Periodic Reporting for period 1 - MRI COMIQSUM (MRI COntrast using MIcrobubbles in Quantitative SUsceptibility Mapping)
Reporting period: 2018-06-01 to 2020-05-31
My overall research goal is to develop Quantitative Susceptibility Mapping (QSM) techniques to improve Magnetic Resonance Imaging (MRI) of microbubbles. Although microbubbles are a well-established intravascular ultrasound contrast agent, a few studies have shown that they can enhance the image contrast in MRI. There is increasing interest in MRI-guided microbubble-mediated focused ultrasound treatments such as thermal surgery and targeted delivery of drugs, antibodies or genes, especially in regions like the brain which are difficult to treat using conventional surgery and treatments. State-of-the-art MRI techniques used to detect microbubbles exploit the local decrease in the MRI signal magnitude due to the magnetic susceptibility difference between microbubbles and the surrounding tissues. Magnetic susceptibility is a property telling us how much a tissue or material becomes magnetised in an external magnetic field and can be measured directly using QSM. Signal magnitude decreases occur not only around microbubbles but also close to veins or other strong susceptibility sources, making it difficult to separate these effects and confounding MRI of microbubbles. Furthermore, the signal magnitude has a non-linear and non-local dependence on microbubble size and volume fraction so it is difficult to relate signal decreases to the quantity of microbubbles at a specific location.
In this fellowship I have developed a fast MRI QSM technique to allow much more direct detection of microbubbles with the potential for dynamic tracking of microbubble concentration, destruction and clearance. MRI signal is complex, i.e. it has two constituents: magnitude and phase. QSM uses the phase, which has previously been discarded in MRI of microbubbles. Use of the phase information to visualise human brain anatomy has resulted in dramatic improvements in the image contrast-to-noise ratio, especially at high magnetic fields (3 Tesla and above). However, the phase contrast is non-local and orientation dependent, making it difficult to interpret. QSM overcomes these problems by calculating from the phase signal the underlying magnetic susceptibility, which more closely represents tissue composition, promising quantification of microbubble concentration or dose calculation in targeted delivery of drugs. The technique I have developed for fast imaging of microbubbles using QSM may facilitate broader application of MRI-guided microbubble-mediated focused ultrasound treatments in the future.
In this fellowship I have developed a fast MRI QSM technique to allow much more direct detection of microbubbles with the potential for dynamic tracking of microbubble concentration, destruction and clearance. MRI signal is complex, i.e. it has two constituents: magnitude and phase. QSM uses the phase, which has previously been discarded in MRI of microbubbles. Use of the phase information to visualise human brain anatomy has resulted in dramatic improvements in the image contrast-to-noise ratio, especially at high magnetic fields (3 Tesla and above). However, the phase contrast is non-local and orientation dependent, making it difficult to interpret. QSM overcomes these problems by calculating from the phase signal the underlying magnetic susceptibility, which more closely represents tissue composition, promising quantification of microbubble concentration or dose calculation in targeted delivery of drugs. The technique I have developed for fast imaging of microbubbles using QSM may facilitate broader application of MRI-guided microbubble-mediated focused ultrasound treatments in the future.
In this fellowship I have optimised an acquisition strategy tailored for quantitative susceptibility mapping of microbubbles in test objects and in a live rat model. I have developed an effective image processing pipeline optimised for susceptibility calculation from phase images of microbubbles in the rat model at 9.4 Tesla. Using measurements in a test object containing different concentrations of microbubbles, I determined the minimum microbubble concentration detectable using an ultra-high field MRI scanner (9.4 Tesla) and derived an empirical relationship between the measured susceptibility and the microbubble concentration. In a live rat model, I showed that it is possible to detect microbubble related signal changes using fast MRI acquisition and processing techniques optimised during this fellowship. Some elements of the image processing pipeline were also tested using a test object at the lower clinically accepted field strength of 3 Tesla. More experiments would need to be performed at 3T MRI scanners to assess the feasibility of detecting microbubbles at lower field strengths.
