Digital Medical Imaging Laboratory
- Presentation
- Services
- Resources
- Outstanding projects
The I3PT Medical Imaging Laboratory develops research, innovation and technological support activities in the field of digital medical imaging and artificial intelligence applied to imaging diagnostics. Its mission is to promote collaborative projects and provide technical and methodological services to I3PT research groups, as well as to other institutions and entities linked to the biomedical and technological fields.
The laboratory has access to the infrastructure and technological resources of medical imaging integrated into the Parc Taulí University Hospital, including PACS tools and technological capacity for the processing and analysis of large volumes of data. This infrastructure allows both the management (extraction, anonymization, structuring and storage) and the processing and exploitation of clinical data for research projects, development and validation of new technologies, in a secure environment for working with medical data.
The laboratory develops methodologies for processing and quantitative analysis of medical images, as well as artificial intelligence algorithms applied to different areas of diagnostic imaging. It also participates in pilot studies and clinical validation of new technologies based on medical imaging and AI.
- Extraction, anonymization, structuring and transfer of medical image databases from PACS, aimed at research projects, innovation and validation of new technologies.
- Storage and management of anonymized and annotated research datasets.
- Support in the planning, design and execution of projects related to medical imaging and the application of new technologies in the field of imaging diagnostics.
- Development of image processing methodologies, quantitative analysis tools and artificial intelligence algorithms applied to medical imaging.
- Validation and deployment of proof of concept, pilot studies and clinical trials linked to new medical imaging technologies and artificial intelligence.
- Collaboration in the preparation of funding applications, technical reports and scientific documentation associated with research and innovation projects.
Medical imaging infrastructure and systems:
- Medical image collections and associated data:
-
- Medical images in DICOM format
-
- Radiology reports in HTML format
- RAIM Server (PACS)
- RAIM Cloud (centralized PACS)
- RAIM SDI (Radiological Information System)
- RAIM SISDI Web (RIS web)
- RAIM Viewer (Medical Image Viewer)
- Orthanc (research-oriented medical image manager and viewer)
- SQL Server Management Studio (PACS data management)
Computational capacity and development:
- High-performance workstation with GPU for processing, medical image analysis and development of artificial intelligence algorithms.
- Server with 6 TB of storage for medical imaging research data.
- Development and analysis environments:
- Visual Studio Code
- Jupyter Notebook
- Matlab
- Python and image processing and AI libraries
- Image processing and segmentation tools:
- ITK-SNAP
- 3D slicer
Optimal lung
Optimal Lung is a diagnostic support tool based on deep learning algorithms for the detection and characterization of pulmonary nodules in medical imaging. The system includes two applications, OPTIMAL CT and OPTIMAL XR, for the analysis of CT and chest radiographs, with the aim of supporting the early identification of potentially relevant lesions. It is a co-development project with Eurecat and Hospital Vall d'Hebron.
UNCAN-Connect
UNCAN-Connect is a European project aimed at creating a decentralized collaboration network for oncology research. The initiative aims to facilitate secure data exchange and joint work between different actors in the healthcare and research ecosystem at European level. The project focuses on boosting research in various types of cancer, such as pediatric tumors, lymphoid neoplasms and carcinomas of the pancreas, ovary, lung and prostate.
RadioVal
RadioVal is an international project aimed at validating artificial intelligence solutions based on radiomics for predicting treatment response in breast cancer. It involves multiple hospital centers in Europe and other regions, with the aim of evaluating the robustness, equity and clinical applicability of the algorithms in real-world settings. Our participation is based on the contribution as a data provider, with the transfer of breast cancer datasets in mammography, tomosynthesis and magnetic resonance.