Olaf Ronneberger

U-Net: Convolutional Networks for Biomedical Image Segmentation

There is large consent that successful training of deep networks requires many thousand annotated training samples. In this paper, we present a network and training strategy that relies on the strong use of data augmentation to use the available annotated samples more efficiently. The architecture consists of a contracting path to capture context and a symmetric expanding path that enables precise localization. We show that such a network can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks. Using the same network trained on transmitted light microscopy images (phase contrast and DIC) we won the ISBI cell tracking challenge 2015 in these categories by a large margin. Moreover, the network is fast. Segmentation of a 512x512 image takes less than a second on a recent GPU. The full implementation (based on Caffe) and the trained networks are available at http://lmb.informatik.uni-freiburg.de/people/ronneber/u-net .

Chemotaxonomic Identification of Single Bacteria by Micro-Raman Spectroscopy: Application to Clean-Room-Relevant Biological Contaminations

ABSTRACT Microorganisms, such as bacteria, which might be present as contamination inside an industrial food or pharmaceutical clean room process need to be identified on short time scales in order to minimize possible health hazards as well as production downtimes causing financial deficits. Here we describe the first results of single-particle micro-Raman measurements in combination with a classification method, the so-called support vector machine technique, allowing for a fast, reliable, and nondestructive online identification method for single bacteria.

3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation

This paper introduces a network for volumetric segmentation that learns from sparsely annotated volumetric images. We outline two attractive use cases of this method: (1) In a semi-automated setup, the user annotates some slices in the volume to be segmented. The network learns from these sparse annotations and provides a dense 3D segmentation. (2) In a fully-automated setup, we assume that a representative, sparsely annotated training set exists. Trained on this data set, the network densely segments new volumetric images. The proposed network extends the previous u-net architecture from Ronneberger et al. by replacing all 2D operations with their 3D counterparts. The implementation performs on-the-fly elastic deformations for efficient data augmentation during training. It is trained end-to-end from scratch, i.e., no pre-trained network is required. We test the performance of the proposed method on a complex, highly variable 3D structure, the Xenopus kidney, and achieve good results for both use cases.

Micro-Raman spectroscopic identification of bacterial cells of the genus Staphylococcus and dependence on their cultivation conditions

Microbial contamination is not only a medical problem, but also plays a large role in pharmaceutical clean room production and food processing technology. Therefore many techniques were developed to achieve differentiation and identification of microorganisms. Among these methods vibrational spectroscopic techniques (IR, Raman and SERS) are useful tools because of their rapidity and sensitivity. Recently we have shown that micro-Raman spectroscopy in combination with a support vector machine is an extremely capable approach for a fast and reliable, non-destructive online identification of single bacteria belonging to different genera. In order to simulate different environmental conditions we analyzed in this contribution different Staphylococcus strains with varying cultivation conditions in order to evaluate our method with a reliable dataset. First, micro-Raman spectra of the bulk material and single bacterial cells that were grown under the same conditions were recorded and used separately for a distinct chemotaxonomic classification of the strains. Furthermore Raman spectra were recorded from single bacterial cells that were cultured under various conditions to study the influence of cultivation on the discrimination ability. This dataset was analyzed both with a hierarchical cluster analysis (HCA) and a support vector machine (SVM).

Vertebrate kidney tubules elongate using a planar cell polarity-dependent, rosette-based mechanism of convergent extension

Cystic kidney diseases are a global public health burden, affecting over 12 million people. Although much is known about the genetics of kidney development and disease, the cellular mechanisms driving normal kidney tubule elongation remain unclear. Here, we used in vivo imaging to show for the first time that mediolaterally oriented cell intercalation is fundamental to vertebrate kidney morphogenesis. Unexpectedly, we found that kidney tubule elongation is driven in large part by a myosin-dependent, multicellular rosette–based mechanism, previously only described in Drosophila melanogaster. In contrast to findings in Drosophila, however, non-canonical Wnt and planar cell polarity (PCP) signaling is required to control rosette topology and orientation during vertebrate kidney tubule elongation. These data resolve long-standing questions concerning the role of PCP signaling in the developing kidney and, moreover, establish rosette-based intercalation as a deeply conserved cellular engine for epithelial morphogenesis.

