This S2V readme.txt file was generated on 2021-01-10 by Johannes Müller. GENERAL INFORMATION 1. Title of Dataset: "Slice2Volume: Fusion of multimodal medical imaging and light microscopy data in 3D of irradiation injured brain tissue." 2. Author Information A. Principal Investigator Contact Information Name: Dr. Antje Dietrich Institution: OncoRay – National Center for Radiation Research in Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Helmholtz-Zentrum Dresden – Rossendorf, Germany German Cancer Consortium (DKTK), Partner Site Dresden, and German Cancer Research Center (DKFZ), Heidelberg, Germany Email: antje.deitrich@ukdd.de B. Associate or Co-investigator Contact Information Name: Johannes Müller Institution: OncoRay – National Center for Radiation Research in Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Helmholtz-Zentrum Dresden – Rossendorf, Germany Helmholtz-Zentrum Dresden - Rossendorf, Institute of Radiooncology – OncoRay, Dresden, Germany Email: johannes.mueller@ukdd.de C. Alternate Contact Information Name: Theresa Suckert Institution: OncoRay – National Center for Radiation Research in Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Helmholtz-Zentrum Dresden – Rossendorf, Germany German Cancer Consortium (DKTK), Partner Site Dresden, and German Cancer Research Center (DKFZ), Heidelberg, Germany Email: theresa.suckert@ukdd.de 3. Date of data collection (range): 2018-2020 4. Information about funding sources that supported the collection of the data: European comission research grant #730983 (INSPIRE) SHARING/ACCESS INFORMATION 1. Licenses/restrictions placed on the data: Creative Commons Attribution 4.0 International 2. Links to publications that cite or use the data: DOI: 10.1016/j.radonc.2020.02.023 DATA & FILE OVERVIEW 1. Introduction: The S2V dataset contains co-aligned image data from several mice, the brains of which were irradiatied with protons. Specifically, the dataset contains: cone-beam computed tomography (CBCT) scans magnetic resonance imaging (MRI) scans over several weeks Monte-Carlo simulation of Dose & linear energy transfer (LET) DSURQE mouse brain atlas Immunofluorescent and histochemically stained tissue sections of the irradiated area. 2. Data types: The S2V dataset provides image data and transformation parameter files. The latter allows free transformations between different used coordinate systems. The following image data types are provided, all of which can be accessed with publically available software: *.czi: Raw image data from Zeiss microscopes. Can be opened with Zen blue light (https://www.zeiss.com/microscopy/int/products/microscope-software/zen-lite.html) *.tif: tagged image format, standard format used for processed image data, Can be opened with Fiji image processing software (https://imagej.net/Fiji/Downloads) *.dcm: "Digital Imaging and Communications in Medicine". Standard format for medical imaging data (e.g. MRI, CT). Can be opened with Fiji or MITK (www.mitk.org) *.nii: "Neuroimaging Informatics Technology Initiative" format. Used for brain atlas data, can be opened with Fiji or MITK. Other filetypes: *.csv, *.txt, open with Excel, Text editor or NotePad++ (https://notepad-plus-plus.org/) 3. Data structure: This paragraph describes the file structure of the S2V dataset. The dataset follows a basic file structure of /mouse_X_modality.zip/data, for easy access. /root /Mouse_identifier (e.g. "P2A_B6_M2") /Atlas.zip /DSURQE_40micron_labels.nii: Raw mouse brain atlas. /DSURQE_40micron_labels.tif: Raw mouse brain atlas in .tif format /DSURQE_40micron_labels_transformed.tif: Transformed brain atlas that was matched with CBCT image data with axial plane orientation. /DSURQE_40micron_labels_transformed_resliced.tif: Transformed brain atlas with coronal plane orientation. /BigWarp_Landmarks.csv: Landmarks used for affine co-registration of Atlas and CBCT. Can be open with the ImageJ plugin "BigWarp" (https://imagej.github.io/BigWarp) /CBCT.zip /CT_0000.dcm .... /CT_XXXX.dcm Axial image slices of mouse brain CT scan. Images were acquired with the