NeuroMiner 1.3 (09/2024)

What is new

Check out the new features!

• New preprocessing options

  • Disparate impact removal

  • Probabilisitic feature extraction techniques

  • JuSpace - Neurotransmitter systems analysis

  • ROI analysis for neuroimaging Data

• Improved integration of algorithms from Python’s sklearn library:

  • Multi-layer Perceptron (MLP)

  • Gradient Boosting

  • Random Forests

  • Independent Compoennt Analysis (ICA) - dimensionality reduction

• Improved synthetic data generation

• Improved simulation analyses

• Shapley Values for model interpretation

• Improved multiclass classification & bug fixes

• Improved subgroup analysis for external validation

• Decision curve analysis for classification models: Assess clinical decision thresholds with the new decision curve analysis tool, making models more actionable for patient care.

• Model calibration analysis: Ensure model probabilities align with clinical realities, improving the reliability of predictive models.

• Resolved bugs and improved usability!

Release notes

• Visualisation of MVRVM for classification currently not implemented