Core versus Periphery OCT Radiomics for Predicting Radiation Damage in Tumor Spheroids

  • chair:Medical Imaging for Modeling and Simulation
  • type:Bachelor thesis
  • tutor:

    M.Sc. Mark Daniel Arndt

  • person in charge:

    Armin Krull

  • This thesis builds a compact OCT analysis pipeline to predict radiation-induced damage in 3D tumour spheroids using machine learning. Spheroids are segmented with core-only and periphery-only masks; from each, interpretable features (intensity, speckle/texture, morphology) are extracted. The most predictive features will be identified for each region and differences will be analysed.