ADDI Alzheimer’s Detection Challenge

This challenge is closed. Click here for updates on our challenge and sign-up to receive future challenge alerts.

SUMMARY

The purpose of the ADDI Alzheimer’s Detection Challenge is to build an automated algorithm that classifies a patient’s phase of Alzheimer’s disease (AD) based on a common cognitive screening tool for dementia—the Clock Drawing Test (CDT).

The result will be shared with users of the AD Workbench to expand the existing suite of analytics for the benefit of the user community and the advancement of AD research.

Prizes will be awarded for best scores. There will be four (4) cash prizes:

1st place - $20,000 USD
2nd place - $15,000 USD
3rd place - $10,000 USD
4th place - $5,000 USD

The 14 non-cash prizes (3 x Sony PlayStation 5, 1 x X-Box Series X, 5 x Oculus Quest 2, 5 x DJI Mavic Mini 2) will be awarded based on score or community contributions.

The aggregate value of all prizes is approximately $60,000 USD.

Click here for detailed instructions on how to participate via the AIcrowd platform.

Review Rules and Guidelines for more information about eligibility, how to enter, submission guidelines, prizes, and selection of winners.

OBJECTIVE

The purpose of this data science challenge is to build an automated algorithm that classifies a patient’s phase of Alzheimer’s based on a common cognitive screening tool for dementia—the Clock Drawing Test (CDT).

The research will leverage the collected data (derived features from the decomposition of clock-drawing images) to build an algorithm that will help in predicting if the patient is in Pre-Alzheimer’s phase (Early Warning), Post-Alzheimer’s phase (Detection) or Normal (Not an Alzheimer’s patient). This is a three-class classification task.

The result will be shared with users of the AD Workbench to expand the existing suite of analytics for the benefit of the user community and the advancement of AD research.

BACKGROUND

The Alzheimer’s Disease Data Initiative (ADDI) is a new global effort that aims to advance innovation for Alzheimer’s disease (AD) and related dementias by connecting researchers with the data needed to generate insights and inform the development of improved treatments and diagnostic tools. ADDI, a nonprofit medical research organization, was created by a coalition of partners to increase the sharing of dementia-related data among researchers and providing new ways to experiment with the most trusted datasets.

The Alzheimer’s Disease (AD) Workbench is a cloud-based platform with tools for scientists to accelerate discoveries and innovations for AD and related dementias. The workbench will help facilitate interoperability across data platforms and enable users to work across multiple datasets. The AD Workbench will allow permissioned researchers to import their datasets, access and transfer data from other platforms and enable the safe and secure analysis of datasets. Users can have a personalized workspace where they can ensure quality control, harmonize data and analyze data within the platform.

The Alzheimer’s Detection Challenge aims to build and mobilize the AD Workbench community, giving data scientists the opportunity to apply open-ended innovation, predictive modeling, and machine learning to some of the biggest problems in the field.

RULES AND GUIDELINES