Lukas Ramrath

A robotic assistant for stereotactic neurosurgery on small animals.

This work presents the development and performance analysis of a robotic system for stereotactic neurosurgery on small animals. The system is dedicated to the precise placement of probes in the small animal brain, thus providing an improved framework for brain research.

Towards Multi-Directional OCT for Speckle Noise Reduction

Multi-directional optical coherence tomography (MD-OCT) applies and extends the concept of angular compounding for speckle noise reduction to the area of OCT imaging. OCT images are acquired from a wide range of angles of view. Averaging of the rotated images therefore requires compensation of the parallax which is achieved by simple image registration for image reconstruction. Test measurements of a sample structure in a low and highly scattering environment show that the method improves the signal-to-noise ratio by a factor of 4 and hence reduces speckle noise significantly. Experimental results also show that the proposed averaging increases the performance of common edge-detection algorithms.

Spherical assistant for stereotactic surgery

This contribution reports the development of a novel robotic manipulator for stereotactic surgery on small animals, the spherical assistant for stereotactic surgery (SAS- SU). A kinematic design is deduced based on the surgical task requirements. Forward and inverse kinematics are derived analytically. As the system is required to position medical probes of varying size and shape, details on the calibration for different probe configurations are provided. The kinematic design of the novel manipulator is compared to an existing stereotactic instrument in terms of kinematic accuracy. Results show that the SASSU systems is less sensitive to translational positioning errors induced by changes in the joint variables.

Towards Automated OCT-based Identification of White Brain Matter

A novel model-based identification of white brain matter in OCT A-scans is proposed. Based on nonlinear energy operators used in the classification of neural activity, candidates for white matter structures are extracted from a baseline-corrected signal. Validation of candidates is done by evaluating the correspondence to a simplified intensity model which is parametrized beforehand. Results for identification of white matter in rat brain in vitro show the capability of the proposed algorithm.

Automated segmentation of tissue structures in optical coherence tomography data

Segmentation of optical coherence tomography (OCT) images provides useful information, especially in medical imaging applications. Because OCT images are subject to speckle noise, the identification of structures is complicated. Addressing this issue, two methods for the automated segmentation of arbitrary structures in OCT images are proposed. The methods perform a seeded region growing, applying a model-based analysis of OCT A-scans for the seeds acquisition. The segmentation therefore avoids any user-intervention dependency. The first region-growing algorithm uses an adaptive neighborhood homogeneity criterion based on a model of an OCT intensity course in tissue and a model of speckle noise corruption. It can be applied to an unfiltered OCT image. The second performs region growing on a filtered OCT image applying the local median as a measure for homogeneity in the region. Performance is compared through the quantitative evaluation of artificial data, showing the capabilities of both in terms of structures detected and leakage. The proposed methods were tested on real OCT data in different scenarios and showed promising results for their application in OCT imaging.

Computer- and robot-assisted stereotaxy for high-precision small animal brain exploration Computer- und robotergestutzte Stereotaxie fur hochprazise Exploration des Kleintierhirns

Abstract This contribution introduces a computer- and robot-assisted framework for stereotactic neurosurgery on small animals. Two major elements of this framework are presented in detail: a robotic stereotactic assistant and the software framework for placement of probes into the brain. The latter integrates modules for registration, insertion control, and preoperative path planning. Two options for path planning are addressed: (a) atlas-based planning and (b) image-based planning based on computed tomography data. The framework is tested performing robot-assisted insertion of microelectrodes and acquisition of electrophysiological recordings in vivo. Concepts for data analysis pointing towards a mapping of position and neural structure to functional data are introduced. Results show that the presented framework allows precise small animal stereotaxy and therefore offers new options for brain research. Zusammenfassung Dieser Beitrag stellt eine computer- und robotergestützte Umgebung für stereotaktische Eingriffe am Kleintier vor. Zwei Bestandteile der Umgebung werden im Detail vorgestellt: ein robotischer Assistent und die Softwareumgebung, um Instrumente im Kleintierhirn einzubringen. Letztere integriert dabei Module zur Registrierung, Steuerung des stereotaktischen Assistenten und zur Planung des Eingriffs. Zwei Optionen für die Planung werden vorgestellt: (a) atlasbasierte und (b) bildbasierte Planung auf Basis von Computertomographiedaten. Die Umgebung wird anhand der Einbringung von Mikroelektroden und der Akquise von elektrophysiologischen Ableitungen in vivo getestet. Konzepte zur Evaluierung der Daten im Hinblick auf eine Zuordnung einer räumlichen Position und einer neuronalen Struktur zu funktionellen Daten werden vorgestellt. Die Ergebnisse zeigen, dass die vorgestellte Umgebung präzise stereotaktische Eingriffe am Kleintier und neue Optionen in der Hirnforschung ermöglicht.

Mathematical identification of a neuronal network consisting of GABA and DA in striatal slices of the rat brain

High frequency stimulation (HFS) has been used to treat various neurological and psychiatric diseases. Although further disorders are under investigation to extend the clinical application of HFS, the complex effect of HFS within a neuronal network is still unknown. Thus, it would be desirable to find a theoretical model that allows an estimation of the expected effect of applied HFS. Based on the neurochemical analysis of effects of the g-aminobutyric acid (GABA)A receptor antagonist bicuculline, the D2-like receptor antagonist sulpiride and the D1-like receptor antagonist SCH-23390 on HFS evoked GABA and dopamine (DA) release from striatal slices of the rat brain, a mathematical network model is proposed including the neurotransmitters GABA, DA and glutamate (GLU). The model reflects inhibitory and excitatory interactions of the neurotransmitters outflow in the presence of HFS. Under the assumption of linear interactions and static measurements, the model is expressed analytically. Numerical identification of inhibition and excitation is performed on a basis of real outflow levels of GABA and DA in the rat striatum. Results validate the nature of the proposed model. Therefore, this leads to an analytical model of the interactions within distinct neural network components of the rat striatum.

Local linear neural networks based on principal component analysis

A new method for the estimation of process parameters based on the principal component analysis is developed. The estimator yields optimal estimation results in the case of errors in variables (EIV) problems which are characterized by corrupted measurements of input and output signals. As the residual generation in fault detection methods often feature EIV characteristics, the estimator can be used to identify linear models for residual calculation. To overcome the limitations on linear models, the developed estimator is integrated into the LOLIMOT approach which is able to identify nonlinear processes. The estimator is used as an alternative to the standard Least Squares estimator to identify the parameters of the local linear models. Comparative results show the better suitability of the developed estimator for the residual generation in EIV-setups

Model-Based Detection of White Matter in Optical Coherence Tomography Data

A method for white matter detection in Optical Coherence Tomography A-Scans is presented. The Kalman filter is used to obtain a slope change estimate of the intensity signal. The estimate is subsequently analyzed by a spike detection algorithm and then evaluated by a neural network binary classifier (Perceptron). The capability of the proposed method is shown through the quantitative evaluation of simulated A-Scans. The method was also applied to data obtained from a rats brain in vitro. Results show that the developed algorithm identifies less false positives than other two spike detection methods, thus, enhancing the robustness and quality of detection.