Daniel Sage

Related publications (77)

Deep Learning Enables Individual Xenograft Cell Classification in Histological Images by Analysis of Contextual Features

Cathrin Brisken, Olivier Burri, Fabio De Martino, Quentin Juppet, Daniel Sage, Michaël Unser, Martin Weigert

Patient-Derived Xenografts (PDXs) are the preclinical models which best recapitulate inter- and intra-patient complexity of human breast malignancies, and are also emerging as useful tools to study the normal breast epithelium. However, data analysis generated with such models is often confounded by the presence of host cells and can give rise to data misinterpretation. For instance, it is important to discriminate between xenografted and host cells in histological sections prior to performing immunostainings. We developed Single Cell Classifier (SCC), a data-driven deep learning-based computational tool that provides an innovative approach for automated cell species discrimination based on a multi-step process entailing nuclei segmentation and single cell classification. We show that human and murine cell contextual features, more than cell-intrinsic ones, can be exploited to discriminate between cell species in both normal and malignant tissues, yielding up to 96% classification accuracy. SCC will facilitate the interpretation of H&E- and DAPI-stained histological sections of xenografted human-in-mouse tissues and it is open to new in-house built models for further applications. SCC is released as an open-source plugin in ImageJ/Fiji available at the following link: https://github.com/Biomedical-Imaging-Group/SingleCellClassifier.

2021

DeepImageJ: A user-friendly environment to run deep learning models in ImageJ

Laurène Donati, Daniel Sage, Michaël Unser

DeepImageJ offers a user-friendly solution in ImageJ to run trained deep learning models for biomedical image analysis. It includes guiding tools for reliable analyses, contributing to the democratization of deep learning in microscopy. DeepImageJ is a user-friendly solution that enables the generic use of pre-trained deep learning models for biomedical image analysis in ImageJ. The deepImageJ environment gives access to the largest bioimage repository of pre-trained deep learning models (BioImage Model Zoo). Hence, nonexperts can easily perform common image processing tasks in life-science research with deep learning-based tools including pixel and object classification, instance segmentation, denoising or virtual staining. DeepImageJ is compatible with existing state of the art solutions and it is equipped with utility tools for developers to include new models. Very recently, several training frameworks have adopted the deepImageJ format to deploy their work in one of the most used softwares in the field (ImageJ). Beyond its direct use, we expect deepImageJ to contribute to the broader dissemination and reuse of deep learning models in life sciences applications and bioimage informatics.

NATURE PORTFOLIO2021

In pancreatic islets from type 2 diabetes patients, the dampened circadian oscillators lead to reduced insulin and glucagon exocytosis

Daniel Sage

Circadian clocks operative in pancreatic islets participate in the regulation of insulin secretion in humans and, if compromised, in the development of type 2 diabetes (T2D) in rodents. Here we demonstrate that human islet alpha- and beta-cells that bear attenuated clocks exhibit strongly disrupted insulin and glucagon granule docking and exocytosis. To examine whether compromised clocks play a role in the pathogenesis of T2D in humans, we quantified parameters of molecular clocks operative in human T2D islets at population, single islet, and single islet cell levels. Strikingly, our experiments reveal that islets from T2D patients contain clocks with diminished circadian amplitudes and reduced in vitro synchronization capacity compared to their nondiabetic counterparts. Moreover, our data suggest that islet clocks orchestrate temporal profiles of insulin and glucagon secretion in a physiological context. This regulation was disrupted in T2D subjects, implying a role for the islet cell-autonomous clocks in T2D progression. Finally, Nobiletin, an agonist of the core-clock proteins ROR alpha/gamma, boosted both circadian amplitude of T2D islet clocks and insulin secretion by these islets. Our study emphasizes a link between the circadian clockwork and T2D and proposes that clock modulators hold promise as putative therapeutic agents for this frequent disorder.

NATL ACAD SCIENCES2020

, , , , ,

In fluorescence microscopy live-cell imaging, there is a critical trade-off between the signal-to-noise ratio and spatial resolution on one side, and the integrity of the biological sample on the other side. To obtain clean high-resolution (HR) images, one can either use microscopy techniques, such as structured-illumination microscopy (SIM), or apply denoising and super-resolution (SR) algorithms. However, the former option requires multiple shots that can damage the samples, and although efficient deep learning based algorithms exist for the latter option, no benchmark exists to evaluate these algorithms on the joint denoising and SR (JDSR) tasks. To study JDSR on microscopy data, we propose such a novel JDSR dataset, Widefield2SIM (W2S), acquired using a conventional fluorescence widefield and SIM imaging. W2S includes 144,000 real fluorescence microscopy images, resulting in a total of 360 sets of images. A set is comprised of noisy low-resolution (LR) widefield images with different noise levels, a noise-free LR image, and a corresponding high-quality HR SIM image. W2S allows us to benchmark the combinations of 6 denoising methods and 6 SR methods. We show that state-of-the-art SR networks perform very poorly on noisy inputs. Our evaluation also reveals that applying the best denoiser in terms of reconstruction error followed by the best SR method does not necessarily yield the best final result. Both quantitative and qualitative results show that SR networks are sensitive to noise and the sequential application of denoising and SR algorithms is sub-optimal. Lastly, we demonstrate that SR networks retrained end-to-end for JDSR outperform any combination of state-of-the-art deep denoising and SR networks

