SonadorAI: Classify Chest X-Rays from the NIH Database
Series of articles describing how to create a machine learning model within Sonador that is capable of classifying conditions on Chest X-Rays.
- Part 1: Introduction and Overview
- Part 2: Import Chest X-Ray Dataset
- Convert JPEG/PNG to DICOM (sonador-client#14)
- Part 3: Import Machine Learning Labels
- Create DICOM-SR documents containing labels for the machine learning model
- Part 4: Train and Assess the model
- Document interface for Sonador AI (#4)
- How should data be structured prior to attempting to train a model?
- What are the inputs to a dataset, how do those work? (#1 (closed))
- What are the methods that need to be overridden in a SonadorAI dataset to provide data to PyTorch?
- Demonstrate example model and results
- How should a PyTorch model with Sonador data be assessed?
- Document interface for Sonador AI (#4)