Thorsten Falk

Bio: Thorsten Falk is an academic researcher from University of Freiburg. The author has contributed to research in topics: Object detection & Segmentation. The author has an hindex of 6, co-authored 9 publications receiving 1093 citations.

U-Net: deep learning for cell counting, detection, and morphometry

Abstract: U-Net is a generic deep-learning solution for frequently occurring quantification tasks such as cell detection and shape measurements in biomedical image data. We present an ImageJ plugin that enables non-machine-learning experts to analyze their data with U-Net on either a local computer or a remote server/cloud service. The plugin comes with pretrained models for single-cell segmentation and allows for U-Net to be adapted to new tasks on the basis of a few annotated samples.

Single-cell profiling identifies myeloid cell subsets with distinct fates during neuroinflammation

Abstract: The innate immune cell compartment is highly diverse in the healthy central nervous system (CNS), including parenchymal and non-parenchymal macrophages. However, this complexity is increased in inflammatory settings by the recruitment of circulating myeloid cells. It is unclear which disease-specific myeloid subsets exist and what their transcriptional profiles and dynamics during CNS pathology are. Combining deep single-cell transcriptome analysis, fate mapping, in vivo imaging, clonal analysis, and transgenic mouse lines, we comprehensively characterized unappreciated myeloid subsets in several CNS compartments during neuroinflammation. During inflammation, CNS macrophage subsets undergo self-renewal, and random proliferation shifts toward clonal expansion. Last, functional studies demonstrated that endogenous CNS tissue macrophages are redundant for antigen presentation. Our results highlight myeloid cell diversity and provide insights into the brain's innate immune system.

Author Correction: U-Net: deep learning for cell counting, detection, and morphometry.

Abstract: In the version of this paper originally published, one of the affiliations for Dominic Mai was incorrect: "Center for Biological Systems Analysis (ZBSA), Albert-Ludwigs-University, Freiburg, Germany" should have been "Life Imaging Center, Center for Biological Systems Analysis, Albert-Ludwigs-University, Freiburg, Germany." This change required some renumbering of subsequent author affiliations. These corrections have been made in the PDF and HTML versions of the article, as well as in any cover sheets for associated Supplementary Information.

Flavonol-mediated stabilization of PIN efflux complexes regulates polar auxin transport.

Abstract: The transport of auxin controls the rate, direction and localization of plant growth and development. The course of auxin transport is defined by the polar subcellular localization of the PIN proteins, a family of auxin efflux transporters. However, little is known about the composition and regulation of the PIN protein complex. Here, using blue-native PAGE and quantitative mass spectrometry, we identify native PIN core transport units as homo- and heteromers assembled from PIN1, PIN2, PIN3, PIN4 and PIN7 subunits only. Furthermore, we show that endogenous flavonols stabilize PIN dimers to regulate auxin efflux in the same way as does the auxin transport inhibitor 1-naphthylphthalamic acid (NPA). This inhibitory mechanism is counteracted both by the natural auxin indole-3-acetic acid and by phosphomimetic amino acids introduced into the PIN1 cytoplasmic domain. Our results lend mechanistic insights into an endogenous control mechanism which regulates PIN function and opens the way for a deeper understanding of the protein environment and regulation of the polar auxin transport complex.

A 3D digital atlas of the Nicotiana tabacum root tip and its use to investigate changes in the root apical meristem induced by the Agrobacterium 6b oncogene.

Abstract: Summary Using the intrinsic Root Coordinate System (iRoCS) Toolbox, a digital atlas at cellular resolution has been constructed for Nicotiana tabacum roots. Mitotic cells and cells labeled for DNA replication with 5-ethynyl-2’-deoxyuridine (EdU) were mapped. The results demonstrate that iRoCS analysis can be applied to thicker roots than those of Arabidopsis thaliana without histological sectioning. A three-dimensional (3-D) analysis of the root tip showed that tobacco roots undergo several irregular periclinal and tangential divisions. Irrespective of cell type, rapid cell elongation starts at the same distance from the quiescent center, however, boundaries between cell proliferation and transition domains are cell-type specific. The data support the existence of a transition domain in tobacco roots. Cell endoreduplication starts in the transition domain and continues into the elongation zone. The tobacco root map was subsequently used to analyze root organization changes caused by inducible expression of the Agrobacterium 6b oncogene. In tobacco roots expressing the 6b gene, the root apical meristem was shorter and radial cell growth was reduced, but the mitotic and DNA replication indexes were not affected. The epidermis of 6b-expressing roots produced less files and underwent abnormal periclinal divisions. The periclinal division leading to mature endodermis and cortex3 cell files was delayed. These findings define additional targets for future studies on the mode of action of the Agrobacterium 6b oncogene. This article is protected by copyright. All rights reserved.

nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation

Abstract: Biomedical imaging is a driver of scientific discovery and a core component of medical care and is being stimulated by the field of deep learning. While semantic segmentation algorithms enable image analysis and quantification in many applications, the design of respective specialized solutions is non-trivial and highly dependent on dataset properties and hardware conditions. We developed nnU-Net, a deep learning-based segmentation method that automatically configures itself, including preprocessing, network architecture, training and post-processing for any new task. The key design choices in this process are modeled as a set of fixed parameters, interdependent rules and empirical decisions. Without manual intervention, nnU-Net surpasses most existing approaches, including highly specialized solutions on 23 public datasets used in international biomedical segmentation competitions. We make nnU-Net publicly available as an out-of-the-box tool, rendering state-of-the-art segmentation accessible to a broad audience by requiring neither expert knowledge nor computing resources beyond standard network training.

UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation

Abstract: The state-of-the-art models for medical image segmentation are variants of U-Net and fully convolutional networks (FCN). Despite their success, these models have two limitations: (1) their optimal depth is apriori unknown, requiring extensive architecture search or inefficient ensemble of models of varying depths; and (2) their skip connections impose an unnecessarily restrictive fusion scheme, forcing aggregation only at the same-scale feature maps of the encoder and decoder sub-networks. To overcome these two limitations, we propose UNet++, a new neural architecture for semantic and instance segmentation, by (1) alleviating the unknown network depth with an efficient ensemble of U-Nets of varying depths, which partially share an encoder and co-learn simultaneously using deep supervision; (2) redesigning skip connections to aggregate features of varying semantic scales at the decoder sub-networks, leading to a highly flexible feature fusion scheme; and (3) devising a pruning scheme to accelerate the inference speed of UNet++. We have evaluated UNet++ using six different medical image segmentation datasets, covering multiple imaging modalities such as computed tomography (CT), magnetic resonance imaging (MRI), and electron microscopy (EM), and demonstrating that (1) UNet++ consistently outperforms the baseline models for the task of semantic segmentation across different datasets and backbone architectures; (2) UNet++ enhances segmentation quality of varying-size objects—an improvement over the fixed-depth U-Net; (3) Mask RCNN++ (Mask R-CNN with UNet++ design) outperforms the original Mask R-CNN for the task of instance segmentation; and (4) pruned UNet++ models achieve significant speedup while showing only modest performance degradation. Our implementation and pre-trained models are available at https://github.com/MrGiovanni/UNetPlusPlus .

U-Net: deep learning for cell counting, detection, and morphometry

Abstract: U-Net is a generic deep-learning solution for frequently occurring quantification tasks such as cell detection and shape measurements in biomedical image data. We present an ImageJ plugin that enables non-machine-learning experts to analyze their data with U-Net on either a local computer or a remote server/cloud service. The plugin comes with pretrained models for single-cell segmentation and allows for U-Net to be adapted to new tasks on the basis of a few annotated samples.

Spatial and temporal heterogeneity of mouse and human microglia at single-cell resolution.

Abstract: Microglia have critical roles not only in neural development and homeostasis, but also in neurodegenerative and neuroinflammatory diseases of the central nervous system1-4. These highly diverse and specialized functions may be executed by subsets of microglia that already exist in situ, or by specific subsets of microglia that develop from a homogeneous pool of cells on demand. However, little is known about the presence of spatially and temporally restricted subclasses of microglia in the central nervous system during development or disease. Here we combine massively parallel single-cell analysis, single-molecule fluorescence in situ hybridization, advanced immunohistochemistry and computational modelling to comprehensively characterize subclasses of microglia in multiple regions of the central nervous system during development and disease. Single-cell analysis of tissues of the central nervous system during homeostasis in mice revealed specific time- and region-dependent subtypes of microglia. Demyelinating and neurodegenerative diseases evoked context-dependent subtypes of microglia with distinct molecular hallmarks and diverse cellular kinetics. Corresponding clusters of microglia were also identified in healthy human brains, and the brains of patients with multiple sclerosis. Our data provide insights into the endogenous immune system of the central nervous system during development, homeostasis and disease, and may also provide new targets for the treatment of neurodegenerative and neuroinflammatory pathologies.