Search

Bioimaging refers to a collection of methods to visualize the internal structures and mechanisms of living organisms. The fundamental tool, the microscope, has enabled seminal discoveries like that of the cell as the smallest unit of life, and continues to expand our understanding of biological processes. Today, we can follow the interaction of single molecules within nanoseconds in a living cell, and the development of complete small organisms like fish and flies over several days starting from the fertilized egg. Each image pixel encodes multiple spatiotemporal and spectral dimensions, compounding the massive volume and complexity of bioimage data. Proper handling of this data is indispensable for analysis and its lack has become a growing hindrance for the many disciplines of the life and biomedical sciences relying on bioimaging. No single domain has the expertise to tackle this bottleneck alone. As a method-specific consortium, NFDI4BIOMAGE seeks to address these issues, enabling bioimaging data to be shared and re-used like they are acquired, i.e., independently of disciplinary boundaries. We will provide solutions for exploiting the full information content of bioimage data and enable new discoveries through sharing and re-analysis. Our RDM strategy is based on a robust needs analysis that derives not only from a community survey but also from over a decade of experience in German BioImaging, the German Society for Microscopy and Image Analysis. It considers the entire lifecycle of bioimaging data, from acquisition to archiving, including analysis and enabling re-use. A foundational element of this strategy is the definition of a common, cloud-compatible, and interoperable digital object that bundles binary images with their descriptive and provenance metadata. With members from plant biology to neuroscience, NFDI4BIOIMAGE will champion the standardization of bioimage data to create a framework that answers discipline-specific needs while ensuring communicati

Contains fulltext : 305368.pdf (Publisher’s version ) (Closed access), Increasing evidence shows that flaws in machine learning (ML) algorithm validation are an underestimated global problem. In biomedical image analysis, chosen performance metrics often do not reflect the domain interest, and thus fail to adequately measure scientific progress and hinder translation of ML techniques into practice. To overcome this, we created Metrics Reloaded, a comprehensive framework guiding researchers in the problem-aware selection of metrics. Developed by a large international consortium in a multistage Delphi process, it is based on the novel concept of a problem fingerprint-a structured representation of the given problem that captures all aspects that are relevant for metric selection, from the domain interest to the properties of the target structure(s), dataset and algorithm output. On the basis of the problem fingerprint, users are guided through the process of choosing and applying appropriate validation metrics while being made aware of potential pitfalls. Metrics Reloaded targets image analysis problems that can be interpreted as classification tasks at image, object or pixel level, namely image-level classification, object detection, semantic segmentation and instance segmentation tasks. To improve the user experience, we implemented the framework in the Metrics Reloaded online tool. Following the convergence of ML methodology across application domains, Metrics Reloaded fosters the convergence of validation methodology. Its applicability is demonstrated for various biomedical use cases.

Contains fulltext : 305370.pdf (Publisher’s version ) (Closed access), Validation metrics are key for tracking scientific progress and bridging the current chasm between artificial intelligence research and its translation into practice. However, increasing evidence shows that, particularly in image analysis, metrics are often chosen inadequately. Although taking into account the individual strengths, weaknesses and limitations of validation metrics is a critical prerequisite to making educated choices, the relevant knowledge is currently scattered and poorly accessible to individual researchers. Based on a multistage Delphi process conducted by a multidisciplinary expert consortium as well as extensive community feedback, the present work provides a reliable and comprehensive common point of access to information on pitfalls related to validation metrics in image analysis. Although focused on biomedical image analysis, the addressed pitfalls generalize across application domains and are categorized according to a newly created, domain-agnostic taxonomy. The work serves to enhance global comprehension of a key topic in image analysis validation.

