FANTAISE: Fostering the safe use of ionising radiation in AI- based medical CT imaging by developing quality assessment and assurance

Project description

AI approaches such as tomographic image reconstruction and pre- or postprocessing techniques have the potential to offer better medical imaging such as CT, interventional imaging, PET and SPECT using potentially less ionising radiation exposure to patients. AI-processed images always appear to be of high fidelity, making it difficult to identify artefacts and missing information. Thus, there is a risk of misdiagnosis. An incorrect diagnosis based on CT, PET or SPECT imaging procedures is a radiation protection issue, as ionising radiation is applied to the patient without medical benefit. Due to the high-fidelity appearance of AI-enhanced images, it is difficult to compare the quality between different CT systems. In addition, quality assurance becomes more challenging as issues with the systems' hardware and software become harder to detect, as standard test phantoms are potentially identified by the algorithms resulting in the obfuscation of problems using AI-enhancement.

This project aims to facilitate the evaluation of CT imaging procedures that are AI-processed and to ascertain any regulatory requirements for these procedures. Adjustable software and hardware phantoms representing relevant pathological and anatomical structures will be developed for CT imaging in conjunction with European experts for radiology. These phantoms will serve as test objects for which the ground truth is known but which the system cannot easily identify. This will enable the safe use of improved imaging technologies by ensuring CT quality assessment and assurance.

Further information PIANTOFORTE Homepage

 

Funding & Duration

The project is funded from 10/2025 to 04/2029 by the European Union