Non Invasive Temperature Estimation for RF safety
Input data for these steps will be available soon on Google Drive in a CST_Data Folder. If any of the links are broken in the future, please open an issue.
- [MATLAB] Data shaping after CST EM/Thermal simulation
- File: nite/MatFolder/TestSlicesSequence.m
- Input: Google Drive links: CST_Data folder
- Output: Mat files for each slices (training and test). Move output files to nite/MatFolder
- [MATLAB] Data preparation for deep learning training
- File: nite/MatFolder/Data_mass_main.m
- Input: x_Data mat file
- Output: xtrain-4.mat, ytrain-4.mat, xtest-4.mat, ytest-4.mat
- Move output files to nite/PyFolder
- [Python] Building and running the neural network for training
- File : nite/PyFolder/Train_Test.ipynb
- Input: xtrain-4.mat, ytrain-4.mat, xtest-4.mat, ytest-4.mat
- Output: trained model (trained_model.h5)
- [MATLAB] Data preparation for deep learning testing
- File: nite/MatFolder/Data_mass_main_prep4test.m
- Input: t_Data mat files [changes with the considered slice]
- Output: NITE_x_test_sl_HUGO4.mat with the variable x_test, NITE_x_test_sl_NELLY4.mat with the variable x_test [changes with the considered slice]
- Move output files to nite/PyFolder
- [Python] Testing slices with trained model
- File: nite/PyFolder/Training_Testing.ipynb
- Input: NITE_x_test_sl_HUGO4.mat, NITE_x_test_sl_NELLY4.mat [changes with the considered slice]
- Output: Reconstructed temperature maps (NITE_recon_HUGO_4-new.mat, NITE_recon_NELLY_4-new.mat)
- Move output files to nite/MatFolder/
- [MATLAB] Temperature maps plot and comparison between NITE map and CST map
- Run nite/MatFolder/Data_mass_main.m again.
- File: nite/MatFolder/Data_mass_main_testresvis.m
- Input: NITE_x_test_sl_HUGO4.mat, NITE_x_test_sl_NELLY4.mat [changes with the considered slice]
- Output: Plot and comparison of temperature maps