Startseite » Research

Radiomics- Classification based on Imaging Data

The complexity of imaging data has been steadily increasing over the past decades. Cross-section imaging methods such as computed tomography (CT) and magnetic resonance imaging (MRI) have added a wealth of additional information to imaging including three- and four-dimensional information and a myriad of different imaging signals and contrasts. Only a limited part of this information can be processed visually by a human observer. To make more of this information amenable to diagnostic decision making, it is essential to establish new post-processing algorithms. Imaging data need to be normalized and segmented and to be eventually translated back into quantitative information. Parameters to be extracted include homogeneity and heterogeneity, texture, signal intensity and information on shape. These extracted features need to be classified by using machine learning and supervised learning algorithms. Our current projects include radiomics approaches for brain tumors (gliomas) as well as for prostate cancer.

Specialization of functional connectivity architecture of the human brain

The intrinsic resting state activity of the brain, measured by means of functional magnetic resonance with blood oxygenation level dependent (BOLD) sequences, provides a measure of brain activation in the state of relaxed consciousness. In collaboration with the Department of Neonatology we are currently examining the lateralisation of intrinsic functional connectivity as a maturation parameter within the human brain. In another project in collaboration with the Department of Psychiatry we are investigating changes in hemispherical specialization in patients with chronic schizophrenia, asymptomatic family members and healthy control subjects. Mueller S, Wang D, Fox MD, Yeo BBT, Sepulcre J, Sabuncu MR, Shafee R, Lu J, Liu H. Individual Variability in Functional Connectivity Architecture of the Human Brain. Neuron, 2013 Feb.

High-resolution structural imaging with 3D-segmentation techniques

MR imaging is able to delineate structures within the neurocranium both in a high contrast resolution and in an excellent spatial resolution. One focus of our group lies on high-resolution structural imaging of the inner ear. A potential morphological correlate of various inner ear pathologies is a disturbed relation between endolymphatic and the perilymphatic fluid spaces, e.g. due to endolymphatic hydrops. Perilymphatic and endolymphatic spaces of the labyrinth can be assessed, differentiated and segmented by high-resolution MR imaging after contrast media application. There is a significant correlation between audio-vestibular function and the degree of endolymphatic hydrops. Gürkov R, Berman A, Dietrich O, Flatz W, Jerin C, Krause E, Keeser D, Ertl-Wagner B. MR volumetric assessment of endolymphatic hydrops. Eur Radiol. 2015 Feb;25(2):585-95.

Structural Connectivity within Neuroimmunological Diseases

Diffusion Tensor Imaging provides information about white matter microstructure and fiber tract integrity by measuring the Brownian motion of water molecules and their degree of anisotropy in the brain. Tractography algorithms follow the strongest diffusion direction, thereby creating a virtual fiber path that represents anatomical tracts. In a collaboration project with the Institute of Clinical Neuroimmunology in the context of "SyNergy- Munich Cluster for Systems Neurology" we investigate the structural connectivity of the visual system in patients with neuoimmunological pathologies such as multiple sclerosis and to correlate structural and functional connectivity patterns to markers of microglial activation. Blaschek A, Keeser D, Müller S, Koerte I, Schröder AS, Müller-Felber W, Heinen F, Ertl-Wagner B. Early white matter changes in childhood multiple sclerosis: A DTI study. AJNR Am J Neuroradiol; 2013

Neurofeedback using realtime fMRI or simultaneous EEG-fMRI

Based on the BOLD-effect, realtime functional magnetic resonance imaging (rt-fMRI) enables the observation of one’s own brain activity while lying in the scanner. In contrast to EEG, rt-fMRI allows the application of neurofeedback to deeper subcortical areas of the brain known to be involved in various emotional and cognitive processes. As many neuropsychiatric disorders are associated with functional changes in specific subcortical brain regions or networks, rt-fMRI neurofeedback appears to be a promising therapeutic tool that may form an integral part of treatment concepts in the future. Karch S, Keeser D, Hümmer S, Paolini M, Kirsch V, Karali T, Kupka M, Rauchmann BS, Chrobok A, Blautzik J, Koller G, Ertl-Wagner B, Pogarell O. Modulation of Craving Related Brain Responses Using Real-Time fMRI in Patients with Alcohol Use Disorder. PLoS One. 2015 Jul 23;10(7):e0133034.

