<> "The repository administrator has not yet configured an RDF license."^^ . <> . . "Decentralized Infrastructure for Medical Image Analysis -\r\nThe Development and Establishment of Kaapana as an Open Framework for Imaging Platforms in Clinical Environments"^^ . "The emergence of new data- and algorithm-driven analysis methods is revolutionizing\r\nmany areas of research and enabling solutions to problems that were previously\r\nconsidered intractable. But does this also apply to medicine?\r\n\r\nImproving diagnosis, fine characterization of patients for personalized medicine,\r\nmonitoring disease progression and its prognosis, or predicting the outcomes of\r\nvarious therapies are just some of the areas that could potentially benefit from such\r\nnew analysis techniques based on Deep Learning, which has enabled major advances\r\nin computer vision and related fields. Such algorithms now enable machines to\r\nbetter understand and interpret visual image data, which on the one hand offers\r\npromising perspectives for medical image processing, but on the other hand also poses\r\nchallenges. As such, large amounts of annotated data are needed to prevent overfitting\r\nand to generate robust, generalizing and reliable models. Multicenter imaging studies\r\ncould greatly improve the availability of such data and also enhance heterogeneity\r\nby obtaining training data from various sites. However, sharing data across multiple\r\nsites has proven to be difficult due to the high level of data protection associated\r\nwith medical records and technical challenges such as interoperability. Consequently,\r\nthis thesis attempts to avoid the necessity of data exchange by following a different\r\napproach: \"Let’s share the algorithms, not the data!\"\r\n\r\nThe objective and central research question here is whether it is possible to shift\r\nthe evaluation and training of modern image analysis methods to the clinical data\r\nowners and how this can be accomplished. Although this approach helps to circumvent\r\nthe data export and the associated data protection challenges, the participating\r\npartners must still be enabled to execute the algorithms from a technical and or-\r\nganizational perspective. For this purpose, this dissertation investigated concepts,\r\nthe development and establishment of a decentralized infrastructure for clinical\r\nmedical image analysis that enables standardized data access and uniform execution\r\nof analysis methods for joint data analysis in the context of multicenter imaging studies.\r\nThe technical realization of this infrastructure was achieved by developing a software\r\nframework called Kaapana, which enables the building of imaging platforms. The\r\nresulting software can be hosted on dedicated servers within the clinical IT environ-\r\nment to be operated isolated from any external connectivity and to be interconnected\r\nwith other local clinical systems. By leveraging modern cloud technologies such as\r\ncontainers and Kubernetes, the local deployment provides a private cloud for image\r\nprocessing that can be accessed from any locally connected workstations via the web\r\nbrowser. Standardized linkage to the clinical PACS via DICOM and the integration\r\nof a research imaging archive enables redundant data management and consistent\r\ndata access for the execution of analysis methods. Dashboards enable efficient data\r\nexploration and filtering by visually presenting DICOM metadata of the platform’s data\r\nand allowing it to be selected via search queries. A uniform execution environment\r\nfor data processing allows algorithms to be applied to such selected data and to be\r\nuniformly packaged and shared with partners. High-performance server hardware\r\nincluding Graphics Processing Units enables a variety of analysis techniques such as\r\nDeep Learning-based model inference, as well as the training of new models to be\r\nshared with partners. Within the infrastructure, the standardized and widely adopted\r\nDICOM format has been prioritized so that also many analysis results can be provided\r\nin a standard-compliant way. Using formats such as DICOM SEG, SR or Encapsu-\r\nlated PDF, data annotations become compatible with clinical workflows and IT systems.\r\nWith support of the Radiological Cooperative Network and the German Cancer Consor-\r\ntium, the resulting infrastructure could be deployed and evaluated within two German\r\nresearch consortia. To this end, all 36 German university hospitals have commissioned\r\ntheir own server on which the platform has been installed and tested. This involved\r\nevaluating the commissioning, integration, operation and maintenance of such a\r\ndecentralized network, as well as various use cases designed to cover the typical tasks\r\nof such a system. As a result, these use cases and the corresponding varying demands\r\ncould be realized with the developed concept and the implemented framework. A\r\nflexible extension mechanism also allows the integration of additional services such as137\r\nthe MITK workbench or processing algorithms such as the nnUNet into the framework.\r\nFurthermore, first analysis pipelines developed by external partners could also be\r\nintegrated and delivered to the partners already.\r\n\r\nWithin the scope of this work, clinics were enabled to apply up-to-date research\r\nmethods to their own data through the development, distribution and support of\r\nthe developed infrastructure. Since the research consortia, which already include\r\nall German university hospitals, have only just started their activities, there are\r\ngreat opportunities to make the high-quality data from the partner sites accessible\r\nand usable for research in the future. Because of Kaapana’s open source code, its\r\narchitecture based on common industry standards, and its already broad deployment,\r\nthis framework could also serve as a foundation for areas other than medical imaging,\r\nand thus offer the potential for tighter data integration for clinical computing in\r\ngeneral."^^ . "2023" . . . . . . . "Jonas"^^ . "Scherer"^^ . "Jonas Scherer"^^ . . . . . . "Decentralized Infrastructure for Medical Image Analysis -\r\nThe Development and Establishment of Kaapana as an Open Framework for Imaging Platforms in Clinical Environments (PDF)"^^ . . . "Jonas_Scherer_Dissertation_pdfA_1a.pdf"^^ . . . "Decentralized Infrastructure for Medical Image Analysis -\r\nThe Development and Establishment of Kaapana as an Open Framework for Imaging Platforms in Clinical Environments (Other)"^^ . . . . . . "indexcodes.txt"^^ . . . "Decentralized Infrastructure for Medical Image Analysis -\r\nThe Development and Establishment of Kaapana as an Open Framework for Imaging Platforms in Clinical Environments (Other)"^^ . . . . . . "lightbox.jpg"^^ . . . "Decentralized Infrastructure for Medical Image Analysis -\r\nThe Development and Establishment of Kaapana as an Open Framework for Imaging Platforms in Clinical Environments (Other)"^^ . . . . . . "preview.jpg"^^ . . . "Decentralized Infrastructure for Medical Image Analysis -\r\nThe Development and Establishment of Kaapana as an Open Framework for Imaging Platforms in Clinical Environments (Other)"^^ . . . . . . "medium.jpg"^^ . . . "Decentralized Infrastructure for Medical Image Analysis -\r\nThe Development and Establishment of Kaapana as an Open Framework for Imaging Platforms in Clinical Environments (Other)"^^ . . . . . . "small.jpg"^^ . . "HTML Summary of #32015 \n\nDecentralized Infrastructure for Medical Image Analysis - \nThe Development and Establishment of Kaapana as an Open Framework for Imaging Platforms in Clinical Environments\n\n" . "text/html" . . . "004 Informatik"@de . "004 Data processing Computer science"@en . . . "600 Technik, Medizin, angewandte Wissenschaften"@de . "600 Technology (Applied sciences)"@en . . . "610 Medizin"@de . "610 Medical sciences Medicine"@en . .