Florian Schulze

Automated landmarking and labeling of fully and partially scanned spinal columns in CT images

The spinal column is one of the most distinguishable structures in CT scans of the superior part of the human body. It is not necessary to segment the spinal column in order to use it as a frame of reference. It is sufficient to place landmarks and the appropriate anatomical labels at intervertebral disks and vertebrae. In this paper, we present an automated system for landmarking and labeling spinal columns in 3D CT datasets. We designed this framework with two goals in mind. First, we relaxed input data requirements found in the literature, and we label both full and partial spine scans. Secondly, we intended to fulfill the performance requirement for daily clinical use and developed a high throughput system capable of processing thousands of slices in just a few minutes. To accomplish the aforementioned goals, we encoded structural knowledge from training data in probabilistic boosting trees and used it to detect efficiently the spinal canal, intervertebral disks, and three reference regions responsible for initializing the landmarking and labeling. Final landmarks and labels are selected by Markov Random Field-based matches of newly introduced 3-disk models. The framework has been tested on 36 CT images having at least one of the regions around the thoracic first ribs, the thoracic twelfth ribs, or the sacrum. In an average time of 2 min, we achieved a correct labeling in 35 cases with precision of 99.0% and recall of 97.2%. Additionally, we present results assuming none of the three reference regions could be detected.

Automatic Segmentation of Drosophila Neural Compartments Using GAL4 Expression Data Reveals Novel Visual Pathways

Summary Identifying distinct anatomical structures within the brain and developing genetic tools to target them are fundamental steps for understanding brain function. We hypothesize that enhancer expression patterns can be used to automatically identify functional units such as neuropils and fiber tracts. We used two recent, genome-scale Drosophila GAL4 libraries and associated confocal image datasets to segment large brain regions into smaller subvolumes. Our results (available at https://strawlab.org/braincode) support this hypothesis because regions with well-known anatomy, namely the antennal lobes and central complex, were automatically segmented into familiar compartments. The basis for the structural assignment is clustering of voxels based on patterns of enhancer expression. These initial clusters are agglomerated to make hierarchical predictions of structure. We applied the algorithm to central brain regions receiving input from the optic lobes. Based on the automated segmentation and manual validation, we can identify and provide promising driver lines for 11 previously identified and 14 novel types of visual projection neurons and their associated optic glomeruli. The same strategy can be used in other brain regions and likely other species, including vertebrates.

neuroMAP — Interactive graph-visualization of the fruit fly's neural circuit

Neuroscientists study the function of neural circuits in the brain of the common fruit fly Drosophila melanogaster to discover how complex behavior is generated. To establish models of neural information processing, knowledge about potential connections between individual neurons is required. Connections can occur when the arborizations of two neurons overlap. Judging connectivity by analyzing overlaps using traditional volumetric visualization is difficult since the examined objects occlude each other. A more abstract form of representation is therefore desirable. In collaboration with a group of neuroscientists, we designed and implemented neuroMap, an interactive two-dimensional graph that renders the brain and its interconnections in the form of a circuit-style wiring diagram. neuroMap provides a clearly structured overview of all possible connections between neurons and offers means for interactive exploration of the underlying neuronal database. In this paper, we discuss the design decisions that formed neuroMap and evaluate its application in discussions with the scientists.

Structure-Based Neuron Retrieval Across Drosophila Brains

Comparing local neural structures across large sets of examples is crucial when studying gene functions, and their effect in the Drosophila brain. The current practice of aligning brain volume data to a joint reference frame is based on the neuropil. However, even after alignment neurons exhibit residual location and shape variability that, together with image noise, hamper direct quantitative comparison and retrieval of similar structures on an intensity basis. In this paper, we propose and evaluate an image-based retrieval method for neurons, relying on local appearance, which can cope with spatial variability across the population. For an object of interest marked in a query case, the method ranks cases drawn from a large data set based on local neuron appearance in confocal microscopy data. The approach is based on capturing the orientation of neurons based on structure tensors and expanding this field via Gradient Vector Flow. During retrieval, the algorithm compares fields across cases, and calculates a corresponding ranking of most similar cases with regard to the local structure of interest. Experimental results demonstrate that the similarity measure and ranking mechanisms yield high precision and recall in realistic search scenarios.

3D object retrieval in an atlas of neuronal structures

Circuit neuroscience tries to solve one of the most challenging questions in biology: How does the brain work? An important step toward an answer to this question is to gather detailed knowledge about the neuronal circuits of the model organism Drosophila melanogaster. Geometric representations of neuronal objects of the Drosophila are acquired using molecular genetic methods, confocal microscopy, nonrigid registration and segmentation. These objects are integrated into a constantly growing common atlas. The comparison of new segmented neuronal objects to already known neuronal structures is a frequent task, which evolves with a growing amount of data into a bottleneck of the knowledge discovery process. Thus, the exploration of the atlas by means of domain specific similarity measures becomes a pressing need. To enable similarity based retrieval of neuronal objects, we defined together with domain experts tailored dissimilarity measures for each of the three typical neuronal structures cell body, projection, and arborization. Moreover, we defined the neuron enhanced similarity for projections and arborizations. According to domain experts, the developed system has big advantages for all tasks, which involve extensive data exploration.

