Segmentation models and experimental results on segmenting slide scans of diatom preparations from river Menne

<p>This archive contains the models and the results of the deep learning experiments published in Kloster et al. 2022: Improving deep learning-based segmentation of diatoms in gigapixel-sized virtual slides by object-based tile positioning and object integrity constraint.</p> <p>The folders contain models and results of each of the 24 training runs, the files &ldquo;metrics.experiments.pt[prediction score threshold].csv contain segmentation score thresholds the models obtained on unknown evaluation data.</p> <p>The data pertaining to each model is stored in a separate folder. Its name follows the convention &ldquo;experiment.[model architecture].[tiling method].[dataset size].[timestamp]&rdquo;. Please note that the naming of the tiling method differs from the manuscript; &ldquo;fixed&rdquo; refers to fixed-stride tiling, &ldquo;objectcentred&rdquo; object-based positioning, and &ldquo;objectcentred_with_cropped&rdquo; to object-based positioning + object integrity constraint. Dataset size &ldquo;10p&rdquo; refers to a 10% subsample of the complete training data set, &ldquo;25p&rdquo; to a 25% subsample and so on.</p> <p>Each folder contains the best performing model of the corresponding training run as pth (Mask R-CNN, PyTorch) or h5 (U-Net, Tensorflow/Keras) file, along with a files describing setup and conduction of the training. Subfolders &ldquo;test_images_segmented*&rdquo; contain the segmentation predicted by the models on the evaluation data. These are supplied as 1.) mask image either with the intensity value representing the prediction score (score 0.0 &ndash; 1.0 = intensities 0 &ndash; 255) or thresholded by the prediction score threshold given in the folder name; 2.) input image overlayed with ground truth (red) and segmentation mask (green), resulting in TP marked in yellow; 3.) a CSV file giving info on the filenames and the segmentation performance metrics.</p> <p>Please refer to the manuscript for further details.</p>