State-of-the-art machine learning methods such as convolutional neural networks (CNNs) are frequently employed in computer vision. Despite their high performance on unseen data, CNNs are often criticized for lacking transparency — that is, providing very limited if any information about the internal decision-making process. In some applications, especially in healthcare, such transparency of algorithms is crucial for end users, as trust in diagnosis and prognosis is important not only for the satisfaction and potential adherence of patients, but also for their health. Explainable artificial intelligence (XAI) aims to open up this “black box,” often perceived as a cryptic and inconceivable algorithm, to increase understanding of the machines’ reasoning.XAI is an emerging field, and techniques for making machine learning explainable are becoming increasingly available. XAI for computer vision mainly focuses on image classification, whereas interpretability in other tasks remains challenging. Here, I examine explainability in computer vision beyond image classification, namely in semantic segmentation and 3D multitarget image regression. This thesis consists of five chapters. In Chapter 1 (Introduction), the background of artificial intelligence (AI), XAI, computer vision, and optics is presented, and the definitions of the terminology for XAI are proposed. Chapter 2 is focused on explaining the predictions of U-Net, a CNN commonly used for semantic image segmentation, and variations of this architecture. To this end, I propose the gradient-weighted class activation mapping for segmentation (Seg-Grad-CAM) method based on the well-known Grad-CAM method for explainable image classification. In Chapter 3, I present the application of deep learning to estimation of optical aberrations in microscopy biodata by identifying the present Zernike aberration modes and their amplitudes. A CNN-based approach PhaseNet can accurately estimate monochromatic aberrations in images of po

Embryo implantation into the uterus marks a key transition in mammalian development. In mice, implantation is mediated by the trophoblast and is accompanied by a morphological transition from the blastocyst to the egg cylinder. However, the roles of trophoblast‐uterine interactions in embryo morphogenesis during implantation are poorly understood due to inaccessibility in utero and the remaining challenges to recapitulate it ex vivo from the blastocyst. Here, we engineer a uterus‐like microenvironment to recapitulate peri‐implantation development of the whole mouse embryo ex vivo and reveal essential roles of the physical embryo‐uterine interaction. We demonstrate that adhesion between the trophoblast and the uterine matrix is required for in utero‐like transition of the blastocyst to the egg cylinder. Modeling the implanting embryo as a wetting droplet links embryo shape dynamics to the underlying changes in trophoblast adhesion and suggests that the adhesion‐mediated tension release facilitates egg cylinder formation. Light‐sheet live imaging and the experimental control of the engineered uterine geometry and trophoblast velocity uncovers the coordination between trophoblast motility and embryo growth, where the trophoblast delineates space for embryo morphogenesis.

Embryo implantation into the uterus marks a key transition in mammalian development. In mice, implantation is mediated by the trophoblast and is accompanied by a morphological transition from the blastocyst to the egg cylinder. However, the roles of trophoblast-uterine interactions in embryo morphogenesis during implantation are poorly understood due to inaccessibility in utero and the remaining challenges to recapitulate it ex vivo from the blastocyst. Here, we engineer a uterus-like microenvironment to recapitulate peri-implantation development of the whole mouse embryo ex vivo and reveal essential roles of the physical embryo-uterine interaction. We demonstrate that adhesion between the trophoblast and the uterine matrix is required for in utero-like transition of the blastocyst to the egg cylinder. Modeling the implanting embryo as a wetting droplet links embryo shape dynamics to the underlying changes in trophoblast adhesion and suggests that the adhesion-mediated tension release facilitates egg cylinder formation. Light-sheet live imaging and the experimental control of the engineered uterine geometry and trophoblast velocity uncovers the coordination between trophoblast motility and embryo growth, where the trophoblast delineates space for embryo morphogenesis.

