返回Google圖書搜尋

Advancements to the Reference Region Model for Dynamic Contrast Enhanced MRI

出版	McGill University Libraries, 2020
URL	http://books.google.com.hk/books?id=Gv8vzgEACAAJ&hl=&source=gbs_api

註釋"Background: Dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) provides quantitative information on tissue blood supply by fitting a model to the acquired data. Conventional models require the arterial input function (AIF) which is difficult to measure. The reference region model (RRM) can quantify perfusion by using a reference tissue to circumvent the AIF. However, the RRM suffers from some major limitations: (i) it is only valid for simple tissues, (ii) it requires prior knowledge of the reference tissue parameters, and (iii) it requires a well-characterized healthy tissue, such as muscle, near the tissue of interest. Methods:Four advances were contributed to reference region modelling of DCE-MRI. The scope of the RRM was expanded by deriving new versions based on the extended Tofts model and the more generalized two-compartment models. The reference region and input function tail (RRIFT) approach was developped for estimating the reference tissue parameters by using the AIF tail. The AIF tail is the washout portion and is considerably easier to measure than the full AIF. Lastly, the self-referenced approach was proposed to automatically identify a reference region from within the tissue of interest. Results:The new RRMs were feasibly applied in highly vascularized tumours and provided perfusion estimates that were consistent with advanced conventional models. The RRIFT approach was able to estimate reference tissue parameters using either a measured AIF tail or an assumed literature-based AIF tail. The self-reference approach provided quantitative perfusion estimates with the only input being the voxels in the tissue of interest. The key advantage of the new RRM models over conventional models was that the RRM does not need the complete measured AIF and they remain feasible at temporal resolutions as low as 30 seconds. Conclusion:The presented advancements extend the range of application of the RRM over a wide spectrum of tissues and acquisition conditions, and can provide quantitative perfusion estimates without the need for a measured AIF or prior knowledge of reference tissue parameters"--