This page accepts zip folders and tests the contained DICOM files for compatibility with MMPS processing. If no errors are found the assessments should all report with a green checkmark. Open the corresponding section to see a detailed list of erroneous entries. Please be aware that MMPS is under continued development, additional tags might be required in the future. Please report such issues to the authors so they can be added to this application.
Multi-Modal Processing Stream (MMPS, (doi)[https://doi.org/10.1016%2Fj.neuroimage.2019.116091]) is a software package developed and maintained in-house at the Center for Multimodal Imaging and Genetics (CMIG) at the University of California, San Diego (UCSD).
General instructions for pseudonymization/anonymization of DICOM data: Use your PACS or a tool like (MIRC Anonymizer)[http://mirc.rsna.org/download/Anonymizer-installer.jar] for data anonymization. Replace both PatientID and PatientName with the study identifier for the participant. Do not remove private tags.
This page only accepts zip folders and parses them for DICOM files. It is safe to use this application on sensitive data as no part of your data will leave your machine. The assessments are performed in your web-browser and at your site.
Number of files: 0 (0 non-DICOM files ignored), Number of series: 0
Assessments
Test if StudyDate is present. The "StudyDate" DICOM tag (0008,0020) should be present and in the correct format, e.g. 20060306 (sixths of march 2006).
Test if Study-, Series- and SOPInstanceUID tags are in each DICOM file. The three identifiers do not contain any patient relevant information. They are needed to identify images that belong together (images in a volume, series in a study).
Private tags CSAImageHeaderInfo and CSASeriesHeaderInfo. If SIEMENS is the manufacturer we expect the CSA header entries (private tags) to have a value. MMPS will use the values to decide if gradient distortion corrections need to be performed.
Manufacturer information. MMPS will use the Manufacturer and ManufacturerModelName tags to decide how to read other tags.
PatientID and PatientName. Both tags should be anonymized but present.
SeriesDescription. Processing with MMPS will fail without a SeriesDescription tag.
