arXiv subjects

Carsten Marr

Publications and source records attributed to Carsten Marr.

At least 55 records · Page 3Linked to original sources

arXiv2023

Active Learning Enhances Classification of Histopathology Whole Slide Images with Attention-based Multiple Instance Learning

In many histopathology tasks, sample classification depends on morphological details in tissue or single cells that are only visible at the highest magnification. For a pathologist, this implies tedious zooming in and out, while for a computational decision support algorithm, it leads to the analysis of a huge number of small image patches per whole slide image (WSI). Attention-based multiple instance learning (MIL), where attention estimation is learned in a weakly supervised manner, has been successfully applied in computational histopathology, but it is challenged by large numbers of irrelevant patches, reducing its accuracy. Here, we present an active learning approach to the problem. Querying the expert to annotate regions of interest in a WSI guides the formation of high-attention regions for MIL. We train an attention-based MIL and calculate a confidence metric for every image in the dataset to select the most uncertain WSIs for expert annotation. We test our approach on the CAMELYON17 dataset classifying metastatic lymph node sections in breast cancer. With a novel attention guiding loss, this leads to an accuracy boost of the trained models with few regions annotated for each class. Active learning thus improves WSIs classification accuracy, leads to faster and more robust convergence, and speeds up the annotation process. It may in the future serve as an important contribution to train MIL models in the clinically relevant context of cancer classification in histopathology.

cs.CV↗

arXiv2023

BEL: A Bag Embedding Loss for Transformer enhances Multiple Instance Whole Slide Image Classification

Multiple Instance Learning (MIL) has become the predominant approach for classification tasks on gigapixel histopathology whole slide images (WSIs). Within the MIL framework, single WSIs (bags) are decomposed into patches (instances), with only WSI-level annotation available. Recent MIL approaches produce highly informative bag level representations by utilizing the transformer architecture's ability to model the dependencies between instances. However, when applied to high magnification datasets, problems emerge due to the large number of instances and the weak supervisory learning signal. To address this problem, we propose to additionally train transformers with a novel Bag Embedding Loss (BEL). BEL forces the model to learn a discriminative bag-level representation by minimizing the distance between bag embeddings of the same class and maximizing the distance between different classes. We evaluate BEL with the Transformer architecture TransMIL on two publicly available histopathology datasets, BRACS and CAMELYON17. We show that with BEL, TransMIL outperforms the baseline models on both datasets, thus contributing to the clinically highly relevant AI-based tumor classification of histological patient material.

cs.CV↗

arXiv2022

Capturing Shape Information with Multi-Scale Topological Loss Terms for 3D Reconstruction

Reconstructing 3D objects from 2D images is both challenging for our brains and machine learning algorithms. To support this spatial reasoning task, contextual information about the overall shape of an object is critical. However, such information is not captured by established loss terms (e.g. Dice loss). We propose to complement geometrical shape information by including multi-scale topological features, such as connected components, cycles, and voids, in the reconstruction loss. Our method uses cubical complexes to calculate topological features of 3D volume data and employs an optimal transport distance to guide the reconstruction process. This topology-aware loss is fully differentiable, computationally efficient, and can be added to any neural network. We demonstrate the utility of our loss by incorporating it into SHAPR, a model for predicting the 3D cell shape of individual cells based on 2D microscopy images. Using a hybrid loss that leverages both geometrical and topological information of single objects to assess their shape, we find that topological information substantially improves the quality of reconstructions, thus highlighting its ability to extract more relevant features from image datasets.

cs.CV↗

arXiv2022

Anomaly-aware multiple instance learning for rare anemia disorder classification

Deep learning-based classification of rare anemia disorders is challenged by the lack of training data and instance-level annotations. Multiple Instance Learning (MIL) has shown to be an effective solution, yet it suffers from low accuracy and limited explainability. Although the inclusion of attention mechanisms has addressed these issues, their effectiveness highly depends on the amount and diversity of cells in the training samples. Consequently, the poor machine learning performance on rare anemia disorder classification from blood samples remains unresolved. In this paper, we propose an interpretable pooling method for MIL to address these limitations. By benefiting from instance-level information of negative bags (i.e., homogeneous benign cells from healthy individuals), our approach increases the contribution of anomalous instances. We show that our strategy outperforms standard MIL classification algorithms and provides a meaningful explanation behind its decisions. Moreover, it can denote anomalous instances of rare blood diseases that are not seen during the training phase.

cs.LG↗

arXiv2022

Unsupervised Cross-Domain Feature Extraction for Single Blood Cell Image Classification