Similarity based object retrieval of composite neuronal structures

Circuit Neuroscience tries to solve one of the most challenging questions in biology: How does the brain work? An important step towards an answer to this question is to gather detailed knowledge about the neuronal circuits of the model organism Drosophila melanogaster. Geometric representations of neuronal objects of the Drosophila are acquired using molecular genetic methods, confocal microscopy, non-rigid registration and segmentation. These objects are integrated into a constantly growing common atlas. The comparison of new segmented neurons to already known neurons is a frequent task which evolves with a growing amount of data into a bottleneck of the knowledge discovery process. Thus, the exploration of the atlas by means of domain specific similarity measures becomes a pressing need. To enable similarity based retrieval of neuronal objects we defined together with domain experts tailored dissimilarity measures for each of the three typical neuronal sub structures cell body, projection, arborization. The dissimilarity measure for composite neurons has been defined as domain specific combination of the sub structure dissimilarities. According to domain experts the developed system has big advantages for all tasks which involve extensive data exploration.

A Statistically Representative Atlas for Mapping Neuronal Circuits in the Drosophila Adult Brain

Imaging the expression patterns of reporter constructs is a powerful tool to dissect the neuronal circuits of perception and behavior in the adult brain of Drosophila, one of the major models for studying brain functions. To date, several Drosophila brain templates and digital atlases have been built to automatically analyze and compare collections of expression pattern images. However, there has been no systematic comparison of performances between alternative atlasing strategies and registration algorithms. Here, we objectively evaluated the performance of different strategies for building adult Drosophila brain templates and atlases. In addition, we used state-of-the-art registration algorithms to generate a new group-wise inter-sex atlas. Our results highlight the benefit of statistical atlases over individual ones and show that the newly proposed inter-sex atlas outperformed existing solutions for automated registration and annotation of expression patterns. Over 3,000 images from the Janelia Farm FlyLight collection were registered using the proposed strategy. These registered expression patterns can be searched and compared with a new version of the BrainBaseWeb system and BrainGazer software. We illustrate the validity of our methodology and brain atlas with registration-based predictions of expression patterns in a subset of clock neurons. The described registration framework should benefit to brain studies in Drosophila and other insect species.

Adaptive and Background-Aware GAL4 Expression Enhancement of Co-registered Confocal Microscopy Images

GAL4 gene expression imaging using confocal microscopy is a common and powerful technique used to study the nervous system of a model organism such as Drosophila melanogaster. Recent research projects focused on high throughput screenings of thousands of different driver lines, resulting in large image databases. The amount of data generated makes manual assessment tedious or even impossible. The first and most important step in any automatic image processing and data extraction pipeline is to enhance areas with relevant signal. However, data acquired via high throughput imaging tends to be less then ideal for this task, often showing high amounts of background signal. Furthermore, neuronal structures and in particular thin and elongated projections with a weak staining signal are easily lost. In this paper we present a method for enhancing the relevant signal by utilizing a Hessian-based filter to augment thin and weak tube-like structures in the image. To get optimal results, we present a novel adaptive background-aware enhancement filter parametrized with the local background intensity, which is estimated based on a common background model. We also integrate recent research on adaptive image enhancement into our approach, allowing us to propose an effective solution for known problems present in confocal microscopy images. We provide an evaluation based on annotated image data and compare our results against current state-of-the-art algorithms. The results show that our algorithm clearly outperforms the existing solutions.

Erratum to “Automated landmarking and labeling of fully and partially scanned spinal columns in CT images” [Med. Image Anal. 17 (2013) 1151–1163]

1361-8415/

Direct Volume Deformation

see front matter 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.media.2014.01.008 DOI of original article: http://dx.doi.org/10.1016/j.media.2013.07.005 E-mail address: david.major@vrvis.at (D. Major) Table 1 Overview of state-of-the-art approaches and our method. Data modality and requirement to contain a specific part of spine (‘‘C’’ – cervical, ‘‘T’’ – thorax, ‘‘L’’ – lumbar, ‘‘f spine scan and ‘‘T12’’ – thoracic twelfth vertebra/rib). Whether tested with pathology data or not. How many imaging vendors do the test data come from. Correct labe (correctly labeled test data/all test data and correctly labeled vertebra or disk/all tested vertebra or disk) and time performance in seconds. ‘‘–’’ means not available.