SAIGRT CBCT platform (Tillner et al., DOI: 10.1088/0031-9155/61/8/3084) /Histology.zip /Slices These directories contain the raw data from bundles of adjacent tissue sections. The physical distance between these bundles is 150µm. /0001_Scene_1 /0001_Scene_2 /0002_Scene_1 ... /XXXX_Scene_Y Raw images: /*_GNI_*.czi: Raw image of tissue section stained with GFAP (Glial fibrillary acidic protein), Iba1 (Allograft inflammatory factor 1), Nestin and DAPI. Referred to as "GNI". /*_NOK_*.czi: Raw image of tissue section stained with NeuN (neuronal cells), OSP (Oligodendrocyte specific protein), Ki67 (Marker of proliferation 67) and DAPI. Referred to as "NOK". /*_HE_*.czi/tif: Raw image of tissue section stained with H&E (hematoxylin & eosin) Co-aligned images: /result /...*.tif: histological stainings of fixed image separated by channel (e.g. DAPI/Iba1/OSP/etc). The images were not transformed. /transformed_*.tif: Transformed histological stainings separated by channel (e.g. transformed_DAPI/etc). The images were transformed with transformix (https://elastix.lumc.nl/index.php) according to the provided transformation parameters (see below). /*_HE_*_to_*_GNI_* Contains transformation parameters that were used to co-align the raw H&E image with the fluorescent GNI image. For this, both images are downsampled and converted to binary masks: MovingMask.tif: Masked & downsampled H&E image TargetMask.tif: Masked & downsampled GNI image Both images are co-aligned with the Elastix registration software (https://elastix.lumc.nl/index.php) result.tif: Resulting transformed version of MovingMask.tif TransformParameters_small.0.txt: Transformix-readable transform parameter file that transforms MovingMask.tif into the coordinate system of TargetMask.tif TransformParameters_altered.0.txt: Altered Transformix-readable parameter file that transforms the H&E image in high resolution into the GNI image's coordinate system. /*_NOK_*_to_*_GNI_* see (*_HE_*_to_*_GNI_*) /Warped This directory contains the output of the co-alignment of histological slices with masked and transformed atlas data (see line 77) as returned from the Slice2Volume script (https://github.com/jo-mueller/Slice2Volume) /results /OutputStack_DAPI_r.tif: Slice2Volume output: DAPI staining color channel aligned with transformed atlas image data. /OutputStack_interpolated_DAPI_r.tif: Slice2Volume output: DAPI staining color channel aligned with atlas image data with interpolated empty voxels. /S2V_LogFile.txt: Contains all parameters which were used as input for Slice2Volume to reproduce the results. /SliceAssignment_Overview: Overview file that lists the pairs of histological sections (only DAPI channel used) and coronal slices of the transformed and masked brain atlas that were co-aligned. /trafo This directory contains the transformix-readable transform parameters to project high resolution histological images into the assigned coronal slice of the CT coordinate system. /0001_Scene_1_trafo.txt /0001_Scene_1_trafo_inverse.txt ... /XXXX_Scene_Y_trafo.txt: transformix-readable transform parameters to project histological sections from the directory .../Histology/Slices/result/ into the CT coordinate system. See .../Warped/results/SliceAssignment_Overview.txt for the correct coronal slice in the CT image. /XXXX_Scene_Y_trafo_inverse.txt: transformix-readable transform parameters to project the assigned coronal slice in the CT coordinate system into the coordinate system of the high resolution microscopy data. The transformed image has to be upsampled with a method of your choice to match the dimensions of the high resolution microscopy data. /MRI.zip The directories contain raw and processed MRI data from different timepoints after irradiation, with Week_00 marking the pre-irradiation timepoint. /Week_00 ... /Week_XX /* /Sequence_1 ... /Sequence_I: Contains raw MRI data (*.dcm) as separate slices. The directory is named according to the respective MRI sequence (e.g. T1_GRE_SP_3D__iso_11min) Sequence_1.dcm ... Sequence_I.dcm: Contains raw MRI data (*dcm) as single file. The