Search

Purpose: To adapt the non-uniform Self-Gating (nuSG) method and compare it to established self-gating approaches for lung imaging in uniform and highly irregular respiratory patterns.Methods: Six healthy volunteers underwent free breathing lung MRI using a radial tiny golden angle ultrashort echo-time sequence. Acquisitions were performed with the volunteer breathing as uniformly as possible and with a deliberately non-uniform respiratory pattern. The acquired data was reconstructed with the nuSG method, previously introduced for cardiac imaging and imaging of the temporomandibular joint (TMJ) and compared to established k-space based and image-based self-gating approaches. Residual motion blur, SNR and functional values were assessed and compared to reference breath-hold acquisitions.Results: nuSG is capable of reconstructing high-quality images for uniform and non-uniform breathing patterns and is furthermore capable of resolving motion in cases where additional motion is superimposed or no clear motion surrogate exists. Derived functional values do not differ significantly from other image-based gated reconstructions - and in the case of non-uniform respiratory patterns replicate the reference BH values.Conclusion: Image based approaches are computationally more demanding but yield better results in all aspects. In scenarios with a direct surrogate for respiratory motion (i.e. the lung-liver interface) the extraction of a one-dimensional navigator is sufficient. When there is no direct surrogate for the motion of the target structure available (e.g., considerable through-plane motion or a different source of motion), the two-dimensional correlation-based measure used in nuSG is able to track the motion more accurately.

Majority of lower grade glioma (LGG) are located eloquently rendering surgical resection challenging. Aim of our study was to assess rate of permanent deficits and its predisposing risk factors. We retrieved 83 patients harboring an eloquently located LGGs from the prospective LoG-Glio Database. Patients without surgery or incomplete postoperative data were excluded. Sign rank test, explorative correlations by Spearman ρ and multivariable regression for new postoperative deficits were calculated. Eloquent region involved predominantly motor (45%) and language (40%). At first follow up after 3 months permanent neuro-logical deficits (NDs) were noted in 39%. Mild deficits remained in 29% and severe deficits in 10%. Complete tumor removal (CTR) was successfully in 62% of intended cases. Postoperative and 3-month follow up National Institute of Health Stroke Score (NIHSS) showed significantly lower values than preoperatively (p0 (p=0.021, OR 8.5) were independent predictors for permanent postoperative deficit according to NIHSS at 3-month according to multivariable regression model. Patients harboring eloquently located LGG are highly vulnerable for permanent deficits. Almost one third of patients have a permanent reduction of their functional status based on ECOG. Risk of an extended resection has to be balanced with the respective oncological benefit. Especially, patients with impaired pre-operative status are at risk for new permanent deficits. There is a relevant improvement of neurological symptoms in the first year after surgery, especially for patients with slight aphasia.

To identify progressive supranuclear palsy (PSP), we combined voxel-based morphometry (VBM) and support vector machine (SVM) classification using disease-specific features in multicentric magnetic resonance imaging (MRI) data. Structural brain differences were investigated at four centers between 20 patients with PSP and 20 age-matched healthy controls with T1-weighted MRI at 3T. To pave the way for future application in personalized medicine, we applied SVM classification to identify PSP on an individual level besides group analyses based on VBM. We found a major decline in gray matter density in the brainstem, insula, and striatum, and also in frontomedian regions, which is in line with current literature. Moreover, SVM classification yielded high accuracy rates above 80% for disease identification in imaging data. Focusing analyses on disease-specific regions-of-interest (ROI) led to higher accuracy rates compared to a whole-brain approach. Using a polynomial kernel (instead of a linear kernel) led to an increased sensitivity and a higher specificity of disease detection. Our study supports the application of MRI for individual diagnosis of PSP, if combined with SVM approaches. We demonstrate that SVM classification provides high accuracy rates in multicentric data—a prerequisite for potential application in diagnostic routine.