Instructions

• To train prediction model, following the steps:
  • \({\rm\color{blue}{S1\_generate\_spectrum.py}}\)
    • Generate \({\rm\mu_s}\) spectra from the A, K fitted factor from literatures.
    • Generate \({\rm\mu_a}\) spectra in the range of literatures.
  • \({\rm\color{blue}{absoprtion\_spectrum\_by\_substance/make\_spectrum.ipynb}}\)
    • Generate \({\rm\mu_a}\) spectra based on the chromophores such as water, blood, subcutaneous, collagen and melanin.
    • copy the result files skin_mua_spectrum.csv, fat_mua_spectrum.csv, muscle_mua_spectrum and cca_mua_spectrum.csv to the directory prediction_model/OPs_used/mua_chromophore
  • \({\rm\color{blue}{S2\_generate\_surrogate\_result.py}}\)
  • \({\rm\color{blue}{S3\_generate\_prediction\_input.py}}\)
  • \({\rm\color{blue}{S4\_train\_prediction\_model.py}}\)