State-of-the-art machine learning methods such as convolutional neural networks (CNNs) are frequently employed in computer vision. Despite their high performance on unseen data, CNNs are often criticized for lacking transparency — that is, providing very limited if any information about the internal decision-making process. In some applications, especially in healthcare, such transparency of algorithms is crucial for end users, as trust in diagnosis and prognosis is important not only for the satisfaction and potential adherence of patients, but also for their health. Explainable artificial intelligence (XAI) aims to open up this “black box,” often perceived as a cryptic and inconceivable algorithm, to increase understanding of the machines’ reasoning.XAI is an emerging field, and techniques for making machine learning explainable are becoming increasingly available. XAI for computer vision mainly focuses on image classification, whereas interpretability in other tasks remains challenging. Here, I examine explainability in computer vision beyond image classification, namely in semantic segmentation and 3D multitarget image regression. This thesis consists of five chapters. In Chapter 1 (Introduction), the background of artificial intelligence (AI), XAI, computer vision, and optics is presented, and the definitions of the terminology for XAI are proposed. Chapter 2 is focused on explaining the predictions of U-Net, a CNN commonly used for semantic image segmentation, and variations of this architecture. To this end, I propose the gradient-weighted class activation mapping for segmentation (Seg-Grad-CAM) method based on the well-known Grad-CAM method for explainable image classification. In Chapter 3, I present the application of deep learning to estimation of optical aberrations in microscopy biodata by identifying the present Zernike aberration modes and their amplitudes. A CNN-based approach PhaseNet can accurately estimate monochromatic aberrations in images of po

Validation metrics are key for the reliable tracking of scientific progress and for bridging the current chasm between artificial intelligence (AI) research and its translation into practice. However, increasing evidence shows that particularly in image analysis, metrics are often chosen inadequately in relation to the underlying research problem. This could be attributed to a lack of accessibility of metric-related knowledge: While taking into account the individual strengths, weaknesses, and limitations of validation metrics is a critical prerequisite to making educated choices, the relevant knowledge is currently scattered and poorly accessible to individual researchers. Based on a multi-stage Delphi process conducted by a multidisciplinary expert consortium as well as extensive community feedback, the present work provides the first reliable and comprehensive common point of access to information on pitfalls related to validation metrics in image analysis. Focusing on biomedical image analysis but with the potential of transfer to other fields, the addressed pitfalls generalize across application domains and are categorized according to a newly created, domain-agnostic taxonomy. To facilitate comprehension, illustrations and specific examples accompany each pitfall. As a structured body of information accessible to researchers of all levels of expertise, this work enhances global comprehension of a key topic in image analysis validation., Comment: Shared first authors: Annika Reinke and Minu D. Tizabi; shared senior authors: Lena Maier-Hein and Paul F. J\"ager. Published in Nature Methods. arXiv admin note: text overlap with arXiv:2206.01653

Upon implantation, mammalian embryos undergo major morphogenesis and key developmental processes such as body axis specification and gastrulation. However, limited accessibility obscures the study of these crucial processes. Here, we develop an ex vivo Matrigel-collagen-based culture to recapitulate mouse development from E4.5 to E6.0. Our system not only recapitulates embryonic growth, axis initiation, and overall 3D architecture in 49% of the cases, but its compatibility with light-sheet microscopy also enables the study of cellular dynamics through automatic cell segmentation. We find that, upon implantation, release of the increasing tension in the polar trophectoderm is necessary for its constriction and invagination. The resulting extra-embryonic ectoderm plays a key role in growth, morphogenesis, and patterning of the neighboring epiblast, which subsequently gives rise to all embryonic tissues. This 3D ex vivo system thus offers unprecedented access to peri-implantation development for in toto monitoring, measurement, and spatiotemporally controlled perturbation, revealing a mechano-chemical interplay between extra-embryonic and embryonic tissues.

Upon implantation, mammalian embryos undergo major morphogenesis and key developmental processes such as body axis specification and gastrulation. However, limited accessibility obscures the study of these crucial processes. Here, we develop an ex vivo Matrigel-collagen-based culture to recapitulate mouse development from E4.5 to E6.0. Our system not only recapitulates embryonic growth, axis initiation, and overall 3D architecture in 49% of the cases, but its compatibility with light-sheet microscopy also enables the study of cellular dynamics through automatic cell segmentation. We find that, upon implantation, release of the increasing tension in the polar trophectoderm is necessary for its constriction and invagination. The resulting extra-embryonic ectoderm plays a key role in growth, morphogenesis, and patterning of the neighboring epiblast, which subsequently gives rise to all embryonic tissues. This 3D ex vivo system thus offers unprecedented access to peri-implantation development for in toto monitoring, measurement, and spatiotemporally controlled perturbation, revealing a mechano-chemical interplay between extra-embryonic and embryonic tissues.

