News

A neuroscientist studying MRI images of the brain of a patient

A completed EMPIR project has continued to bring quantitative measurements to a widely used medical imaging technique

The project
Magnetic Resonance Imaging (MRI) is a medical technique that uses a strong magnetic field and radio waves to diagnose a range of conditions in patients without the need for ionising radiation. Around 30- 40 million MRI scans are used in Europe each year. Despite its extensive use, MRI is mainly a qualitative technique and relies on the experience of clinicians to interpret the MRI maps of tissues that are generated.

Recently, new MR scanning methods have become available such as Electric Properties Tomography (EPT) and Magnetic Resonance Fingerprinting (MRF) which offer the ability to produce quantitative images that could be analysed by software or AI – making their results much more objective. However, these new approaches have lacked the metrological validation required for clinical use.

The EMPIR project Quantitative MR-based imaging of physical biomarkers (18HLT05, QUIERO) addressed this problem developing new techniques and methodology.

Tissue phantoms

Most MR test objects lacked the heterogeneity of real tissues. The project developed a protocol for 3D printing phantoms out of semi-solid materials that exhibited dielectric and relaxation properties, similar to those of the white and grey matter, including a brain-shaped phantom in combination with a calcium chloride (CaCl2) solution mimicking cerebrospinal fluid.

Electric Properties Tomography advances

Open-source software

An open-source, state-of-the-art library of EPT methods was established: EPTlib. The library has continued to be developed and improved since the end of the project in 2022. It is aimed at those who are interested in EPT research but don’t want to start from zero and clinical groups who want to apply EPT but don’t want to develop the tools.

Measurement uncertainty

Working with the EMPIR project Advancing measurement uncertainty - comprehensive examples for key international standards (17NRM05, EMUE), the QUIERO project, through a model implemented in EPTlib, applied the first measurement uncertainty for the commonly used Helmholtz-EPT technique.

MRF advances

In the heart, two MR-specific parameters, T1 and T2 times, are of interest because T1 can indicate the presence of fibrosis (i.e., scar tissue) and T2 can indicate inflammation. Information from both can tell if condition is due to injury, such as infarction, is acute (e.g., inflammation present but no scar yet) or not. Commonly two different scans are needed, one showing T1 and one showing T2.

QUIERO developed techniques which obtained T1 and T2 directly from raw MR data. This improves the quality and accuracy of the obtained results in one MR scan. The work on the use of MRF in cardiac imaging has since been published.

Continuing impact

Consortium members have been involved in an A4IM project, looking at MRF-based techniques and extending the methods developed in QUIERO by combining them with the power of deep learning. This project is developing portable open-source MR scanners to provide this service to a wide-range of patients, especially in communities which so far do not have access to MRI.

INRiM, the National Metrology Institute (NMI) of Italy, who coordinated the QUIERO project, are now working with the Italian Associazione Italiana di Fisica Medica (AIFM). INRiM will develop around 100 phantoms to send to hospitals around Italy to make clinical measurements and increase the pool of MR data available.

Participation in the first “EPT Challenge”

From an initiative from the University Medical Center of Utrecht (UMC), members of the consortium have participated in the first ever EPT Challenge, organised by the Electro-Magnetic Tissue Properties Study Group of International Society for Magnetic Resonance in Medicine (ISMRM). Lasting for 1-2 months the aim of the challenge was to benchmark EPT reconstruction methods and involved 3 phases:

1) Reconstructions from a simulated (blind) dataset. This phase mainly involved data sharing to allow results to be returned in a consistent form.

2) Reconstructions from several simulated datasets. This phase used simulated data of a healthy brain or tumour brain with a tumour inclusion. Participants applied algorithms to generate EPT reconstructions.

3) Reconstructions from measured data. The third phase related to real MR data on homogenous and heterogenous phantoms and volunteers.

Results on phases 1 & 2 are anticipated to be released by the end of 2024 and phase 3, on in vivo data, will be released as a separate analysis in 2025.

Data from the project has also contributed to the European Metrology Partnership project Standardisation for safe implant scanning in MRI (21NRM05,STASIS) which is investigating and evaluating interactions between patient implants and the MRI magnetic fields.

Luca Zilberti (INRiM) who coordinated the successful project said about the work

“The development of quantitative MRI runs fast and, with QUIERO, European NMIs have become part of this revolution. As for any other measurement, uncertainty evaluation plays a fundamental role in giving the measured data a practical interpretation. We will go on working to bring this aspect to full maturity”.
 

The EMPIR project was co-funded by the European Union's Horizon 2020 research and innovation programme and the EMPIR Participating States

This Metrology Partnership project has received funding from the European Partnership on Metrology, co-financed by the European Union Horizon Europe Research and Innovation Programme and from the Participating State

Want to hear more about EURAMET?

Sign up for EURAMET newsletters and other information

Follow us on LinkedIn and X/Twitter