INCISIVE is a 42-months research project, funded under the call DT-TDS-05-2020 – AI for Health Images. Throughout the project's lifecycle, a vast amount of cancer-related data has been collected, both in the form of imaging examinations as well as accompanying clinical and diagnostic information. The repository is maintained after the project's lifecycle to enable the secure donation and sharing of data in compliance with ethical, legal and privacy demands, increasing accessibility to datasets and enabling experimentation of AI-based solutions, towards the large-scale adoption of such solutions in cancer diagnosis, prediction and follow-up.
Currently, the INCISIVE repository contains anonymized data for four major cancer types; lung cancer, breast cancer, prostate cancer and colorectal cancer. Imaging examinations are stored in DICOM format, whereas clinical and diagnostic information is semantically encoded in SNOMED and LOINC codes, thus enabling data interoperability and adherence to standardized medical protocols. Pixel-level annotations accompany the imaging examinations, thus offering the option to develop precise AI-based tools, accelerating research in the medical domain.
The INCISIVE repository also operates as part of the EUCAIM project, which provides a robust, trustworthy platform for researchers, clinicians, and innovators to access diverse cancer images, enabling the benchmarking, testing, and piloting of AI-driven technologies. By connecting high-quality cancer image data and AI experts, Cancer Image Europe facilitates collaboration and accelerates the development of cutting-edge solutions for cancer diagnosis and treatment.
INCISIVE offers 2 levels of data access, restricted and unrestricted, depending on the respective data providers’ request. Restricted means that the data of certain data providers can be utilized for federated processing, without the option of data visualization, whereas unrestricted also allows for data visualization and pre-processing. Depending on the level of access that each dataset posesses, a request for use may be submitted either in centralized form or as part of the EUCAIM federated node network.
For inquiries related to obtaining data access, please contact Gianna Tsakou ([email protected]).
For more detailed information about data availability, please contact Stavros Sykiotis ([email protected])