Isocitrate dehydrogenase (IDH)-1 mutation is an important prognostic factor and a potential therapeutic target in glioma. Immunohistological and molecular diagnosis of IDH mutation status is invasive. To avoid tumor biopsy, dedicated spectroscopic techniques have been proposed to detect D-2-hydroxyglutarate (2-HG), the main metabolite of IDH, directly in vivo. However, these methods are technically challenging and not broadly available. Therefore, we explored the use of machine learning for the non-invasive, inexpensive and fast diagnosis of IDH status in standard ¹H-magnetic resonance spectroscopy (¹H-MRS). To this end, 30 of 34 consecutive patients with known or suspected glioma WHO grade II-IV were subjected to metabolic positron emission tomography (PET) imaging with O-(2-¹⁸F-fluoroethyl)-L-tyrosine (¹⁸F-FET) for optimized voxel placement in ¹H-MRS. Routine ¹H-magnetic resonance (¹H-MR) spectra of tumor and contralateral healthy brain regions were acquired on a 3 Tesla magnetic resonance (3T-MR) scanner, prior to surgical tumor resection and molecular analysis of IDH status. Since 2-HG spectral signals were too overlapped for reliable discrimination of IDH mutated (IDHmut) and IDH wild-type (IDHwt) glioma, we used a nested cross-validation approach, whereby we trained a linear support vector machine (SVM) on the complete spectral information of the ¹H-MRS data to predict IDH status. Using this approach, we predicted IDH status with an accuracy of 88.2%, a sensitivity of 95.5% (95% CI, 77.2–99.9%) and a specificity of 75.0% (95% CI, 42.9–94.5%), respectively. The area under the curve (AUC) amounted to 0.83. Subsequent ex vivo ¹H-nuclear magnetic resonance (¹H-NMR) measurements performed on metabolite extracts of resected tumor material (eight specimens) revealed myo-inositol (M-ins) and glycine (Gly) to be the major discriminators of IDH status. We conclude that our approach allows a reliable, non-invasive, fast and cost-effective prediction of IDH status in a standard clinical setting.

Objective: The aim of our study was to systematically compare two-point Dixon fat suppression (FS) and spectral FS techniques in contrast enhanced imaging of the head and neck region. Methods: Three independent readers analysed coronal T 1 weighted images recorded after contrast medium injection with Dixon and spectral FS techniques with regard to FS homogeneity, motion artefacts, lesion contrast, image sharpness and overall image quality. Results: 85 patients were prospectively enrolled in the study. Images generated with Dixon-FS technique were of higher overall image quality and had a more homogenous FS over the whole field of view compared with the standard spectral fat-suppressed images (p < 0.001). Concerning motion artefacts, flow artefacts, lesion contrast and image sharpness no statistically significant difference was observed. Conclusion: The Dixon-FS technique is superior to the spectral technique due to improved homogeneity of FS and overall image quality while maintaining lesion contrast. Advances in knowledge: T 1 with Dixon FS technique offers, compared to spectral FS, significantly improved FS homogeneity and over all image quality in imaging of the head and neck region.

(1) Background: T1 mapping in magnetic resonance imaging (MRI) of the liver has been proposed to estimate liver function or to detect the stage of liver disease, among others. Thus far, the impact of intrahepatic fat on T1 quantification has only been sparsely discussed. Therefore, the aim of this study was to evaluate the potential of water–fat separated T1 mapping of the liver. (2) Methods: A total of 386 patients underwent MRI of the liver at 3 T. In addition to routine imaging techniques, a 3D variable flip angle (VFA) gradient echo technique combined with a two-point Dixon method was acquired to calculate T1 maps from an in-phase (T1_in) and water-only (T1_W) signal. The results were correlated with proton density fat fraction using multi-echo 3D gradient echo imaging (PDFF) and multi-echo single voxel spectroscopy (PDFF_MRS). Using T1_in and T1_W, a novel parameter FF_T1 was defined and compared with PDFF and PDFF_MRS. Furthermore, the value of retrospectively calculated T1_W (T1_W_calc) based on T1_in and PDFF was assessed. Wilcoxon test, Pearson correlation coefficient and Bland–Altman analysis were applied as statistical tools. (3) Results: T1_in was significantly shorter than T1_W and the difference of both T1 values was correlated with PDFF (R = 0.890). FF_T1 was significantly correlated with PDFF (R = 0.930) and PDFF_MRS (R = 0.922) and yielded only minor bias compared to both established PDFF methods (0.78 and 0.21). T1_W and T1_W_calc were also significantly correlated (R = 0.986). (4) Conclusion: T1_W acquired with a water–fat separated VFA technique allows to minimize the influence of fat on liver T1. Alternatively, T1_W can be estimated retrospectively from T1_in and PDFF, if a Dixon technique is not available for T1 mapping.

