As described in my about me, my research has two main aspects:
1) Translating MRI developments to the clinic
2) Translating between clinical and pre-clinical research.
Below, I describe my published data in projects relating to these two themes.
Translating MRI Developments to the Clinic
1: Standardized T1w/T2w Ratio
In some neurological disorders (for example multiple sclerosis and mutliple system atrophy), there is an increasing awareness that there is damage in the white matter that appears healthy on a typical MRI image. This “normal-appearing” white matter can typically only be investigated using advanced MRI techniques. While these advances have allowed us to better understand the pathology underlying NAWM, the advanced nature of these techniques limit their applicability in the clinic - they typically require additional scan time and analytic expertise that can be limited in a pressured radiology clinic.
It has recently been shown that the ratio of two typically-acquired scans, T1-weighted and T2-weighted scans (T1w/T2w ratio), may reflect microstructural properties of white matter. As these are typically required as part of the clinical routine, the T1w/T2w ratio shows promise as a measure of normal-appearing white matter that could be applied in the clinic. However, the application of the T1w/T2w ratio in the clinic is limited because of some important technical issues - the quantitative value calculated by the ratio is sensitive to scanner-specific variation and choice of paramaters to acquire the T2-weighted image.
Elsewhere in the MRI field it was shown that standardizing the ratio significantly reduces scanner-specific effects and produces a scaled value between -1 and 1. This overcomes the barriers to clinical application as the values output can be compared with values output from another scanner or a different individual.
In our work, we are investigating the clinical applicability of the standardized T1w/T2w ratio as a measure of NAWM in multiple sclerosis. As a first step, we showed that the standardization method was robust against disease-related irregularities in the scan and that it was sensitive to tissue damage in established multiple sclerosis patients. Our next steps will involve evaluating this measure in early MS, investigating how it changes over time and in different centers as well as using it in retrospective analyses where only clinical scan data is available.
In another disease, multiple system atrophy, it is known that hyperintensities in the middle cerebellar peduncle can be visually detected on clinical MRI scans and that the presence of these hyperintensities is sensitive to the diagnosis of multiple system atrophy. However, this visual assessment is not very quantifiable and is time-consuming for the radiologist if viewing multiple scans. In a collaboration with colleagues in Chiba, Japan, we calculated the standardized T1w/T2w ratio in the middle cerebellar peduncle of these patients and found that this quantification of damage was also highly sensitive to diagnosis and was related to clinical disability. You can read more about this in the paper. We are further investigating applications of the standardized T1w/T2w in multiple system atrophy in this exciting international collaboration.
If you are interested in applying the technique on your data or have any questions about our approach, please check out the paper and feel free to email me using the link below!
2: Quantitative Multi-Parameter Mapping
As mentioned above, advanced quantitative MRI techniques can have limited clinical applicability due to a need for increased scan time, despite their many benefits. In this project, we optimize a quantitative multi-parameter mapping technique for the clinic. This technqiue can be used to measure four different whole-brain quantitative maps, with high scan-rescan stability and high reliability between scanners and over time. The original method is applied at high-resolution (0.8 mm isotropic or 1.0 mm isotropic) and takes around 20 minutes per patient. The relatively long scan time limits the clinical applicability, although it is still very useful for research purposes. We optimized this method for the clinic by reducing the resolution to 1.6 mm isotropic, which reduced the scan time to around 7 minutes. The reduction in resolution also increased the scan-rescan stability by around 40%. A further optimization was to correct for Gibb’s ringing artefacts, which further improved stability. In the paper we describe this optimization in detail and also provide stability data for 52 regions of the brain to aid researchers and clinicians in interpreting data and planning future studies. Please check out the paper and our website for more information. We are very excited about the clinical potential of this optimized technique!
Translating between Clinical and Pre-Clinical Research (Bench to Bedside and Back Again)
I haven’t published any papers in this area yet but watch this space!