Élodie Puybareau

Automated heart rate estimation in fish embryo

Transparent organisms such as fish embryos are being increasingly used for environmental toxicology studies. These studies require estimating a number of physiological parameters. These estimations may be diverse in nature and can be a challenge to automate. Among these, an example is the development of reliable and repeatable automated assays for the determination of heart rates. To achieve this, most existing method rely on cyclical luminance variations, since as the heart fills and empties, it become respectively brighter and darker. However, sometimes direct measurement of the heart rate may be difficult, depending on the age of the embryo, its actual transparency, and its aspect under the microscope. It may be easier to seek an indirect measurement. In this article, we estimate the heart function parameters, such as heart frequency, either from measuring the heart motion or from blood flow in arteries. This measurement is more complex from the image analysis point of view, but it is more precise, more physically meaningful and easier to use in practice and to automate than measuring illumination changes. It may also be more informative. We illustrate on medaka embryos.

A regionalized automated measurement of ciliary beating frequency

Cilia are slender, microscopic, hair-like structures or organelles that extend from the surface of nearly all mammalian cells. Motile cilia, such as those found in the lungs and respiratory tract, present a beating motion that keep the airways clear of mucus and dirt. They are thus of primary importance in many respiratory diseases. The performance of mucous transportation in the nasal cavity can be represented by a ciliary beating frequency. In this paper, we propose a fully automated method that computes the beating frequency from a sequence of images taken with high-speed videomicroscopy. The advantage of our approach is its capacity in computing regionalized frequencies, i.e., various frequencies each associated with one region in the image. Moreover we propose a preprocessing pipeline to alleviate acquistion artefacts due to the camera or to the cell proper motions. We demonstrate the robustness of our approach, and illustrate its performance in comparison to the state-of-the-art.

An automated assay for the assessment of cardiac arrest in fish embryo

Studies on fish embryo models are widely developed in research. They are used in several research fields including drug discovery or environmental toxicology. In this article, we propose an entirely automated assay to detect cardiac arrest in Medaka (Oryzias latipes) based on image analysis. We propose a multi-scale pipeline based on mathematical morphology. Starting from video sequences of entire wells in 24-well plates, we focus on the embryo, detect its heart, and ascertain whether or not the heart is beating based on intensity variation analysis. Our image analysis pipeline only uses commonly available operators. It has a low computational cost, allowing analysis at the same rate as acquisition. From an initial dataset of 3192 videos, 660 were discarded as unusable (20.7%), 655 of them correctly so (99.25%) and only 5 incorrectly so (0.75%). The 2532 remaining videos were used for our test. On these, 45 errors were made, leading to a success rate of 98.23%.

Automating the measurement of physiological parameters: A case study in the image analysis of cilia motion

As image processing and analysis techniques improve, an increasing number of procedures in bio-medical analyses can be automated. This brings many benefits, e.g improved speed and accuracy, leading to more reliable diagnoses and follow-up, ultimately improving patients outcome. Many automated procedures in bio-medical imaging are well established and typically consist of detecting and counting various types of cells (e.g. blood cells, abnormal cells in Pap smears, and so on). In this article we propose to automate a different and difficult set of measurements, which is conducted on the cilia of people suffering from a variety of respiratory tract diseases. Cilia are slender, microscopic, hair-like structures or organelles that extend from the surface of nearly all mammalian cells. Motile cilia, such as those found in the lungs and respiratory tract, present a periodic beating motion that keep the airways clear of mucus and dirt. In this paper, we propose a fully automated method that computes various measurements regarding the motion of cilia, taken with high-speed video-microscopy. The advantage of our approach is its capacity to automatically compute robust, adaptive and regionalized measurements, i.e. associated with different regions in the image. We validate the robustness of our approach, and illustrate its performance in comparison to the state-of-the-art.

An Automated Assay for the Evaluation of Mortality in Fish Embryo

Fish embryo models are used increasingly for human disease modeling, chemical toxicology screening, drug discovery and environmental toxicology studies. These studies are devoted to the analysis of a wide spectrum of physiological parameters, such as mortality ratio. In this article, we develop an assay to determine Medaka (Oryzias latipes) embryo mortality. Based on video sequences, our purpose is to obtain reliable, repeatable results in a fully automated fashion. To reach that challenging goal, we develop an efficient morphological pipeline that analyses image sequences in a multiscale paradigm, from the global scene to the embryo, and then to its heart, finally analysing its putative motion, characterized by intensity variations. Our pipeline, based on robust morphological operators, has a low computational cost, and was experimentally assessed on a dataset consisting of 660 images, providing a success ratio higher than 99%.

The challenge of cerebral magnetic resonance imaging in neonates: A new method using mathematical morphology for the segmentation of structures including diffuse excessive high signal intensities.

