External validation of automated focal cortical dysplasia detection using morphometric analysis.

OBJECTIVE: Focal cortical dysplasias (FCDs) are a common cause of drug-resistant focal epilepsy but frequently remain undetected by conventional magnetic resonance imaging (MRI) assessment. The visual detection can be facilitated by morphometric analysis of T1-weighted images, for example, using the Morphometric Analysis Program (v2018; MAP18), which was introduced in 2005, independently validated for its clinical benefits, and successfully integrated in standard presurgical workflows of numerous epilepsy centers worldwide. Here we aimed to develop an artificial neural network (ANN) classifier for robust automated detection of FCDs based on these morphometric maps and probe its generalization performance in a large, independent data set. METHODS: In this retrospective study, we created a feed-forward ANN for FCD detection based on the morphometric output maps of MAP18. The ANN was trained and cross-validated on 113 patients (62 female, mean age ± SD =29.5 ± 13.6 years) with manually segmented FCDs and 362 healthy controls (161 female, mean age ± SD =30.2 ± 9.6 years) acquired on 13 different scanners. In addition, we validated the performance of the trained ANN on an independent, unseen data set of 60 FCD patients (28 female, mean age ± SD =30 ± 15.26 years) and 70 healthy controls (42 females, mean age ± SD = 40.0 ± 12.54 years). RESULTS: In the cross-validation, the ANN achieved a sensitivity of 87.4% at a specificity of 85.4% on the training data set. On the independent validation data set, our method still reached a sensitivity of 81.0% at a comparably high specificity of 84.3%. SIGNIFICANCE: Our method shows a robust automated detection of FCDs and performance generalizability, largely independent of scanning site or MR-sequence parameters. Taken together with the minimal input requirements of a standard T1 image, our approach constitutes a clinically viable and useful tool in the presurgical diagnostic routine for drug-resistant focal epilepsy.

MR evaluation of vessel size imaging of human gliomas: Validation by histopathology.

PURPOSE: To compare the vessel size and the cerebral blood volume in human gliomas with histopathology. Vessel size imaging (VSI) is a dynamic susceptibility contrast method for the assessment of the vessel size in normal and pathological tissue. Previous publications in rodents showed a satisfactory conformity with the vessel size derived from histopathology. To assess the clinical value, further, the progression-free interval was determined and correlated. MATERIALS AND METHODS: Twenty-five gliomas (WHO grade °II [n = 10], °III [n = 3], °IV [n = 12]) were prospectively included and received a stereotaxic biopsy after VSI. The vessel size and the cerebral blood volume (CBV) were calculated in regions of interest at the tumor edge and correlated with the vessel size measured by histopathology. RESULTS: Both VSI and CBV showed a good correlation with the vessel size in histopathology (up to r = 0.84, P < 0.001, and r = 0.62, P < 0.001, respectively). Slope and offset of the linear regression (y = 0.77x + 0.36 µm) suggest that the size of normal capillaries is overestimated with VSI, while for grossly enlarged vessels an underestimation occurs. Both VSI and CBV were negatively correlated with the progression-free interval (r = -0.57, P = 0.008, and r = -0.50, P = 0.025, respectively). CONCLUSION: The correlation between VSI and vessel size from histopathology is in good accordance with the animal studies. The overestimation of small capillary sizes is also known from the animal trials. Vessel size and CBV showed similar results, both for the correlation with the histopathological vessel size and the progression-free interval.

The value of dual-energy CTA for control of surgically clipped aneurysms.