I have disseminated my work at three international conferences: two Annual Meetings of International Society for Magnetic Resonance in Medicine (ISMRM) in 2019 and 2020, and the 5th International Workshop on MRI Phase Contrast & QSM in 2019. I also attended and presented my work at two national meetings of the British Chapter of the ISMRM in 2018 and 2019. A manuscript presenting one of the image processing methods, namely the ROMEO phase unwrapping algorithm, developed during my fellowship was published in the Magnetic Resonance in Medicine journal. A manuscript summarising my results showing detection of the microbubbles in test objects and in vivo at 9.4 Tesla is currently in preparation.
I have disseminated my work at three international conferences: two Annual Meetings of International Society for Magnetic Resonance in Medicine (ISMRM) in 2019 and 2020, and the 5th International Workshop on MRI Phase Contrast & QSM in 2019. I also attended and presented my work at two national meetings of the British Chapter of the ISMRM in 2018 and 2019. A manuscript presenting one of the image processing methods, namely the ROMEO phase unwrapping algorithm, developed during my fellowship was published in the Magnetic Resonance in Medicine journal. A manuscript summarising my results showing detection of the microbubbles in test objects and in vivo at 9.4 Tesla is currently in preparation.
QSM of microbubbles is a novel idea as only conventional magnitude imaging has been used to detect this contrast agent with MRI to date. QSM offers several exciting advantages over MRI magnitude images:
1 Susceptibility maps calculated from phase images have improved contrast-to-noise ratio over magnitude images, particularly at high magnetic fields such as 9.4 Tesla.
2 QSM yields quantitative maps of an intrinsic tissue property, whereas magnitude images provide only qualitative information: QSM will allow quantification of microbubble concentration.
3 MRI techniques to be used for QSM provide a complex signal. The signal phase is used for susceptibility calculation and the magnitude is always available in addition so the standard contrast can still be obtained ‘for free’.
Clinical translation of fast imaging and quantification of microbubbles using QSM would greatly facilitate the use of microbubbles as a dual-modality ultrasound-MRI contrast agent, enabling more accurate diagnosis than each modality separately. Additionally, this new technique could potentially facilitate MRI-guided ultrasound therapies such as: bleeding-free treatment of thrombosis¬, targeted drug, antibody and gene delivery and focused ultrasound surgery which is approved for treating uterine fibroids or bone metastases and is being developed to treat prostate, breast and brain diseases. QSM of microbubbles will benefit all these applications by enabling MRI to accurately guide and time focused ultrasound exposure by measuring the concentration and detecting the destruction of microbubbles as a marker of drug delivery. Techniques developed during this fellowship have potential to broaden the applicability of these therapies and facilitate their translation into the clinic.
1 Susceptibility maps calculated from phase images have improved contrast-to-noise ratio over magnitude images, particularly at high magnetic fields such as 9.4 Tesla.
2 QSM yields quantitative maps of an intrinsic tissue property, whereas magnitude images provide only qualitative information: QSM will allow quantification of microbubble concentration.
3 MRI techniques to be used for QSM provide a complex signal. The signal phase is used for susceptibility calculation and the magnitude is always available in addition so the standard contrast can still be obtained ‘for free’.
Clinical translation of fast imaging and quantification of microbubbles using QSM would greatly facilitate the use of microbubbles as a dual-modality ultrasound-MRI contrast agent, enabling more accurate diagnosis than each modality separately. Additionally, this new technique could potentially facilitate MRI-guided ultrasound therapies such as: bleeding-free treatment of thrombosis¬, targeted drug, antibody and gene delivery and focused ultrasound surgery which is approved for treating uterine fibroids or bone metastases and is being developed to treat prostate, breast and brain diseases. QSM of microbubbles will benefit all these applications by enabling MRI to accurately guide and time focused ultrasound exposure by measuring the concentration and detecting the destruction of microbubbles as a marker of drug delivery. Techniques developed during this fellowship have potential to broaden the applicability of these therapies and facilitate their translation into the clinic.