This repository contains the code code used to define and train the model presented in the paper: Cu fractionation, isotopic analysis, and data processing via machine learning: new approaches for the diagnosis and follow up of Wilson’s Disease via ICP-MS It also includes the code used to generate the uncertainty metric.
JAAS manuscript is a preliminar version of the manuscript published on the Journal of Analytical Atomic Spectrometry. (The final version can be found in: https://doi.org/10.1039/D2JA00267A)
Wilson_ML.py: includes the code used to train the models and generate the results.
Anlysis_Uncertainty.py: includes the functions to analize the uncertainty present in the outputs of the 50 NNs ensemble
Models includes all the models used in the final versión of the work, and the generated results.