Open-source tools

Spectral preprocessing

Baseline correction, cosmic ray removal, smoothing, intensity normalisation and resampling onto a common wavenumber grid.

• Language: PythonLicence: MIT

  raman-skincancer-augmentation

  Open-source implementation of physically constrained data augmentation and interpretable machine learning for class-imbalanced Raman diagnosis of skin cancer. Full preprocessing chain, augmentation pipeline, PCA-30 + logistic regression classifier, and leakage-safe cross-validation. Regenerates every figure and metric of the paper.

  Source code