MELD classifier Documentation

Installation

  • Docker container
  • Native installation
  • Singularity container

MELD FCD prediction pipeline

  • Prepare the data
  • Compute the harmonisation parameters for a new scanner
  • Predict FCD-lesion on MRI data
  • Interpretation of the MELD Graph pipeline results

Other usage

  • Notebooks to recreate figures
  • Training and evaluating a graph model
  • Contributing guide

Help

  • FAQs
MELD classifier Documentation
  • Notebooks to recreate figures
  • View page source

Notebooks to recreate figures

Figures:
- Figure 1. Overview
- Figure 2A. Comparison of predictions & cluster numbers
- Figure 2B. Comparison cluster numbers
- Figure 3. Examples of interpretable patient reports

Tables:
- Table 1. results MELD graph model vs MLP model
- Table 2. breakdown performances

eTables:
- eTable 1. Demographics
- eTable 2. Reduction in false positive clusters
- eTable 3. MELD Graph model with harmonized vs non-harmonized MRI features

eFigures:
- eFigure 1. MELD Graph model architecture
- eFigure 2. Pipeline for running a new patient’s MRI scan through MELD Graph
- eFigure 3. Stability of NeuroCombat
- eFigure 4. Characterisation of predictions\

Previous Next

© Copyright 2024, Mathilde Ripart, Hannah Spitzer, Sophie Adler, Konrad Wagstyl.

Built with Sphinx using a theme provided by Read the Docs.