Increasing evidence shows that flaws in machine learning (ML) algorithm validation are an underestimated global problem. Particularly in automatic biomedical image analysis, chosen performance metrics often do not reflect the domain interest, thus failing to adequately measure scientific progress and hindering translation of ML techniques into practice. To overcome this, our large international expert consortium created Metrics Reloaded, a comprehensive framework guiding researchers in the problem-aware selection of metrics. Following the convergence of ML methodology across application domains, Metrics Reloaded fosters the convergence of validation methodology. The framework was developed in a multi-stage Delphi process and is based on the novel concept of a problem fingerprint - a structured representation of the given problem that captures all aspects that are relevant for metric selection, from the domain interest to the properties of the target structure(s), data set and algorithm output. Based on the problem fingerprint, users are guided through the process of choosing and applying appropriate validation metrics while being made aware of potential pitfalls. Metrics Reloaded targets image analysis problems that can be interpreted as a classification task at image, object or pixel level, namely image-level classification, object detection, semantic segmentation, and instance segmentation tasks. To improve the user experience, we implemented the framework in the Metrics Reloaded online tool, which also provides a point of access to explore weaknesses, strengths and specific recommendations for the most common validation metrics. The broad applicability of our framework across domains is demonstrated by an instantiation for various biological and medical image analysis use cases., Comment: Shared first authors: Lena Maier-Hein, Annika Reinke. arXiv admin note: substantial text overlap with arXiv:2104.05642 Published in Nature Methods

In this thesis I will use state-of-the-art (SOTA) image denoising methods to denoise electron microscopy (EM) data. Then, I will present NoiseVoid a deep learning based self-supervised image denoising approach which is trained on single noisy observations. Eventually, I approach the missing wedge problem in tomography and introduce a novel image encoding, based on the Fourier transform which I am using to predict missing Fourier coefficients directly in Fourier space with Fourier Image Transformer (FIT). In the next paragraphs I will summarize the individual contributions briefly. Electron microscopy is the go to method for high-resolution images in biological research. Modern scanning electron microscopy (SEM) setups are used to obtain neural connectivity maps, allowing us to identify individual synapses. However, slow scanning speeds are required to obtain SEM images of sufficient quality. In (Weigert et al. 2018) the authors show, for fluorescence microscopy, how pairs of low- and high-quality images can be obtained from biological samples and use them to train content-aware image restoration (CARE) networks. Once such a network is trained, it can be applied to noisy data to restore high quality images. With SEM-CARE I present how this approach can be directly applied to SEM data, allowing us to scan the samples faster, resulting in $40$- to $50$-fold imaging speedups for SEM imaging. In structural biology cryo transmission electron microscopy (cryo TEM) is used to resolve protein structures and describe molecular interactions. However, missing contrast agents as well as beam induced sample damage (Knapek and Dubochet 1980) prevent acquisition of high quality projection images. Hence, reconstructed tomograms suffer from low signal-to-noise ratio (SNR) and low contrast, which makes post-processing of such data difficult and often has to be done manually. To facilitate down stream analysis and manual data browsing of cryo tomograms I present cryoCARE a Noise2Noise

