ML Classifiers to distinguish healthy from cancer patient-derived EVs' Raman spectra

The Google Colab/Python codes for AdaBoost Random Forest, SVM, and Decision trees with hyperparameter tuning (Scikit-learn) to classify patient liquid biopsy-derived cancer EVs (extracellular vesicles) are provided. The compressed file contains the codes (4 for Raman spectra and 1 for FTIR spectra) and three sample text files of Raman spectra are provided (2 from cancer patients and 1 from healthy patient sera-EVs).

The paper associated to this Github code is available in https://arxiv.org/abs/2107.10332 and has been published in Applied Intelligence (2022) as found in the citation below:

Uthamacumaran, A., Elouatik, S., Abdouh, M. et al. Machine learning characterization of cancer patients-derived extracellular vesicles using vibrational spectroscopies: results from a pilot study.. Appl Intell (2022). https://doi.org/10.1007/s10489-022-03203-1

Keywords: Cancer; Artificial Intelligence; Spectroscopy; Exosomes; Precision Oncology; Early Cancer Detection.