Standardized evaluation of algorithms for computer-aided diagnosis of dementia based on structural MRI: The CADDementia challenge

Algorithms for computer-aided diagnosis of dementia based on structural MRI have demonstrated high performance in the literature, but are difficult to compare as different data sets and methodology were used for evaluation. In addition, it is unclear how the algorithms would perform on previously unseen data, and thus, how they would perform in clinical practice when there is no real opportunity to adapt the algorithm to the data at hand. To address these comparability, generalizability and clinical applicability issues, we organized a grand challenge that aimed to objectively compare algorithms based on a clinically representative multi-center data set. Using clinical practice as the starting point, the goal was to reproduce the clinical diagnosis. Therefore, we evaluated algorithms for multi-class classification of three diagnostic groups: patients with probable Alzheimers disease, patients with mild cognitive impairment and healthy controls. The diagnosis based on clinical criteria was used as reference standard, as it was the best available reference despite its known limitations. For evaluation, a previously unseen test set was used consisting of 354 T1-weighted MRI scans with the diagnoses blinded. Fifteen research teams participated with a total of 29 algorithms. The algorithms were trained on a small training set (n=30) and optionally on data from other sources (e.g., the Alzheimers Disease Neuroimaging Initiative, the Australian Imaging Biomarkers and Lifestyle flagship study of aging). The best performing algorithm yielded an accuracy of 63.0% and an area under the receiver-operating-characteristic curve (AUC) of 78.8%. In general, the best performances were achieved using feature extraction based on voxel-based morphometry or a combination of features that included volume, cortical thickness, shape and intensity. The challenge is open for new submissions via the web-based framework: http://caddementia.grand-challenge.org.

XuvTools: free, fast and reliable stitching of large 3D datasets

Current biomedical research increasingly requires imaging large and thick 3D structures at high resolution. Prominent examples are the tracking of fine filaments over long distances in brain slices, or the localization of gene expression or cell migration in whole animals like Caenorhabditis elegans or zebrafish. To obtain both high resolution and a large field of view (FOV), a combination of multiple recordings (‘tiles’) is one of the options. Although hardware solutions exist for fast and reproducible acquisition of multiple 3D tiles, generic software solutions are missing to assemble (‘stitch’) these tiles quickly and accurately.

Early functional magnetic resonance imaging activations predict language outcome after stroke

An accurate prediction of system-specific recovery after stroke is essential to provide rehabilitation therapy based on the individual needs. We explored the usefulness of functional magnetic resonance imaging scans from an auditory language comprehension experiment to predict individual language recovery in 21 aphasic stroke patients. Subjects with an at least moderate language impairment received extensive language testing 2 weeks and 6 months after left-hemispheric stroke. A multivariate machine learning technique was used to predict language outcome 6 months after stroke. In addition, we aimed to predict the degree of language improvement over 6 months. 76% of patients were correctly separated into those with good and bad language performance 6 months after stroke when based on functional magnetic resonance imaging data from language relevant areas. Accuracy further improved (86% correct assignments) when age and language score were entered alongside functional magnetic resonance imaging data into the fully automatic classifier. A similar accuracy was reached when predicting the degree of language improvement based on imaging, age and language performance. No prediction better than chance level was achieved when exploring the usefulness of diffusion weighted imaging as well as functional magnetic resonance imaging acquired two days after stroke. This study demonstrates the high potential of current machine learning techniques to predict system-specific clinical outcome even for a disease as heterogeneous as stroke. Best prediction of language recovery is achieved when the brain activation potential after system-specific stimulation is assessed in the second week post stroke. More intensive early rehabilitation could be provided for those with a predicted poor recovery and the extension to other systems, for example, motor and attention seems feasible.

Rotation-Invariant HOG Descriptors Using Fourier Analysis in Polar and Spherical Coordinates

The histogram of oriented gradients (HOG) is widely used for image description and proves to be very effective. In many vision problems, rotation-invariant analysis is necessary or preferred. Popular solutions are mainly based on pose normalization or learning, neglecting some intrinsic properties of rotations. This paper presents a method to build rotation-invariant HOG descriptors using Fourier analysis in polar/spherical coordinates, which are closely related to the irreducible representation of the 2D/3D rotation groups. This is achieved by considering a gradient histogram as a continuous angular signal which can be well represented by the Fourier basis (2D) or spherical harmonics (3D). As rotation-invariance is established in an analytical way, we can avoid discretization artifacts and create a continuous mapping from the image to the feature space. In the experiments, we first show that our method outperforms the state-of-the-art in a public dataset for a car detection task in aerial images. We further use the Princeton Shape Benchmark and the SHREC 2009 Generic Shape Benchmark to demonstrate the high performance of our method for similarity measures of 3D shapes. Finally, we show an application on microscopic volumetric data.

ViBE-Z: a framework for 3D virtual colocalization analysis in zebrafish larval brains.

Precise three-dimensional (3D) mapping of a large number of gene expression patterns, neuronal types and connections to an anatomical reference helps us to understand the vertebrate brain and its development. We developed the Virtual Brain Explorer (ViBE-Z), a software that automatically maps gene expression data with cellular resolution to a 3D standard larval zebrafish (Danio rerio) brain. ViBE-Z enhances the data quality through fusion and attenuation correction of multiple confocal microscope stacks per specimen and uses a fluorescent stain of cell nuclei for image registration. It automatically detects 14 predefined anatomical landmarks for aligning new data with the reference brain. ViBE-Z performs colocalization analysis in expression databases for anatomical domains or subdomains defined by any specific pattern; here we demonstrate its utility for mapping neurons of the dopaminergic system. The ViBE-Z database, atlas and software are provided via a web interface.