In this thesis I will use state-of-the-art (SOTA) image denoising methods to denoise electron microscopy (EM) data. Then, I will present NoiseVoid a deep learning based self-supervised image denoising approach which is trained on single noisy observations. Eventually, I approach the missing wedge problem in tomography and introduce a novel image encoding, based on the Fourier transform which I am using to predict missing Fourier coefficients directly in Fourier space with Fourier Image Transformer (FIT). In the next paragraphs I will summarize the individual contributions briefly. Electron microscopy is the go to method for high-resolution images in biological research. Modern scanning electron microscopy (SEM) setups are used to obtain neural connectivity maps, allowing us to identify individual synapses. However, slow scanning speeds are required to obtain SEM images of sufficient quality. In (Weigert et al. 2018) the authors show, for fluorescence microscopy, how pairs of low- and high-quality images can be obtained from biological samples and use them to train content-aware image restoration (CARE) networks. Once such a network is trained, it can be applied to noisy data to restore high quality images. With SEM-CARE I present how this approach can be directly applied to SEM data, allowing us to scan the samples faster, resulting in $40$- to $50$-fold imaging speedups for SEM imaging. In structural biology cryo transmission electron microscopy (cryo TEM) is used to resolve protein structures and describe molecular interactions. However, missing contrast agents as well as beam induced sample damage (Knapek and Dubochet 1980) prevent acquisition of high quality projection images. Hence, reconstructed tomograms suffer from low signal-to-noise ratio (SNR) and low contrast, which makes post-processing of such data difficult and often has to be done manually. To facilitate down stream analysis and manual data browsing of cryo tomograms I present cryoCARE a Noise2Noise

The evolutionary origin of metazoan cell types such as neurons and muscles is not known. Using whole-body single-cell RNA sequencing in a sponge, an animal without nervous system and musculature, we identified 18 distinct cell types. These include nitric oxide-sensitive contractile pinacocytes, amoeboid phagocytes, and secretory neuroid cells that reside in close contact with digestive choanocytes that express scaffolding and receptor proteins. Visualizing neuroid cells by correlative x-ray and electron microscopy revealed secretory vesicles and cellular projections enwrapping choanocyte microvilli and cilia. Our data show a communication system that is organized around sponge digestive chambers, using conserved modules that became incorporated into the pre- and postsynapse in the nervous systems of other animals.

In this thesis I will use state-of-the-art (SOTA) image denoising methods to denoise electron microscopy (EM) data. Then, I will present NoiseVoid a deep learning based self-supervised image denoising approach which is trained on single noisy observations. Eventually, I approach the missing wedge problem in tomography and introduce a novel image encoding, based on the Fourier transform which I am using to predict missing Fourier coefficients directly in Fourier space with Fourier Image Transformer (FIT). In the next paragraphs I will summarize the individual contributions briefly. Electron microscopy is the go to method for high-resolution images in biological research. Modern scanning electron microscopy (SEM) setups are used to obtain neural connectivity maps, allowing us to identify individual synapses. However, slow scanning speeds are required to obtain SEM images of sufficient quality. In (Weigert et al. 2018) the authors show, for fluorescence microscopy, how pairs of low- and high-quality images can be obtained from biological samples and use them to train content-aware image restoration (CARE) networks. Once such a network is trained, it can be applied to noisy data to restore high quality images. With SEM-CARE I present how this approach can be directly applied to SEM data, allowing us to scan the samples faster, resulting in $40$- to $50$-fold imaging speedups for SEM imaging. In structural biology cryo transmission electron microscopy (cryo TEM) is used to resolve protein structures and describe molecular interactions. However, missing contrast agents as well as beam induced sample damage (Knapek and Dubochet 1980) prevent acquisition of high quality projection images. Hence, reconstructed tomograms suffer from low signal-to-noise ratio (SNR) and low contrast, which makes post-processing of such data difficult and often has to be done manually. To facilitate down stream analysis and manual data browsing of cryo tomograms I present cryoCARE a Noise2Noise

In the version of this Comment initially published, an error appeared in the “Specificity” equation displayed in the middle-right panel of Fig. 2. Originally reading “ fp/fp+tn”, the equation has been corrected to read: “ tn/tn+fp”. The error has been corrected in the online version of the Article. *A list of authors and their affiliations appears online., SCOPUS: no.j, SCOPUS: er.j, info:eu-repo/semantics/published

Instance segmentation is an important computer vision problem which remains challenging despite impressive recent advances due to deep learning-based methods. Given sufficient training data, fully supervised methods can yield excellent performance, but annotation of ground-truth data remains a major bottleneck, especially for biomedical applications where it has to be performed by domain experts. The amount of labels required can be drastically reduced by using rules derived from prior knowledge to guide the segmentation. However, these rules are in general not differentiable and thus cannot be used with existing methods. Here, we relax this requirement by using stateless actor critic reinforcement learning, which enables non-differentiable rewards. We formulate the instance segmentation problem as graph partitioning and the actor critic predicts the edge weights driven by the rewards, which are based on the conformity of segmented instances to high-level priors on object shape, position or size. The experiments on toy and real datasets demonstrate that we can achieve excellent performance without any direct supervision based only on a rich set of priors.

