The MR-STAT technology circumvents the current status quo in MRI, whereby quantitative parameters are produced after raw data have been transposed to the image domain. Instead, MR-STAT reconstructs quantitative parameters directly from the raw time domain. This approach, fueled by recent advances in computing and algorithm development, effectively maximizes the use of every data point collected and removes the need for priori knowledge such as signal dictionaries, thereby opening a wealth of opportunities in standardizing quantitative MRI, at ultra/very short scan times. MR-STAT was developed with a VIDI grant (850,000 euros) and the Demonstrator grant (150,000 euros), both from the Dutch Research Council (NWO). The technology is optimized with a 1-million-euro grant from the NWO and Philips.

With the introduction of this innovative technology based on advanced mathematical algorithms, a single scan is sufficient. This one scan yields not only images but also quantitative measurements. Radiologists obtain more – and more objective – information about the tissue through MR-STAT and patients spend a significantly shorter time in the MRI scanner. The technology can be applied to existing MRI scanners, enabling other hospitals to start using it as well. In the future, MR-STAT MRI examinations may become faster, more personal and more accurate for all patients.

This animation explains the technique: 

MR-STAT (in English)

Researchers

Associate professor Dr. Alessandro Sbrizzi conceived the MR-STAT algorithm in collaboration with Prof. Dr. Nico van den Berg (chair of the computational imaging group at UMCU). 

On the technology development side, Sbrizzi and van den Berg supervise a team of researchers comprising Oscar van der Heide, Hongyan Liu, Dr. Stefano Mandija, Dr. Federico D’Agata, Mike Eijbersen and Guus Grimbergen.

The clinical component of the project is supervised by Prof. Dr. Jeroen Hendrikse with principal investigator Dr. Anja van der Kolk (both neuroradiologists at the UMCU) and research assistant Miss Beyza Koktas.

The researchers explain the development in an interview, you can read their story here.

News

August 2020, Philips and UMC Utrecht have a research partnership to advance the field of precision diagnosis through breakthrough quantitative MRI technology MR-STAT. This exclusive, multi-year research partnership to create a clinical research network and bring MR-STAT to market.

MR-STAT in the media

• 27 februari 2020, ICT&health: Een betere MRI-scan in vijf minuten met MR-STAT (Dutch)

• 27 februari 2020, Nu.nl: UMC Utrecht komt met primeur: MRI-scan in 5 in plaats van 25 minuten (Dutch)

• 27 februari 2020, RTV Utrecht: Nieuwe MRI-techniek UMC Utrecht: 5 minuten in de scanner in plaats van 25 (Dutch)

• 27 februari 2020, AD: Geen 25 maar slechts 5 minuten in de MRI-scan dankzij Utrechtse onderzoekers (Dutch)

• Oncologie up-to-date: Duur van MRI-scan kan veel korter (Dutch)

Clinical trial

The MR-STAT project is currently running a preliminary clinical trial in which we will investigate the image quality of MR-STAT when applied to the brain and brain pathology. A total of forty patients with 4 different neurological diseases (multiple sclerosis, ischemic or hemorrhagic stroke, epilepsy and brain tumors) as well as 10 healthy volunteers will participate in this study and undergo both an MR-STAT examination and standard MRI-scans of the brain, to compare MR-STAT images with current standard clinical care. This clinical trial has been registered in the Dutch Trialregister; more information can be found using the following link: https://www.trialregister.nl/trial/8437 or by accessing the website www.trialregister.nl and searching for trial no. NL8437. For questions about this study please contact us at mr-stat@umcutrecht.nl. 

Klinische studie

Het team van het MR-STAT-project voert op dit moment een klinische studie uit waarin de kwaliteit van MR-STAT wordt beoordeeld voor het in beeld brengen van zowel gezonde hersenen als hersenaandoeningen. In totaal zullen veertig patienten met verschillende hersenaandoeningen (multiple sclerosis (MS), herseninfarct of hersenbloeding, epilepsie en hersentumoren) en 10 gezonde vrijwilligers meedoen aan dit onderzoek. Deelnemers zullen zowel een MR-STAT-scan als een normale MRI-scan van de hersenen ondergaan; op deze manier kunnen de MR-STAT-beelden worden vergeleken met standaard klinische zorg. Deze klinische studie staat geregisteerd bij het Nederlandse Trialregister; meer informatie vindt u via de volgende link: https://www.trialregister.nl/trial/8437 of door naar de website www.trialregister.nl te gaan en te zoeken naar trialnummer NL8437. Voor vragen over deze studie kunt u contact met ons opnemen via mr-stat@umcutrecht.nl.

Journal Publications

• Sbrizzi, A., Van der Heide, O., Cloos, M., Van der Toorn, A., Hoogduin, H., Luijten, P.R. and Van den Berg, C.A., (2018). Fast quantitative MRI as a nonlinear tomography problem. Magnetic resonance imaging, 46, pp.56-63.

• Van der Heide, O., Sbrizzi, A., Luijten, P.R. and Van den Berg, C.A., (2020). High‐resolution in vivo MR‐STAT using a matrix‐free and parallelized reconstruction algorithm. NMR in Biomedicine, 33(4), p.e4251.

• Van der Heide, O., Sbrizzi, A., Luijten, P.R. and Van den Berg, C.A., (2020). Accelerated MR-STAT reconstructions using sparse Hessian approximations. IEEE Transactions on Medical Imaging. In press. Doi: 10.1109/TMI.2020.3003893