Diagnosing hematological malignancies requires identification and classification of white blood cells in peripheral blood smears. Domain shifts caused by different lab procedures, staining, illumination, and microscope settings hamper the re-usability of recently developed machine learning methods on data collected from different sites. Here, we propose a cross-domain adapted autoencoder to extract features in an unsupervised manner on three different datasets of single white blood cells scanned from peripheral blood smears. The autoencoder is based on an R-CNN architecture allowing it to focus on the relevant white blood cell and eliminate artifacts in the image. To evaluate the quality of the extracted features we use a simple random forest to classify single cells. We show that thanks to the rich features extracted by the autoencoder trained on only one of the datasets, the random forest classifier performs satisfactorily on the unseen datasets, and outperforms published oracle networks in the cross-domain task. Our results suggest the possibility of employing this unsupervised approach in more complicated diagnosis and prognosis tasks without the need to add expensive expert labels to unseen data.

cs.CV↗

arXiv2022

DeStripe: A Self2Self Spatio-Spectral Graph Neural Network with Unfolded Hessian for Stripe Artifact Removal in Light-sheet Microscopy

Light-sheet fluorescence microscopy (LSFM) is a cutting-edge volumetric imaging technique that allows for three-dimensional imaging of mesoscopic samples with decoupled illumination and detection paths. Although the selective excitation scheme of such a microscope provides intrinsic optical sectioning that minimizes out-of-focus fluorescence background and sample photodamage, it is prone to light absorption and scattering effects, which results in uneven illumination and striping artifacts in the images adversely. To tackle this issue, in this paper, we propose a blind stripe artifact removal algorithm in LSFM, called DeStripe, which combines a self-supervised spatio-spectral graph neural network with unfolded Hessian prior. Specifically, inspired by the desirable properties of Fourier transform in condensing striping information into isolated values in the frequency domain, DeStripe firstly localizes the potentially corrupted Fourier coefficients by exploiting the structural difference between unidirectional stripe artifacts and more isotropic foreground images. Affected Fourier coefficients can then be fed into a graph neural network for recovery, with a Hessian regularization unrolled to further ensure structures in the standard image space are well preserved. Since in realistic, stripe-free LSFM barely exists with a standard image acquisition protocol, DeStripe is equipped with a Self2Self denoising loss term, enabling artifact elimination without access to stripe-free ground truth images. Competitive experimental results demonstrate the efficacy of DeStripe in recovering corrupted biomarkers in LSFM with both synthetic and real stripe artifacts.

eess.IV↗

arXiv2021

Sickle Cell Disease Severity Prediction from Percoll Gradient Images using Graph Convolutional Networks

Sickle cell disease (SCD) is a severe genetic hemoglobin disorder that results in premature destruction of red blood cells. Assessment of the severity of the disease is a challenging task in clinical routine since the causes of broad variance in SCD manifestation despite the common genetic cause remain unclear. Identification of the biomarkers that would predict the severity grade is of importance for prognosis and assessment of responsiveness of patients to therapy. Detection of the changes in red blood cell (RBC) density through separation of Percoll density gradient could be such marker as it allows to resolve intercellular differences and follow the most damaged dense cells prone to destruction and vaso-occlusion. Quantification of the images obtained from the distribution of RBCs in Percoll gradient and interpretation of the obtained is an important prerequisite for establishment of this approach. Here, we propose a novel approach combining a graph convolutional network, a convolutional neural network, fast Fourier transform, and recursive feature elimination to predict the severity of SCD directly from a Percoll image. Two important but expensive laboratory blood test parameters measurements are used for training the graph convolutional network. To make the model independent from such tests during prediction, the two parameters are estimated by a neural network from the Percoll image directly. On a cohort of 216 subjects, we achieve a prediction performance that is only slightly below an approach where the groundtruth laboratory measurements are used. Our proposed method is the first computational approach for the difficult task of SCD severity prediction. The two-step approach relies solely on inexpensive and simple blood analysis tools and can have a significant impact on the patients' survival in underdeveloped countries where access to medical instruments and doctors is limited

eess.IV↗

arXiv2021

Structure-Preserving Multi-Domain Stain Color Augmentation using Style-Transfer with Disentangled Representations

In digital pathology, different staining procedures and scanners cause substantial color variations in whole-slide images (WSIs), especially across different laboratories. These color shifts result in a poor generalization of deep learning-based methods from the training domain to external pathology data. To increase test performance, stain normalization techniques are used to reduce the variance between training and test domain. Alternatively, color augmentation can be applied during training leading to a more robust model without the extra step of color normalization at test time. We propose a novel color augmentation technique, HistAuGAN, that can simulate a wide variety of realistic histology stain colors, thus making neural networks stain-invariant when applied during training. Based on a generative adversarial network (GAN) for image-to-image translation, our model disentangles the content of the image, i.e., the morphological tissue structure, from the stain color attributes. It can be trained on multiple domains and, therefore, learns to cover different stain colors as well as other domain-specific variations introduced in the slide preparation and imaging process. We demonstrate that HistAuGAN outperforms conventional color augmentation techniques on a classification task on the publicly available dataset Camelyon17 and show that it is able to mitigate present batch effects.

