Luminescence-enabled three-dimensional temperature bioimaging

This dataset supports the study titled "Luminescence-enabled three-dimensional temperature bioimaging", in which we developed a machine learning-assisted 3D thermometry approach based on the near-infrared (NIR) emission of Ag₂S nanothermometers (NTs). The dataset includes: Calibration (training) data: Hyperspectral emission spectra of Ag₂S NTs collected under varying tissue types, tissue thicknesses, and temperature conditions (360 combinations total). Each condition includes 500 extracted spectra for robust model training. Trained neural network model: The CNN+DNN model used to extract temperature and depth information from hyperspectral images. Validation data: 20% calibration data, and independent datasets used to validate the performance of the model, including 3D thermometry experiments in tissue phantoms, ex vivo tissues, and in vivo mouse models. Temperature simulation data: Results from a thermal transfer simulation conducted independently to support the validation experiments. The data are organized into folders corresponding to training, validation, and simulation. Spectral data are stored in .csv format, while model files are in .h5. Code can be accessed using Python. These data support key findings in the paper. The dataset enables verification of 3D temperature mapping performance in various conditions. File: all_data_phantom_and_real_tissue.dat Description: all calibration data (emission spectra) for training model File: validation-d.csv Description: violin plotting of d predictions in figure 2 e File: New_3D_all-loss.txt Description: training loss File: 3capillaries_silmulation_0.13mm.csv Description: the temperature transfer simulation of validation experiments: cube phantom tissue (Fig.3e) File: validation-T.csv Description: violin plotting of T predictions in figure 2 e File: labels_phantom_and_real_tissue.dat Description: the labels of the calibration emission spectra File: 3cappilaries-binary.csv Description: prediction of phantom tissue validation (Fig.3d) File: 3cappilaries-binary.gif Description: gif of Fig.3d File: 3cappilaries_spectra_with_empty_data.csv Description: spectra pass to the algoritm(neural networks) for predictions File: 2capillaries_silmulation_0.09mm.csv Description: the temperature transfer simulation of validation experiments: ex vivo tissue (Fig.2J) File: 2_capillaries-emission_with_empty_data.csv Description: spectra pass to the algoritm(neural networks) for predictions File: 2cappilaries-binary.csv Description: predict results (Fig.3i) File: 2cappilaries-binary.gif Description: gif file of Fig.3i File: cone_prediction_center.csv Description: the center capillary in cone. predictions of d and T (Supplementary Figure 19e) File: ReconstructionCone_Coords_X_Y_Z.csv Description: Reconstruction of Cone Supplementary Figure 20 File: cone_animation.gif Description: gif of Supplementary Figure 19e File: enbemded_capillaries_animation_10min.gif Description: gif of Supplementary Figure 21d File: cone_simulation-9mm.csv Description: heat transfer simulation of cone, Supplementary Figure 19f File: enbemded_capillaries_fitting_3D_simulation_T.csv Description: simulations of enbemded capillaries (Supplementary Figure 21e) File: enbemded_capillaries_T_d_pred_down.csv Description: predictions of enbemded_capillaries, Supplementary Figure 22 A File: enbemded_capillaries_T_d_pred_up.csv Description: predictions of enbemded_capillaries, Supplementary Figure 22 B File: raton2_vessels_3D_plot.csv Description: prediction of in vivo experiments Fig 4c File: raton2_vessels_spectra_correct.csv Description: Corrected spectra from blood vessels for predictions File: raton2_vessels_3D_plot_projection.gif Description: gif of fig