Magnetic Resonance Spectroscopy in combination with EEG

Magnetic Resonance Spectroscopy uses signal frequency changes of hydrogen-nuclei within different molecules to gain quantitative information about the amount of certain metabolites. In a current study in collaboration with the Nuffield Department of Clinical Neurosciences- University of Oxford, we used simultaneous EEG-MRS recordings to assess the course of in vivo GABA levels of the human motorcortex (M1) during different sleep stages and wakefulness. Stagg CJ. Magnetic Resonance Spectroscopy as a tool to study the role of GABA in motor-cortical plasticity. Neuroimage. 2014 Feb 1;86:19-27.

TDCS in combination with fMRI

Transcranial direct current stimulation (tDCS) has been proposed for experimental and therapeutic modulation of regional brain function. Daniel Keeser is project leader investigating disorder-tailored Direct Current Stimulation of the prefrontal cortex using multimodal imaging methods and supported by the Federal Ministery of Education and Research. Different prefrontal tDCS protocols will be investigated in healthy subjects and major depressive/anxiety disordered patients with the aim to study the interaction of noninvasive brain stimulation and functional MRI connectivity. Keeser D et al., Prefrontal transcranial direct current stimulation changes connectivity of resting-state networks during fMRI. J Neurosci. 2011 Oct 26;31(43):15284-93.

Assessment of Perfusion and Permeability Parameters with Dynamic Contrast-Enhanced (DCE) MRI

Cerebral hemodynamics are based on a variety of parameters such as cerebral blood flow (CBF), cerebral blood volume (CBV) and the integrity or leakage of the blood-brain (BB) barrier. Our research focuses on the investigation of cerebral hemodynamics by means of T1-weighted, dynamic, contrast-enhanced (DCE) MRI. Ingrisch M, Sourbron S, Morhard D, Ertl-Wagner B, Kümpfel T, Hohlfeld R, Reiser M, Glaser C. Quantification of perfusion and permeability in multiple sclerosis: dynamic contrast-enhanced MRI in 3D at 3T. Invest Radiol. 2012 Apr;47(4):252-8.

Assessment of cerebral hemo- and hydrodynamic parameters and intracranial pressure by phase contrast MRI techniques

Phase contrast MR sequences are able to quantify flow – a concept that has long been applied in cardiac MR imaging. In addition to measuring arterial and venous flow, phase contrast MR imaging can also quantify cerebrospinal fluid (CSF) flow. In collaboration with the Physiologic Imaging and Modelling Laboratory of the University of Miami, FL, USA (Noam Alperin, PhD) we have established and tested this methodology at our institute and have applied it in various diseases including hydrocephalus, migraine, arteriovenous malformations and traumatic brain injuries. Pomschar A, (...), Ertl-Wagner B, Alperin N. MRI evidence for altered venous drainage and intracranial compliance in mild traumatic brain injury. PLoS One. 2013;8(2)

Assessment of Acute Stroke with CT-Perfusion and CT Angiography

In acute stroke, a fast and precise assessment is of utmost importance for rapid therapeutic decision-making. In various projects we have optimized CT-angiographic and CT perfusion parameters and have assessed techniques to differentiate arterial and venous vessels by linear combinations of quantitative time-density curve characteristics. Havla L, Schneider M, Thierfelder KM, Beyer SE, Ertl-Wagner B, Sommer WH, Dietrich O. Validation of a method to differentiate arterial and venous vessels in CT perfusion data using linear combinations of quantitative time-density curve characteristics. Eur Radiol. 2015 Oct;25(10):2937-44.

Assessment of cerebral hemo- and hydrodynamic parameters and intracranial pressure by phase contrast MRI techniques

In acute stroke, a fast and precise assessment is of utmost importance for rapid therapeutic decision-making. In various projects we have optimized CT-angiographic and CT perfusion parameters and have assessed techniques to differentiate arterial and venous vessels by linear combinations of quantitative time-density curve characteristics. Havla L, Schneider M, Thierfelder KM, Beyer SE, Ertl-Wagner B, Sommer WH, Dietrich O. Validation of a method to differentiate arterial and venous vessels in CT perfusion data using linear combinations of quantitative time-density curve characteristics. Eur Radiol. 2015 Oct;25(10):2937-44.
 
 

Interdisciplinary Multimodal Imaging

Univ. Prof. Dr. med. Birgit Ertl-Wagner, MHBA

Institute for Clinical Radiology

Ludwig-Maximilians-University

Marchioninistr. 15

81377 Munich


Phone: +49 89 4400 7 3250
Mail: Birgit.Ertl-Wagner@med.lmu.de