eess.IV↗

arXiv2021

Fourier Transform of Percoll Gradients Boosts CNN Classification of Hereditary Hemolytic Anemias

Hereditary hemolytic anemias are genetic disorders that affect the shape and density of red blood cells. Genetic tests currently used to diagnose such anemias are expensive and unavailable in the majority of clinical labs. Here, we propose a method for identifying hereditary hemolytic anemias based on a standard biochemistry method, called Percoll gradient, obtained by centrifuging a patient's blood. Our hybrid approach consists on using spatial data-driven features, extracted with a convolutional neural network and spectral handcrafted features obtained from fast Fourier transform. We compare late and early feature fusion with AlexNet and VGG16 architectures. AlexNet with late fusion of spectral features performs better compared to other approaches. We achieved an average F1-score of 88% on different classes suggesting the possibility of diagnosing of hereditary hemolytic anemias from Percoll gradients. Finally, we utilize Grad-CAM to explore the spatial features used for classification.

cs.CV↗

arXiv2020

Attention based Multiple Instance Learning for Classification of Blood Cell Disorders

Red blood cells are highly deformable and present in various shapes. In blood cell disorders, only a subset of all cells is morphologically altered and relevant for the diagnosis. However, manually labeling of all cells is laborious, complicated and introduces inter-expert variability. We propose an attention based multiple instance learning method to classify blood samples of patients suffering from blood cell disorders. Cells are detected using an R-CNN architecture. With the features extracted for each cell, a multiple instance learning method classifies patient samples into one out of four blood cell disorders. The attention mechanism provides a measure of the contribution of each cell to the overall classification and significantly improves the network's classification accuracy as well as its interpretability for the medical expert.

cs.CV↗

arXiv2014

MCA: Multiresolution Correlation Analysis, a graphical tool for subpopulation identification in single-cell gene expression data

Background: Biological data often originate from samples containing mixtures of subpopulations, corresponding e.g. to distinct cellular phenotypes. However, identification of distinct subpopulations may be difficult if biological measurements yield distributions that are not easily separable. Results: We present Multiresolution Correlation Analysis (MCA), a method for visually identifying subpopulations based on the local pairwise correlation between covariates, without needing to define an a priori interaction scale. We demonstrate that MCA facilitates the identification of differentially regulated subpopulations in simulated data from a small gene regulatory network, followed by application to previously published single-cell qPCR data from mouse embryonic stem cells. We show that MCA recovers previously identified subpopulations, provides additional insight into the underlying correlation structure, reveals potentially spurious compartmentalizations, and provides insight into novel subpopulations. Conclusions: MCA is a useful method for the identification of subpopulations in low-dimensional expression data, as emerging from qPCR or FACS measurements. With MCA it is possible to investigate the robustness of covariate correlations with respect subpopulations, graphically identify outliers, and identify factors contributing to differential regulation between pairs of covariates. MCA thus provides a framework for investigation of expression correlations for genes of interests and biological hypothesis generation.

cs.GR↗

arXiv2011

Stability and multi-attractor dynamics of a toggle switch based on a two-stage model of stochastic gene expression

A toggle switch consists of two genes that mutually repress each other. This regulatory motif is active during cell differentiation and is thought to act as a memory device, being able to choose and maintain cell fate decisions. In this contribution, we study the stability and dynamics of a two-stage gene expression switch within a probabilistic framework inspired by the properties of the Pu/Gata toggle switch in myeloid progenitor cells. We focus on low mRNA numbers, high protein abundance and monomeric transcription factor binding. Contrary to the expectation from a deterministic description, this switch shows complex multi-attractor dy- namics without autoactivation and cooperativity. Most importantly, the four attractors of the system, which only emerge in a probabilistic two-stage description, can be identified with committed and primed states in cell differentiation. We first study the dynamics of the system and infer the mechanisms that move the system between attractors using both the quasi-potential and the probability flux of the system. Second, we show that the residence times of the system in one of the committed attractors are geometrically distributed and provide an analytical expression of the distribution. Most importantly we find that the mean residence time increases linearly with the mean protein level. Finally, we study the implications of this distribution for the stability of a switch and discuss the influence of the stability on a specific cell differentiation mechanism. Our model explains lineage priming and proposes the need of either high protein numbers or long term modifications such as chromatin remodeling to achieve stable cell fate decisions. Notably we present a system with high protein abundance that nevertheless requires a probabilistic description to exhibit multistability, complex switching dynamics and lineage priming.

q-bio.MN↗

arXiv2010

Patterns of subnet usage reveal distinct scales of regulation in the transcriptional regulatory network of Escherichia coli

