cells. The hematopoietic compartment of the bone marrow is composed of a heterogeneous population of cells in various stages of development, which presents a challenge for pathologists to identify and quantify the specific cell types (lineages) and developmental stages preferentially affected by different types of oncology therapeutics. AstraZeneca is seeking an automated tool to accurately detect and quantify the 3 major lineages of hematopoietic cells from scanned images of bone marrow. Please refer to the attached slides in the Detailed Description & Requirements section for more details and annotated images.

The submission to the Challenge should include the following:

  1. A detailed description of the proposed Solution and how it addresses each Technical Requirement presented in the Detailed Description of the Challenge. This should also include a thorough description of the algorithm used in the Solution accompanied by a well-articulated rationale for the method employed.
  2. Output from the proposed algorithm will need to be applied to the images in the test folder that will not be used for building the computer model. Output must be in the form described in the Detailed Description of the Challenge. Submissions will be ranked by the Seeker based on the technical merit of the work and when needed, by using evaluations against an internal data set.
  3. For the top ranked submissions, the Seeker may request the source code and/or an executable with sufficient documentation to enable the Seeker to compile, execute the algorithm, and validate the method using additional test data sets. Furthermore, the Solution must rely on open source or similar licensing models that would allow the Seeker to use the tool without incurring additional licensing costs.

The Challenge award is contingent upon evaluation and validation of the submitted Solutions by the Seeker. During the evaluation period, the Seeker will validate submissions using additional images similar to the example images provided in the Challenge.

To receive an award, the Solvers will not have to transfer their exclusive IP rights to the Seeker. Instead, Solvers will grant to the Seeker a non-exclusive license to practice their solutions.

Submissions to this Challenge must be received by 11:59 PM (US Eastern Time) on January 3, 2020. 

Late submissions will not be considered.


AstraZeneca is a global, science-led, biopharmaceutical company that focuses on the discovery, development and commercialization of prescription medicines, primarily for the treatment of disease in three main therapy areas – Oncology, Cardiovascular & Metabolic Diseases and Respiratory. AstraZeneca also is selectively active in the areas of autoimmunity, neuroscience and infection. As an innovation-driven, research organization, AstraZeneca recognizes that great ideas come from many sources. Open innovation is an avenue by which ideas can be shared and AstraZeneca recently launched a pavilion to further its commitment to facilitate the advancement of pharmaceutical research.

Use the slider to explore how the Challenge process works:


Review & Accept



To start solving this Challenge, log in to the Challenge Center or register as a Solver

Visit the Wazoku Crowd Community to start the conversation