While the importance of automatic image analysis is continuously increasing, recent meta-research revealed major flaws with respect to algorithm validation. Performance metrics are particularly key for meaningful, objective, and transparent performance assessment and validation of the used automatic algorithms, but relatively little attention has been given to the practical pitfalls when using specific metrics for a given image analysis task. These are typically related to (1) the disregard of inherent metric properties, such as the behaviour in the presence of class imbalance or small target structures, (2) the disregard of inherent data set properties, such as the non-independence of the test cases, and (3) the disregard of the actual biomedical domain interest that the metrics should reflect. This living dynamically document has the purpose to illustrate important limitations of performance metrics commonly applied in the field of image analysis. In this context, it focuses on biomedical image analysis problems that can be phrased as image-level classification, semantic segmentation, instance segmentation, or object detection task. The current version is based on a Delphi process on metrics conducted by an international consortium of image analysis experts from more than 60 institutions worldwide., Comment: Shared first authors: Annika Reinke and Minu D. Tizabi. This is a dynamic paper on limitations of commonly used metrics. It discusses metrics for image-level classification, semantic and instance segmentation, and object detection. For missing use cases, comments or questions, please contact a.reinke@dkfz.de. Substantial contributions to this document will be acknowledged with a co-authorship

Most state-of-the-art instance segmentation methods have to be trained on densely annotated images. While difficult in general, this requirement is especially daunting for biomedical images, where domain expertise is often required for annotation and no large public data collections are available for pre-training. We propose to address the dense annotation bottleneck by introducing a proposal-free segmentation approach based on non-spatial embeddings, which exploits the structure of the learned embedding space to extract individual instances in a differentiable way. The segmentation loss can then be applied directly to instances and the overall pipeline can be trained in a fully- or weakly supervised manner. We consider the challenging case of positive-unlabeled supervision, where a novel self-supervised consistency loss is introduced for the unlabeled parts of the training data. We evaluate the proposed method on 2D and 3D segmentation problems in different microscopy modalities as well as on the Cityscapes and CVPPP instance segmentation benchmarks, achieving state-of-the-art results on the latter. The code is available at: https://github.com/kreshuklab/spoco, Comment: CVPR 2022 camera-ready

This work introduces a new proposal-free instance segmentation method that builds on single-instance segmentation masks predicted across the entire image in a sliding window style. In contrast to related approaches, our method concurrently predicts all masks, one for each pixel, and thus resolves any conflict jointly across the entire image. Specifically, predictions from overlapping masks are combined into edge weights of a signed graph that is subsequently partitioned to obtain all final instances concurrently. The result is a parameter-free method that is strongly robust to noise and prioritizes predictions with the highest consensus across overlapping masks. All masks are decoded from a low dimensional latent representation, which results in great memory savings strictly required for applications to large volumetric images. We test our method on the challenging CREMI 2016 neuron segmentation benchmark where it achieves competitive scores., Comment: Presented at GCPR 2020

Semantic instance segmentation is the task of simultaneously partitioning an image into distinct segments while associating each pixel with a class label. In commonly used pipelines, segmentation and label assignment are solved separately since joint optimization is computationally expensive. We propose a greedy algorithm for joint graph partitioning and labeling derived from the efficient Mutex Watershed partitioning algorithm. It optimizes an objective function closely related to the Symmetric Multiway Cut objective and empirically shows efficient scaling behavior. Due to the algorithm's efficiency it can operate directly on pixels without prior over-segmentation of the image into superpixels. We evaluate the performance on the Cityscapes dataset (2D urban scenes) and on a 3D microscopy volume. In urban scenes, the proposed algorithm combined with current deep neural networks outperforms the strong baseline of `Panoptic Feature Pyramid Networks' by Kirillov et al. (2019). In the 3D electron microscopy images, we show explicitly that our joint formulation outperforms a separate optimization of the partitioning and labeling problems.