The set of regulatory interactions between genes, mediated by transcription factors, forms a species' transcriptional regulatory network (TRN). By comparing this network with measured gene expression data one can identify functional properties of the TRN and gain general insight into transcriptional control. We define the subnet of a node as the subgraph consisting of all nodes topologically downstream of the node, including itself. Using a large set of microarray expression data of the bacterium Escherichia coli, we find that the gene expression in different subnets exhibits a structured pattern in response to environmental changes and genotypic mutation. Subnets with less changes in their expression pattern have a higher fraction of feed-forward loop motifs and a lower fraction of small RNA targets within them. Our study implies that the TRN consists of several scales of regulatory organization: 1) subnets with more varying gene expression controlled by both transcription factors and post-transcriptional RNA regulation, and 2) subnets with less varying gene expression having more feed-forward loops and less post-transcriptional RNA regulation.

q-bio.MN↗

arXiv2008

Outer-totalistic cellular automata on graphs

We present an intuitive formalism for implementing cellular automata on arbitrary topologies. By that means, we identify a symmetry operation in the class of elementary cellular automata. Moreover, we determine the subset of topologically sensitive elementary cellular automata and find that the overall number of complex patterns decreases under increasing neighborhood size in regular graphs. As exemplary applications, we apply the formalism to complex networks and compare the potential of scale-free graphs and metabolic networks to generate complex dynamics.

nlin.CG↗

arXiv2007

Two distinct logical types of network control in gene expression profiles

In unicellular organisms such as bacteria the same acquired mutations beneficial in one environment can be restrictive in another. However, evolving Escherichia coli populations demonstrate remarkable flexibility in adaptation. The mechanisms sustaining genetic flexibility remain unclear. In E. coli the transcriptional regulation of gene expression involves both dedicated regulators binding specific DNA sites with high affinity and also global regulators - abundant DNA architectural proteins of the bacterial chromoid binding multiple low affinity sites and thus modulating the superhelical density of DNA. The first form of transcriptional regulation is dominantly pairwise and specific, representing digitial control, while the second form is (in strength and distribution) continuous, representing analog control. Here we look at the properties of effective networks derived from significant gene expression changes under variation of the two forms of control and find that upon limitations of one type of control (caused e.g. by mutation of a global DNA architectural factor) the other type can compensate for compromised regulation. Mutations of global regulators significantly enhance the digital control; in the presence of global DNA architectural proteins regulation is mostly of the analog type, coupling spatially neighboring genomic loci; together our data suggest that two logically distinct types of control are balancing each other. By revealing two distinct logical types of control, our approach provides basic insights into both the organizational principles of transcriptional regulation and the mechanisms buffering genetic flexibility. We anticipate that the general concept of distinguishing logical types of control will apply to many complex biological networks.

q-bio.GN↗

arXiv2007

Reply to ''Comment on 'Regularizing Capacity of Metabolic Networks' ''

In a recent paper [C. Marr, M. Mueller-Linow, and M.-T. Huett, Phys. Rev. E 75, 041917 (2007)] we discuss the pronounced potential of real metabolic network topologies, compared to randomized counterparts, to regularize complex binary dynamics. In their comment [P. Holme and M. Huss, arXiv:0705.4084v1], Holme and Huss criticize our approach and repeat our study with more realistic dynamics, where stylized reaction kinetics are implemented on sets of pairwise reactions. The authors find no dynamic difference between the reaction sets recreated from the metabolic networks and randomized counterparts. We reproduce the author's observation and find that their algorithm leads to a dynamical fragmentation and thus eliminates the topological information contained in the graphs. Hence, their approach cannot rule out a connection between the topology of metabolic networks and the ubiquity of steady states.

q-bio.MN↗

arXiv2006

The Regularizing Capacity of Metabolic Networks

Despite their topological complexity almost all functional properties of metabolic networks can be derived from steady-state dynamics. Indeed, many theoretical investigations (like flux-balance analysis) rely on extracting function from steady states. This leads to the interesting question, how metabolic networks avoid complex dynamics and maintain a steady-state behavior. Here, we expose metabolic network topologies to binary dynamics generated by simple local rules. We find that the networks' response is highly specific: Complex dynamics are systematically reduced on metabolic networks compared to randomized networks with identical degree sequences. Already small topological modifications substantially enhance the capacity of a network to host complex dynamic behavior and thus reduce its regularizing potential. This exceptionally pronounced regularization of dynamics encoded in the topology may explain, why steady-state behavior is ubiquitous in metabolism.

q-bio.MN↗

arXiv2005

Similar impact of topological and dynamic noise on complex patterns

Shortcuts in a regular architecture affect the information transport through the system due to the severe decrease in average path length. A fundamental new perspective in terms of pattern formation is the destabilizing effect of topological perturbations by processing distant uncorrelated information, similarly to stochastic noise. We study the functional coincidence of rewiring and noisy communication on patterns of binary cellular automata.

cond-mat.stat-mech↗