MONAI Label - Training from Scratch

In this video, you’ll learn how to train your first model from scratch using MONAI Label and 3D Slicer. First, you’ll download the COVID-19 CT Dataset and Radiology. You’ll prepare the dataset and start the MONAI Label Server using the radiology and segmentation model. Then you’ll use 3D Slicer and the “Grow from Seeds” functionality to segment your first two volumes and start training your model. Last, you’ll examine the training logs and get recommendations on the number of epochs and labels needed to create a good model. COVID-19 CT Lung and Infection Segmentation Dataset: https://zenodo.org/record/3757476 If you’re interested in getting started with MONAI Label, you can find the links to our GitHub Organization below. Once you’ve installed MONAI Label, stay involved in the community by submitting bugs, starting discussions on GitHub, or joining the MONAI Slack server. Resources: https://monai.io/ https://github.com/Project-MONAI/monailabel https://www.slicer.org/ https://zenodo.org/record/3757476 Chapters: 00:00 Intro 00:09 Download COVID-19 CT Dataset 00:29 Download Radiology App 00:55 Set Label Names and No Pretrained Model 01:15 Prepare Dataset 01:43 Start MONAI Label Server 02:03 Open 3D Slicer 02:13 Use Grow From Seeds Functionality 03:17 Submit First Label and Start Training 03:47 Annotate Second Volume 04:32 Submit Second Label and Train 04:46 Training Logs and Recommendations 05:12 Outro