OBJECTIVE: Comparison of image quality in DE-CTA with and without automatic head bone removal (BR) versus CTA with 16-detectors as a tool in postoperative evaluation of patients after neurosurgical clipping. METHODS: In this study 30 aneurysms that had undergone neurosurgical clipping were included: 18 with DE-CTA and 12 with conventional CTA. The images were further processed using the volume rendering technique (VRT) and BR. Two experienced neuroradiologists reviewed the images regarding the severity of artefacts surrounding the clip, visibility of the vessels and remnant necks. The results were compared with DSA images, if performed. RESULTS: Significantly fewer disturbances by artefacts were observed in DE-CTA versus CTA in a 16-row system. Visibility of the surrounding vessels was satisfying in both techniques and there were comparable results with DSA with only one exception. All images produced with 140 kV provided fewer artefacts than those with 80 kV. CONCLUSION: DE-CTA provides better image quality with fewer disturbances by clip artefact, a satisfying evaluation of remnant aneurysm necks and the surrounding vessels. As this method is easily performed and readily accessible with fast image post-processing using BR it provides an opportunity to avoid invasive DSA in the evaluation of suspected aneurysm rests.

Dynamic magnetic resonance imaging of swallowing and laryngeal motion using parallel imaging at 3 T.

OBJECT: To evaluate the feasibility of an optimized MRI protocol based on high field imaging at 3 T in combination with accelerated data acquisition by parallel imaging for the analysis of oropharyngeal and laryngeal function. MATERIALS AND METHODS: Fast 2D gradient echo (GRE) MRI with different spatial resolutions (1.7x2.7 and 1.1x1.5 mm2) and image update rates (4 and 10 frames per second) was employed to assess pharyngeal movements and visualize swallowing via tracking of an oral contrast bolus (blueberry juice). In a study with 10 normal volunteers, image quality was semi-quantitatively graded by three independent observers with respect to the delineation of anatomical detail and depiction of oropharynx and larynx function. Additionally, the feasibility of the technique for the visualization of pathological pre- and post-surgical oropharynx and larynx function was evaluated in a patient with inspiratory stridor. RESULTS: Image grading demonstrated the feasibility of dynamic MRI for the assessment of normal oropharynx and larynx anatomy and function. Superior image quality (P<.05) was found for data acquisition with four frames per second and higher spatial resolution. In the patient, dynamic MRI detected pathological hypermobility of the epiglottis resulting in airway obstruction. Additional post-surgical MRI for one clinical case revealed morphological changes of the epiglottis and improved function, i.e., absence of airway obstruction and normal swallowing. CONCLUSION: Results of the volunteer study demonstrated the feasibility of dynamic MRI at 3 T for the visualization of the oropharynx and larynx function during breathing, movements of the tongue and swallowing. Future studies are necessary to evaluate its clinical value compared to existing modalities based on endoscopy or radiographic techniques.

Vocal tract and register changes analysed by real-time MRI in male professional singers-a pilot study.

Changes of vocal tract shape accompanying changes of vocal register and pitch in singing have remained an unclear field. Dynamic real-time magnetic resonance imaging (MRI) was applied to two professional classical singers (a tenor and a baritone) in this pilot study. The singers sang ascending scales from B3 to G#4 on the vowel /a/, keeping the modal register throughout or shifting to falsetto register for the highest pitches. The results show that these singers made few and minor modifications of vocal tract shape when they changed from modal to falsetto and some clear modifications when they kept the register. In this case the baritone increased his tongue dorsum height, widened his jaw opening, and decreased his jaw protrusion, while the tenor merely lifted his uvula. The method used seems promising and should be applied to a greater number of singer subjects in the future.

Treatment of complex neurovascular lesions: an interdisciplinary angio suite approach.

OBJECTIVE: The objective of this study was to analyse our initial experience using an interdisciplinary angio suite approach to neurosurgical treatment of complex neurovascular lesions and expound technical feasibility and possible applications. SUBJECTS: Six out of 451 patients with cranial or spinal neurovascular lesions were surgically treated in the angio suite (biplane angiographic system) during a 28-month observation period. Clinical baseline data, radiological and intraoperative findings as well as clinical and radiological outcome were assessed. RESULTS: A ventral spinal perimedullary arteriovenous malformation, a ventral spinal perimedullary fistula, two diffuse frontal dural arteriovenous fistulas, a multifocal temporal arteriovenous malformation and a partially embolized fronto-temporo-basal dural arteriovenous fistula were successfully treated with angiographically confirmed complete occlusion and unimpaired neurological condition of the patients at the 12-month follow up. CONCLUSION: This study demonstrates the feasibility of this approach and points out possible indications, namely ventrally located spinal lesions and diffuse, deep seated cranial lesions.

