What this challenge is all about
Since the outbreak of the global Covid-19 pandemic, the number of confirmed COVID-19 cases has reached over 16 million globally, affecting virtually every territory, and with a fatality rate ~2-3% among the cohort of PCR-positive cases. Given the high demand for effective diagnosis and treatment of cases, the WHO recently released a rapid advice guide in July 2020, in which chest imaging is conditionally recommended for several purposes, e.g. to aid diagnosis in the absence/delay of PCR testing, to assess the need for ICU admission and to inform the therapeutic management of patients.
In this challenge, we aim to aid radiologists and physicians through objective and quantitative computational assessment of chest imaging in the context of COVID-19. We provide access to a large dataset of 3D chest CT imaging of the lung, collected from several European and international radiological centers. We call the international research community to develop and test artificial intelligence algorithms on this dataset.
We provide access to low-dose chest CT imaging volumes from a mixed cohort of COVID-19 and non-COVID-19 cases. The dataset contains ~150 labeled/segmented cases (~100 COVID-19, ~50 non-COVID-19), and ~200 unlabeled volumes. A particular scientific challenge will lie in the effective use of unlabeled data through semi- and self-supervised training techniques. Labels represent five lung lobes and two lesions types, consolidation and ground-glass opacities. Labels are provided in a multi-hot encoding to allow region overlaps (e.g. lesions within lung lobes). For local development, we provide a realistic toy dataset of 96 synthetic volumes with 4D labelmaps.
To maintain privacy, the anonymized imaging data remains non-disclosed within a biobank. Participating teams can design their algorithms locally using the representative synthetic dataset. Once ready, teams can submit training & validation jobs on the real dataset through Eisen, a deep learning framework based on pyTorch. Models are trained in the cloud by sponsorship of AWS. We actively promote open science, and require all participating teams to provide their solutions open-source to the technical and medical research community.
Hey! We are thrilled to have you here and we hope that you will consider participating in the Covid-19 Challenge. Together, we will have a great experience making new friends, learning useful stuff and creating AI solutions to fight the disease!
Various attempts have already been made, using AI to help identify, quantify and predict this disease. Most of it was done by single entities. We believe however that the crowd - YOU! - can do much better than that and we want to show that this works.
To make it very convenient for you, we have spent the last months searching Covid-19 datasets throughout the world and meticulously curating them. In this process, many medical students and radiologists helped to create what we call a "gold standard" training set. So we hope you can focus on algorithm development and you don't need to deal with bad data.
After the registration, there will be a “micro-challenge” for all the teams in order to qualify for the main task. This micro-challenge with the task of segmentation based on our synthetic toy dataset has to be submitted until 14th of August. We only have resources for a limited number of teams. Our jury will select the rangers and hunter teams based on the micro-challenge.
You are eager to participate in this great challenge? Then please sign up, either as a Hunter (the teams that actually develop) or as a Ranger (if you have knowledge to share and you like to support the teams):