file is named according to the respective MRI sequence (e.g. T1_GRE_SP_3D__iso_11min.dcm) /*_denoised /T1_GRE_SP_3D__iso_11min_BM3D_15xsigma.tif: Denoised T1-weighted MRI image. Denoising was done with the BM3D method with a sigma of 1.5x /T2_FSE_3D_Centric_Partial_iso_small_6min_BM3D_15xsigma.tif: Denoised T2-weighted MRI image. Denoising was done with the BM3D method with a sigma of 1.5x /*_registered /T1_GRE_SP_3D__iso_11min_BM3D_15xsigma.txt: Denoised T1-weighted image which was co-aligned with the respective image from Week_00. Co-alignment was done with Elastix. /T1_GRE_SP_3D__iso_11min_BM3D_15xsigma_trafo.txt: Transformix-readable transform parameters that align the denoised T1-weighted image (../*_denoised) with its counterpart from Week_00. /T2_FSE_3D_Centric_Partial_iso_small_6min_BM3D_15xsigma.txt: Denoised T2-weighted image which was co-aligned with the respective image from Week_00. Co-alignment was done with Elastix. /T2_FSE_3D_Centric_Partial_iso_small_6min_BM3D_15xsigma_trafo.txt: Transformix-readable transform parameters that align the denoised T2-weighted image (../*_denoised) with its counterpart from Week_00. /*_warped2CT /T1_GRE_SP_3D__iso_11min_BM3D_15xsigma_xfm_0.tif: Denoised & registered T1-weighted image which is now further aligned with the CT image. The co-alignment was done with the ImageJ plugin BigWarp based on the provided landmarks. /T1_GRE_SP_3D__iso_11min_BM3D_15xsigma_landmarks.csv: BigWarp-readable landmarks to affine-transform the denoised and aligned T1-weighted image into the coordinate system of the CT image. /T2_FSE_3D_Centric_Partial_iso_small_6min_BM3D_15xsigma_xfm_0.tif: Denoised & registered T2-weighted image which is now further aligned with the CT image. The co-alignment was done with the ImageJ plugin BigWarp based on the provided landmarks. T2_FSE_3D_Centric_Partial_iso_small_6min_BM3D_15xsigma_landmarks.csv: BigWarp-readable landmarks to affine-transform the denoised and aligned T2-weighted image into the coordinate system of the CT image. /Simulation.zip /Dose.tif: Dose in relative units (scaled from 0-255). Determined with TOPAS Monte-Carlo simulation. /LET.tif: LET in calibrated units (keV/um). Determined with TOPAS Monte-Carlo simulation. Simulations were performed as decribed by Suckert et al. (DOI: 10.1016/j.radonc.2020.02.023 ) /Dose Folder with raw scored dose output of Monte-Carlo simulation in *.dcm format. Each dcm-file contains data from one independent simulation job. The dose.tif image array is composed of the sum of ten randomly picked simulation runs. Pixel values have to multiplied with the DoseGridScaling value (see dicom metadata) /LET Folder with raw scored LET of Monte-Carlo simulation in *.dcm format. Each dcm-file contains data from one independent simulation job. The LET.tif image array is composed of the average of ten randomly picked simulation runs. Pixel values have to multiplied with the DoseGridScaling value (see dicom metadata) /QA.zip /tex Contains all .tex documents that were used to create the provided quality-assurance protocol QA_protocol.pdf. Document can be re-created with MikTex (https://miktex.org/) and TeXMaker as editor (https://www.xm1math.net/texmaker/) /images All figures displayed in the QA-protocol can be accessed separately in this directory as png files. /P2A_ControlStainings.zip /Brain_BLG_X_2nd_only_Y.czi: negative control image stained only with secondary antibody Y (e.g., donkey_aRat_AF488) /Brain_BLG_10_no_AB.czi: negative control image stained with no antibody to assess autofluorescence. CHANGELOG 0.1.1 Added first data of single sample to repository 0.2 Changed zip-packing structure to separate directories for different modalities (CBCT, MRI, Atlas, Histology, Simulation) 0.2.1 Added data from further animals and included QA assurance protocols 0.2.2 Pinned QA protocol to top 0.3 - Incorporated results and updated data from QA-round of co-authors. Changes were made to co-registered MRI-data and histology co-alignment. - Added negative control stainings to repository 0.3.1 - Minor file corrections