Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

MONAI Medical Image Deep Learning: A 3-Minute Introduction


Published on

These slides are meant as a "teaser", to get people interested in learning more about MONAI. They provide a brief summary of the motivation for and the potential of MONAI.

These were presented at the start of the 2020 3D Slicer Project Week. The recording of that presentation is available online:

Published in: Data & Analytics
  • Be the first to comment

  • Be the first to like this

MONAI Medical Image Deep Learning: A 3-Minute Introduction

  1. 1. The Open Source Platform for Reproducible Deep Learning in Medical Imaging Stephen R. Aylward, Ph.D. Chair of MONAI External Advisory Board Senior Directory of Strategic Initiatives, Kitware
  2. 2. Medical Open Network for A. I. (MONAI) Goal: Accelerate the pace of research and development by providing a common software foundation and a vibrant community for medical imaging deep learning. ■ Began as a collaboration between Nvidia and King’s College London ■ Prerna Dogra (Nvidia) and Jorge Cardoso (KCL) ■ Optimized for biomedical applications ■ Medical formats, medical images, transforms, loss functions, metrics ■ Strong emphasis on reproducibility
  3. 3. MONAI IS A GROWING COMMUNITY (Since April 2020) 41
  4. 4. Encapsulating a COVID-19 Algorithm into an Integrated AI Application Nvidia CLARA
  5. 5. MONAI:End-End Training Workflow in 10 Lines of Code from monai.application import MedNISTDataset from import DataLoader from monai.transforms import LoadPNGd, AddChanneld, ScaleIntensityd, ToTensord, Compose from monai.networks.nets import densenet121 from monai.inferers import SimpleInferer from monai.engines import SupervisedTrainer transform = Compose( [ LoadPNGd(keys="image"), AddChanneld(keys="image"), ScaleIntensityd(keys="image"), ToTensord(keys=["image", "label"]) ] ) dataset = MedNISTDataset(root_dir="./", transform=transform, section="training", download=True) trainer = SupervisedTrainer( max_epochs=5, train_data_loader=DataLoader(dataset, batch_size=2, shuffle=True, num_workers=4), network=densenet121(spatial_dims=2, in_channels=1, out_channels=6), optimizer=torch.optim.Adam(model.parameters(),lr=1e-5), loss_function=torch.nn.CrossEntropyLoss(), inferer=SimpleInferer() )