Clog Loss: Advance Alzheimer’s Research with Stall Catchers

Help accelerate Alzheimer's research by automatically classifying which blood vessels are flowing and which are stalled. #health

$10,000 in prizes
aug 2020
915 joined

Stall Catchers example image


5.8 million Americans live with Alzheimer’s dementia, including 10% of all seniors 65 and older. Scientists at Cornell have discovered links between “stalls,” or clogged blood vessels in the brain, and Alzheimer’s. Stalls can reduce overall blood flow in the brain by 30%. The ability to prevent or remove stalls may transform how Alzheimer’s disease is treated.

Stall Catchers is a citizen science project that crowdsources the analysis of Alzheimer’s disease research data provided by Cornell University’s Department of Biomedical Engineering. It resolves a pressing analytic bottleneck: for each hour of data collection it would take an entire week to analyze the results in the lab, which means an entire experimental dataset would take 6-12 months to analyze. Today, the Stall Catchers players are collectively analyzing data 5x faster than the lab while exceeding data quality requirements.

The research team has realized there are aspects of this task that are best suited to uniquely human cognitive faculties as explained in this blog post. However, some portion of the data, the “low-hanging fruit,” may be within reach of machine learning models that are able to distinguish between easy and difficult data and are applied only in cases where they have been validated to meet the researchers’ data quality requirements. If a machine learning classifier could be used for 50% of the data, it would double the analytic throughput of Stall Catchers and could achieve the original goal of analyzing the data 10x faster than the lab. This could ultimately put finding an Alzheimer’s treatment target in reach within the next year or two.

Through the Stall Catchers project, there is now a one-of-a-kind dataset that can be used to train and evaluate ML models on this task. Each exemplar is an image stack (a 3D image) taken from a live mouse brain showing blood vessels and blood flow. Each stack has an outline drawn around a target vessel segment and has been converted to an mp4 video file. The objective of this competition is to classify the outlined vessel segment as flowing—if blood is moving through the vessel—or stalled if the vessel has no blood flow.

Competition End Date:

Aug. 3, 2020, midnight UTC

Place Prize Amount
1st $5,000
2nd $2,000
3rd $1,000
Bonus $2,000

MathWorks employees are not eligible for prizes.

Bonus Award: Top MATLAB User

We're offering a bonus prize of $2,000 to the top contributor using MATLAB. Just make sure to fill out the Software Environment form to be considered for the prize. This is required to be eligible for the bonus prize, though we'd love to hear from you regardless of what environment you're using!

See the MathWorks competition page for more information about complimentary software licenses and learning resources for this challenge.

Note: Recipient can also be a winner of the general prize pool.

Prize generously supplied by our friends at MathWorks.

Image courtesy of the Human Computation Institute.