Since late 2019, the world been in health crisis because of the COVID-19 pandemic. In fact, using Medical Imagery has proved to be efficient in detecting Covid-19 Infection. These Medical Imaging include: X-ray, CT-scans and Ultrasounds. The use of CT-scans is not only limited to the detection of COVID-19 cases, but they can also be used for other important tasks such quantifying the infection and monitoring the evolution of the disease, which can help in treatment and save the patient’s life. In this challenge, the participants will use a dataset labelled by two expert radiologists, who estimated the Covid-19 infection, to train and validate their approaches. In the testing phase, participants will test their approaches using a test dataset collected from various CT-scanners and recording settings.
The challenge has three sets: Train, Val, and Test. The Train set is obtained from 132 CT-scans, from which 128 CT-scans has confirmed to have Covid-19 based on positive reverse transcription polymerase chain reaction (RT-PCR) and CT scan manifestations identified by two experienced thoracic radiologists. The rest four CT-scans have not any infection type (Healthy). The Val set is obtained from 57 CT-scans, from which 55 CT-scans has confirmed to have Covid-19 based on positive reverse transcription polymerase chain reaction (RT-PCR) and CT scan manifestations identified by two experienced thoracic radiologists. The rest two CT-scans have not any infection type (Healthy).
The Train split has two files: Images (Slices) Folder and Labeling Folder ('.csv' file) that contains the labels for each Slice (Image). The Train Split can be downloaded from the link.
The validation dataset contains 1034 images with no label. The Validation Split can be downloaded from the link.