Recent advances in high-throughput electron microscopy imaging enable detailed study of centrosome aberrations in cancer cells. While the image acquisition in such pipelines is automated, manual detection of centrioles is still necessary to select cells for re-imaging at higher magnification. In this contribution we propose an algorithm which performs this step automatically and with high accuracy. From the image labels produced by human experts and a 3D model of a centriole we construct an additional training set with patch-level labels. A two-level DenseNet is trained on the hybrid training data with synthetic patches and real images, achieving much better results on real patient data than training only at the image-level. The code can be found at https://github.com/kreshuklab/centriole_detection., Comment: Accepted at MICCAI 2019

We propose a theoretical framework that generalizes simple and fast algorithms for hierarchical agglomerative clustering to weighted graphs with both attractive and repulsive interactions between the nodes. This framework defines GASP, a Generalized Algorithm for Signed graph Partitioning, and allows us to explore many combinations of different linkage criteria and cannot-link constraints. We prove the equivalence of existing clustering methods to some of those combinations and introduce new algorithms for combinations that have not been studied before. We study both theoretical and empirical properties of these combinations and prove that some of these define an ultrametric on the graph. We conduct a systematic comparison of various instantiations of GASP on a large variety of both synthetic and existing signed clustering problems, in terms of accuracy but also efficiency and robustness to noise. Lastly, we show that some of the algorithms included in our framework, when combined with the predictions from a CNN model, result in a simple bottom-up instance segmentation pipeline. Going all the way from pixels to final segments with a simple procedure, we achieve state-of-the-art accuracy on the CREMI 2016 EM segmentation benchmark without requiring domain-specific superpixels., Comment: Published in CVPR 2022

The throughput of electron microscopes has increased significantly in recent years, enabling detailed analysis of cell morphology and ultrastructure. Analysis of neural circuits at single-synapse resolution remains the flagship target of this technique, but applications to cell and developmental biology are also starting to emerge at scale. The amount of data acquired in such studies makes manual instance segmentation, a fundamental step in many analysis pipelines, impossible. While automatic segmentation approaches have improved significantly thanks to the adoption of convolutional neural networks, their accuracy still lags behind human annotations and requires additional manual proof-reading. A major hindrance to further improvements is the limited field of view of the segmentation networks preventing them from exploiting the expected cell morphology or other prior biological knowledge which humans use to inform their segmentation decisions. In this contribution, we show how such domain-specific information can be leveraged by expressing it as long-range interactions in a graph partitioning problem known as the lifted multicut problem. Using this formulation, we demonstrate significant improvement in segmentation accuracy for three challenging EM segmentation problems from neuroscience and cell biology.

Image partitioning, or segmentation without semantics, is the task of decomposing an image into distinct segments, or equivalently to detect closed contours. Most prior work either requires seeds, one per segment; or a threshold; or formulates the task as multicut / correlation clustering, an NP-hard problem. Here, we propose an efficient algorithm for graph partitioning, the "Mutex Watershed''. Unlike seeded watershed, the algorithm can accommodate not only attractive but also repulsive cues, allowing it to find a previously unspecified number of segments without the need for explicit seeds or a tunable threshold. We also prove that this simple algorithm solves to global optimality an objective function that is intimately related to the multicut / correlation clustering integer linear programming formulation. The algorithm is deterministic, very simple to implement, and has empirically linearithmic complexity. When presented with short-range attractive and long-range repulsive cues from a deep neural network, the Mutex Watershed gives the best results currently known for the competitive ISBI 2012 EM segmentation benchmark.