HighlightsNew morphological method to segment neonatal brain tissues and hyperintensities.Max‐tree, a contrast invariant hierarchical representation, is used to model images.No atlas is required and no nonlinear registration is involved.The proposed method achieves top results on both 1.5T and 3T T2 weighted images.A user‐friendly interface integrating the complete proposed method is provided. Graphical abstract Figure. No caption available. ABSTRACT Preterm birth is a multifactorial condition associated with increased morbidity and mortality. Diffuse excessive high signal intensity (DEHSI) has been recently described on T2‐weighted MR sequences in this population and thought to be associated with neuropathologies. To date, no robust and reproducible method to assess the presence of white matter hyperintensities has been developed, perhaps explaining the current controversy over their prognostic value. The aim of this paper is to propose a new semi‐automated framework to detect DEHSI on neonatal brain MR images having a particular pattern due to the physiological lack of complete myelination of the white matter. A novel method for semi‐ automatic segmentation of neonatal brain structures and DEHSI, based on mathematical morphology and on max‐tree representations of the images is thus described. It is a mandatory first step to identify and clinically assess homogeneous cohorts of neonates for DEHSI and/or volume of any other segmented structures. Implemented in a user‐friendly interface, the method makes it straightforward to select relevant markers of structures to be segmented, and if needed, apply eventually manual corrections. This method responds to the increasing need for providing medical experts with semi‐automatic tools for image analysis, and overcomes the limitations of visual analysis alone, prone to subjectivity and variability. Experimental results demonstrate that the method is accurate, with excellent reproducibility and with very few manual corrections needed. Although the method was intended initially for images acquired at 1.5T, which corresponds to the usual clinical practice, preliminary results on images acquired at 3T suggest that the proposed approach can be generalized.

Morphological Analysis of Brownian Motion for Physical Measurements

Brownian motion is a well-known, apparently chaotic motion affecting microscopic objects in fluid media. The mathematical and physical basis of Brownian motion have been well studied but not often exploited. In this article we propose a particle tracking methodology based on mathematical morphology, suitable for Brownian motion analysis, which can provide difficult physical measurements such as the local temperature and viscosity. We illustrate our methodology on simulation and real data, showing that interesting phenomena and good precision can be achieved.

Periodic area-of-motion characterization for bio-medical applications

Many bio-medical applications involve the analysis of sequences for motion characterization. In this article, we consider 2D+t sequences where a particular motion (e.g. a blood flow) is associated with a specific area of the 2D image (e.g. an artery) but multiple motions may exist simultaneously in the same sequences (e.g. there may be several blood vessels present, each with their specific flow). The characterization of this type of motion typically involves first finding the areas where motion is present, followed by an analysis of these motions: speed, regularity, frequency, etc. In this article, we propose a methodology called “area-of-motion characterization” suitable for simultaneously detecting and characterizing areas where motion is present in a sequence. We can then classify this motion into consistent areas using unsupervised learning and produce directly usable metrics for various applications. We illustrate this methodology for the analysis of cilia motion on ex-vivo human samples, and we apply and validate the same methodology for blood flow analysis in fish embryo.

White Matter Hyperintensities Segmentation in a Few Seconds Using Fully Convolutional Network and Transfer Learning

In this paper, we propose a fast automatic method that segments white matter hyperintensities (WMH) in 3D brain MR images, using a fully convolutional network (FCN) and transfer learning. This FCN is the Visual Geometry Group neural network (VGG for short) pre-trained on ImageNet for natural image classification, and fine tuned with the training dataset of the MICCAI WMH Challenge. We consider three images for each slice of the volume to segment: the T1 slice, the FLAIR slice, and the result of a morphological operator that emphasizes small bright structures. These three 2D images are assembled to form a 2D color image, that inputs the FCN to obtain the 2D segmentation of the corresponding slice. We process all slices, and stack the results to form the 3D output segmentation. With such a technique, the segmentation of WMH on a 3D brain volume takes about 10 s including pre-processing. Our technique was ranked 6-th over 20 participants at the MICCAI WMH Challenge.

Automatic detection of beating cilia with frequencies estimations

Muco-ciliary clearance is the airway first mechanism of defence against environmental attacks such as microorganisms or pollution. Cilia motility impairment can be either of genetic (primary ciliary dyskinesia) or acquired origin (environmental attacks), entailing chronic diseases. It is of interest for practitioners to evaluate cilia beating frequency easily, robustly and reliably. As yet, no fully automatized method is available. 2 Methods Ciliated cells were sampled in patients by brushing nasal mucosa and cilia beating was recorded using high speed video microscopy. We first estimated and removed the sensor pattern. We then stabilized the sequence assuming rigid transforms. We retained only the moving parts of the sequence and, after deblurring, characterized and segmented the moving parts in several regions of interest. The frequency was estimated for each region.