CNI - Challenge




Training data now available here.



Validation data now available here.



Large, open source datasets, such as the Human Connectome Project (HCP) and the Autism Brain Imaging Data Exchange (ABIDE), have spurred the development of new and increasingly powerful machine learning strategies in brain connectomics. However, one key question remains: are we capturing biologically relevant and generalizable information about the brain, or are we simply overfitting to the data? The CNI Challenge 2019 will address the issues of generalizability and clinical relevance for functional connectomes. We will leverage a unique resting-state fMRI (rsfMRI) dataset of attention deficit hyperactivity disorder (ADHD) and neurotypical controls (NC). Participants will be asked to design a classification framework that can predict subject diagnosis (ADHD vs. Neurotypical Control) based on brain connectivity data. We will provide 120 examples of each class for training and validation.

In a surprise twist, we will also evaluate the classification performance on a related clinical population with an ADHD comorbidity. This challenge will allow us to assess (1) whether the method is extracting functional connectivity patterns related to ADHD symptomatology, and (2) how much of this information “transfers” between clinical populations.

This is the first MICCAI challenge on functional connectomics. Together with the 3rd CNI workshop featuring the latest connectomic advancements, our challenge presents a necessary step toward reproducible and translational research in the field.

The expected outcomes of this challenge are as follows:

  • Evaluate the reliability and performance of different analytical strategies for healthy vs disordered connectome classification.
  • Evaluate the generalizability or "transferability" of these classification methods between clinical populations.

Important Dates

  • Validation data release: July 23rd, 2019
  • Submission deadline: August 15th Sept 1st, 2019, 23:59 EST


An accepted challenge submission must be accompanied by at least one author registered to the CNI-TL Challenge through the MICCAI satellite events registration. However, participants will have the option to submit a short video presentation about their method in lieu of in-person attendance.