In the last years, automated segmentation has become a necessary tool for volume electron microscopy (EM) imaging. So far, the best performing techniques have been largely based on fully supervised encoder-decoder CNNs, requiring a substantial amount of annotated images. Domain Adaptation (DA) aims to alleviate the annotation burden by 'adapting' the networks trained on existing groundtruth data (source domain) to work on a different (target) domain with as little additional annotation as possible. Most DA research is focused on the classification task, whereas volume EM segmentation remains rather unexplored. In this work, we extend recently proposed classification DA techniques to an encoder-decoder layout and propose a novel method that adds a reconstruction decoder to the classical encoder-decoder segmentation in order to align source and target encoder features. The method has been validated on the task of segmenting mitochondria in EM volumes. We have performed DA from brain EM images to HeLa cells and from isotropic FIB/SEM volumes to anisotropic TEM volumes. In all cases, the proposed method has outperformed the extended classification DA techniques and the finetuning baseline. An implementation of our work can be found on https://github.com/JorisRoels/domain-adaptive-segmentation., Comment: ISBI 2019 (accepted)

Automated reconstruction of neural connectivity graphs from electron microscopy image stacks is an essential step towards large-scale neural circuit mapping. While significant progress has recently been made in automated segmentation of neurons and detection of synapses, the problem of synaptic partner assignment for polyadic (one-to-many) synapses, prevalent in the Drosophila brain, remains unsolved. In this contribution, we propose a method which automatically assigns pre- and postsynaptic roles to neurites adjacent to a synaptic site. The method constructs a probabilistic graphical model over potential synaptic partner pairs which includes factors to account for a high rate of one-to-many connections, as well as the possibility of the same neuron to be pre-synaptic in one synapse and post-synaptic in another. The algorithm has been validated on a publicly available stack of ssTEM images of Drosophila neural tissue and has been shown to reconstruct most of the synaptic relations correctly.

ROOT is an object-oriented C++ framework conceived in the high-energy physics (HEP) community, designed for storing and analyzing petabytes of data in an efficient way. Any instance of a C++ class can be stored into a ROOT file in a machine-independent compressed binary format. In ROOT the TTree object container is optimized for statistical data analysis over very large data sets by using vertical data storage techniques. These containers can span a large number of files on local disks, the web, or a number of different shared file systems. In order to analyze this data, the user can chose out of a wide set of mathematical and statistical functions, including linear algebra classes, numerical algorithms such as integration and minimization, and various methods for performing regression analysis (fitting). In particular, ROOT offers packages for complex data modeling and fitting, as well as multivariate classification based on machine learning techniques. A central piece in these analysis tools are the histogram classes which provide binning of one- and multi-dimensional data. Results can be saved in high-quality graphical formats like Postscript and PDF or in bitmap formats like JPG or GIF. The result can also be stored into ROOT macros that allow a full recreation and rework of the graphics. Users typically create their analysis macros step by step, making use of the interactive C++ interpreter CINT, while running over small data samples. Once the development is finished, they can run these macros at full compiled speed over large data sets, using on-the-fly compilation, or by creating a stand-alone batch program. Finally, if processing farms are available, the user can reduce the execution time of intrinsically parallel tasks - e.g. data mining in HEP - by using PROOF, which will take care of optimally distributing the work over the available resources in a transparent way.

For the last several months the main focus of development in the ROOT I/O package has been code consolidation and performance improvements. Access to remote files is affected both by bandwidth and latency. We introduced a pre-fetch mechanism to minimize the number of transactions between client and server and hence reducing the effect of latency. We will review the implementation and how well it works in different conditions (gain of an order of magnitude for remote file access). We will also review new utilities, including a faster implementation of TTree cloning (gain of an order of magnitude), a generic mechanism for object references, and a new entry list mechanism tuned both for small and large number of selections. In addition to reducing the coupling with the core module and becoming its owns library (libRIO) (as part of the general restructuration of the ROOT libraries), the I/O package has been enhanced in the area of XML and SQL support, thread safety, schema evolution, TTreeFormula, and many other areas. We will also discuss various ways, ROOT will be able to benefit from multi-core architecture to improve I/O performances.