Background: Tractography based on diffusion-weighted magnetic resonance imaging (DWI) models the structural connectivity of the human brain. Deep brain stimulation (DBS) targeting the subthalamic nucleus is an effective treatment for advanced Parkinson's disease, but may induce adverse effects. This study investigated the relationship between structural connectivity patterns of DBS electrodes and stimulation-induced side effects. Materials and Methods: Twenty-one patients with Parkinson's disease treated with bilateral subthalamic DBS were examined. Overall, 168 electrode contacts were categorized as inducing or noninducing depending on their capability for inducing side effects such as motor effects, paresthesia, dysarthria, oculomotor effects, hyperkinesia, and other complications as assessed during the initial programming session. Furthermore, the connectivity of each contact with target regions was evaluated by probabilistic tractography based on DWI. Finally, stimulation sites and structural connectivity patterns of inducing and noninducing contacts were compared. Results: Inducing contacts differed across the various side effects and from those mitigating Parkinson's symptoms. Although contacts showed a largely overlapping spatial distribution within the subthalamic region, they could be distinguished by their connectivity patterns. In particular, inducing contacts were more likely connected with supplementary motor areas (hyperkinesia, dysarthria), frontal cortex (oculomotor), fibers of the internal capsule (paresthesia), and the basal ganglia-thalamo-cortical circuitry (dysarthria). Discussion: Side effects induced by DBS seem to be associated with distinct connectivity patterns. Cerebellar connections are hardly associated with side effects, although they seem relevant for mitigating motor symptoms in Parkinson's disease. A symptom-specific, connectivity-based approach for target planning in DBS may enhance treatment outcomes and reduce adverse effects. Impact statement: Tractography based on diffusion-weighted magnetic resonance imaging has become a prominent technique for investigating the connectivity of human brain networks in vivo . However, the relationship between structural connections and brain function is still hardly known. The present study examined the relationship between adverse behavioral effects induced by deep brain stimulation (DBS) and tractography patterns in individual brains. The results suggest that DBS-based side effects depend on the structural connections of electrode contacts rather than their location. Network-based target planning in DBS may improve treatment by avoiding side effects. Moreover, the adopted approach may serve as a paragon for investigating structure/function relationships.

The isocitrate dehydrogenase (IDH) mutation status is an indispensable prerequisite for diagnosis of glioma (astrocytoma and oligodendroglioma) according to the WHO classification of brain tumors 2021 and is a potential therapeutic target. Usually, immunohistochemistry followed by sequencing of tumor tissue is performed for this purpose. In clinical routine, however, non-invasive determination of IDH mutation status is desirable in cases where tumor biopsy is not possible and for monitoring neuro-oncological therapies. In a previous publication, we presented reliable prediction of IDH mutation status employing proton magnetic resonance spectroscopy (¹H-MRS) on a 3.0 Tesla (T) scanner and machine learning in a prospective cohort of 34 glioma patients. Here, we validated this approach in an independent cohort of 67 patients, for which ¹H-MR spectra were acquired at 1.5 T between 2002 and 2007, using the same data analysis approach. Despite different technical conditions, a sensitivity of 82.6% (95% CI, 61.2–95.1%) and a specificity of 72.7% (95% CI, 57.2–85.0%) could be achieved. We concluded that our ¹H-MRS based approach can be established in a routine clinical setting with affordable effort and time, independent of technical conditions employed. Therefore, the method provides a non-invasive tool for determining IDH status that is well-applicable in an everyday clinical setting.

We aimed to evaluate whether U-shaped convolutional neuronal networks can be used to segment liver parenchyma and indicate the degree of liver fibrosis/cirrhosis at the voxel level using contrast-enhanced magnetic resonance imaging. This retrospective study included 112 examinations with histologically determined liver fibrosis/cirrhosis grade (Ishak score) as the ground truth. The T1-weighted volume-interpolated breath-hold examination sequences of native, arterial, late arterial, portal venous, and hepatobiliary phases were semi-automatically segmented and co-registered. The segmentations were assigned the corresponding Ishak score. In a nested cross-validation procedure, five models of a convolutional neural network with U-Net architecture (nnU-Net) were trained, with the dataset being divided into stratified training/validation (n = 89/90) and holdout test datasets (n = 23/22). The trained models precisely segmented the test data (mean dice similarity coefficient = 0.938) and assigned separate fibrosis scores to each voxel, allowing localization-dependent determination of the degree of fibrosis. The per voxel results were evaluated by the histologically determined fibrosis score. The micro-average area under the receiver operating characteristic curve of this seven-class classification problem (Ishak score 0 to 6) was 0.752 for the test data. The top-three-accuracy-score was 0.750. We conclude that determining fibrosis grade or cirrhosis based on multiphase Gd-EOB-DTPA-enhanced liver MRI seems feasible using a 2D U-Net. Prospective studies with localized biopsies are needed to evaluate the reliability of this model in a clinical setting.