A transmit/receive radiofrequency array for imaging the carotid arteries at 7 Tesla: coil design and first in vivo results.

OBJECTIVE: To develop a transmit/receive radiofrequency (RF) array for magnetic resonance imaging (MRI) of the carotid arteries at 7 T. The prototype is characterized in numerical simulations and bench measurements, and the feasibility of plaque imaging at 7 T is demonstrated in first in vivo images. MATERIALS AND METHODS: The RF phased array coil consists of 8 surface loop coils. To allow imaging of both sides of the neck, the RF array is divided into 2 coil clusters, each with 4 overlapping loop elements. For safety validation, numerical computations of the RF field distribution and the corresponding specific absorption rate were performed on the basis of a heterogeneous human body model. To validate the coil model, maps of the transmit B1(+) field were compared between simulation and measurement. In vivo images of a healthy volunteer and a patient (ulcerating plaque and a 50% stenosis of the right internal carotid artery) were acquired using a 3-dimensional FLASH sequence with a high isotropic spatial resolution of 0.54 mm as well as using pulse-triggered proton density (PD)/T2-weighted turbo spin echo sequences. RESULTS: Measurements of the S-parameters yielded a reflection and isolation of the coil elements of better than -18 and -13 dB, respectively. Measurements of the g-factor indicated good image quality for parallel imaging acceleration factors up to 2.4. A similar distribution and a very good match of the absolute values were found between the measured and simulated B1(+) transmit RF field for the validation of the coil model. In vivo images revealed good signal excitation of both sides of the neck and a high vessel-to-background image contrast for the noncontrast-enhanced 3-dimensional FLASH sequence. Imaging at 7 T could depict the extent of stenosis, and revealed the disruption and ulcer of the plaque. CONCLUSIONS: This study demonstrates that 2 four-channel transmit/receive RF arrays for each side of the neck is a suitable concept for in vivo MRI of the carotid arteries at 7 Tesla. Further studies are needed to explore and exploit the full potential of 7 T high-field MRI for carotid atherosclerotic plaque imaging.

Imaging of patients with hippocampal sclerosis at 7 Tesla: initial results.

RATIONALE AND OBJECTIVES: Focal epilepsies potentially can be cured by neurosurgery; other treatment options usually remain symptomatic. High-resolution magnetic resonance (MR) imaging is the central imaging strategy in the evaluation of focal epilepsy. The most common substrate of temporal epilepsies is hippocampal sclerosis (HS), which cannot always be sufficiently characterized with current MR field strengths. Therefore, the purpose of our study was to demonstrate the feasibility of high-resolution MR imaging at 7 Tesla in patients with focal epilepsy resulting from a HS and to improve image resolution at 7 Tesla in patients with HS. MATERIALS AND METHODS: Six patients with known HS were investigated with T1-, T2-, T2(*)-, and fluid-attenuated inversion recovery-weighted sequences at 7 Tesla with an eight-channel transmit-receive head coil. Total imaging time did not exceed 90 minutes per patient. RESULTS: High-resolution imaging at 7 Tesla is feasible and reveals high resolution of intrahippocampal structures in vivo. HS was confirmed in all patients. The maximum non-interpolated in-plane resolution reached 0.2 x 0.2 mm(2) in T2(*)-weighted images. The increased susceptibility effects at 7 Tesla revealed identification of intrahippocampal structures in more detail than at 1.5 Tesla, but otherwise led to stronger artifacts. Imaging revealed regional differences in hippocampal atrophy between patients. The scan volume was limited because of specific absorption rate restrictions, scanning time was reasonable. CONCLUSIONS: High-resolution imaging at 7 Tesla is promising in presurgical epilepsy imaging. "New" contrasts may further improve detection of even very small intrahippocampal structural changes. Therefore, further investigations will be necessary to demonstrate the potential benefit for presurgical selection of patients with various lesion patterns in mesial temporal epilepsies resulting from a unilateral HS.