2020

, , ,

A two-platform measurement system for realizing airborne thermography of the Lake Surface Water Temperature (LSWT) with ~0.8 m pixel resolution (sub-pixel satellite scale) is presented. It consists of a tethered Balloon Launched Imaging and Monitoring Platform (BLIMP) that records LSWT images and an autonomously operating catamaran (called ZiviCat) that measures in situ surface/near surface temperatures within the image area, thus permitting simultaneous ground-truthing of the BLIMP data. The BLIMP was equipped with an uncooled InfraRed (IR) camera. The ZiviCat was designed to measure along predefined trajectories on a lake. Since LSWT spatial variability in each image is expected to be low, a poor estimation of the common spatial and temporal noise of the IR camera (nonuniformity and shutter-based drift, respectively) leads to errors in the thermal maps obtained. Nonuniformity was corrected by applying a pixelwise two-point linear correction method based on laboratory experiments. A Probability Density Function (PDF) matching in regions of overlap between sequential images was used for the drift correction. A feature matching-based algorithm, combining blob and region detectors, was implemented to create composite thermal images, and a mean value of the overlapped images at each location was considered as a representative value of that pixel in the final map. The results indicate that a high overlapping field of view (~95%) is essential for image fusion and noise reduction over such low-contrast scenes. The in situ temperatures measured by the ZiviCat were then used for the radiometric calibration. This resulted in the generation of LSWT maps at sub-pixel satellite scale resolution that revealed spatial LSWT variability, organized in narrow streaks hundreds of meters long and coherent patches of different size, with unprecedented detail.

2019

Achieving high-resolution thermal imagery in low-contrast lake surface waters by aerial remote sensing and image registration

David Andrew Barry, Abolfazl Irani Rahaghi, Ulrich Lemmin, Daniel Sage

2019

Closed-Form Expression of the Fourier Ring-Correlation for Single-Molecule Localization Microscopy

Thanh-An Michel Pham, Daniel Sage, Emmanuel Emilien Louis Soubies, Michaël Unser

Single-molecule localization microscopy (SMLM) is a popular microscopic technique that achieves super resolution imaging by localizing individual blinking molecules in thousands of frames. Therefore, the reconstructed high-resolution image is a combination of millions of point sources. This particular computational reconstruction leads to the question of the estimation of the image resolution. Fourier-ring correlation (FRC) is the standard tool for assessing the resolution. It has been proposed for SMLM by computing a discrete correlation in the Fourier domain. In this work, we derive a closed-form expression to compute the continuous FRC. Our implementation provides an exact FRC and an alternative to compute a parameter-free FRC. In addition, it gives insights on the discrepancy of the discrete FRC and yields a rule to select its parameters such as the spatial sampling step or the width of the kernel used as density estimator.

IEEE2019

, , ,

We present a new active contour to segment cell aggregates. We describe it by a smooth tessellation that is attracted toward the cell membranes. Our approach relies on subdivision schemes that are tightly linked to the theory of wavelets. The shape is encoded by control points grouped in tiles. The smooth and continuously defined boundary of each tile is generated by recursively applying a refinement process to its control points. We deform the smooth tessellation in a global manner using a ridge-based energy that we have designed for that purpose. By construction, cells are segmented without overlap and the tessellation structure is maintained even on dim membranes. Leakage, which afflicts usual image-processing methods (e.g., watershed), is thus prevented. We validate our framework on both synthetic and real microscopy images, showing that the proposed method is robust to membrane gaps and to high levels of noise.

IEEE2019

, , , , , , ,

GlobalBioIm is an open-source MATLAB (R) library for solving inverse problems. The library capitalizes on the strong commonalities between forward models to standardize the resolution of a wide range of imaging inverse problems. Endowed with an operator-algebra mechanism, GlobalBioIm allows one to easily solve inverse problems by combining elementary modules in a lego-like fashion. This user-friendly toolbox gives access to cutting-edge reconstruction algorithms, while its high modularity makes it easily extensible to new modalities and novel reconstruction methods. We expect GlobalBioIm to respond to the needs of imaging scientists looking for reliable and easy-to-use computational tools for solving their inverse problems. In this paper, we present in detail the structure and main features of the library. We also illustrate its flexibility with examples from multichannel deconvolution microscopy.

IOP PUBLISHING LTD2019

Super-resolution fight club: assessment of 2D and 3D single-molecule localization microscopy software

Anurag Agrawal, Anna Archetti, Silvia Colabrese, Ricardo Henriques, Seamus John Holden, Tomas Lukes, Thomas Pengo, Thanh-An Michel Pham, Daniel Sage, Michaël Unser

With the widespread uptake of two-dimensional (2D) and three-dimensional (3D) single-molecule localization microscopy (SMLM), a large set of different data analysis packages have been developed to generate super-resolution images. In a large community effort, we designed a competition to extensively characterize and rank the performance of 2D and 3D SMLM software packages. We generated realistic simulated datasets for popular imaging modalities-2D, astigmatic 3D, biplane 3D and double-helix 3D-and evaluated 36 participant packages against these data. This provides the first broad assessment of 3D SMLM software and provides a holistic view of how the latest 2D and 3D SMLM packages perform in realistic conditions. This resource allows researchers to identify optimal analytical software for their experiments, allows 3D SMLM software developers to benchmark new software against the current state of the art, and provides insight into the current limits of the field.

2019