Nicolas Loménie

Mitosis detection in breast cancer histological images An ICPR 2012 contest.

Introduction: In the framework of the Cognitive Microscope (MICO) project, we have set up a contest about mitosis detection in images of H and E stained slides of breast cancer for the conference ICPR 2012. Mitotic count is an important parameter for the prognosis of breast cancer. However, mitosis detection in digital histopathology is a challenging problem that needs a deeper study. Indeed, mitosis detection is difficult because mitosis are small objects with a large variety of shapes, and they can thus be easily confused with some other objects or artefacts present in the image. We added a further dimension to the contest by using two different slide scanners having different resolutions and producing red-green-blue (RGB) images, and a multi-spectral microscope producing images in 10 different spectral bands and 17 layers Z-stack. 17 teams participated in the study and the best team achieved a recall rate of 0.7 and precision of 0.89. Context: Several studies on automatic tools to process digitized slides have been reported focusing mainly on nuclei or tubule detection. Mitosis detection is a challenging problem that has not yet been addressed well in the literature. Aims: Mitotic count is an important parameter in breast cancer grading as it gives an evaluation of the aggressiveness of the tumor. However, consistency, reproducibility and agreement on mitotic count for the same slide can vary largely among pathologists. An automatic tool for this task may help for reaching a better consistency, and at the same time reducing the burden of this demanding task for the pathologists. Subjects and Methods: Professor Frιdιrique Capron team of the pathology department at Pitiι-Salpκtriθre Hospital in Paris, France, has selected a set of five slides of breast cancer. The slides are stained with H and E. They have been scanned by three different equipments: Aperio ScanScope XT slide scanner, Hamamatsu NanoZoomer 2.0-HT slide scanner and 10 bands multispectral microscope. The data set is made up of 50 high power fields (HPF) coming from 5 different slides scanned at ×40 magnification. There are 10 HPFs/slide. The pathologist has annotated all the mitotic cells manually. A HPF has a size of 512 μm × 512 μm (that is an area of 0.262 mm 2 , which is a surface equivalent to that of a microscope field diameter of 0.58 mm. These 50 HPFs contain a total of 326 mitotic cells on images of both scanners, and 322 mitotic cells on the multispectral microscope. Results : Up to 129 teams have registered to the contest. However, only 17 teams submitted their detection of mitotic cells. The performance of the best team is very promising, with F-measure as high as 0.78. However, the database we provided is by far too small for a good assessment of reliability and robustness of the proposed algorithms. Conclusions : Mitotic count is an important criterion in the grading of many types of cancers, however, very little research has been made on automatic mitotic cell detection, mainly because of a lack of available data. A main objective of this contest was to propose a database of mitotic cells on digitized breast cancer histopathology slides to initiate works on automated mitotic cell detection. In the future, we would like to extend this database to have much more images from different patients and also for different types of cancers. In addition, mitotic cells should be annotated by several pathologists to reflect the partial agreement among them.

Time-efficient sparse analysis of histopathological whole slide images

Histopathological examination is a powerful standard for the prognosis of critical diseases. But, despite significant advances in high-speed and high-resolution scanning devices or in virtual exploration capabilities, the clinical analysis of whole slide images (WSI) largely remains the work of human experts. We propose an innovative platform in which multi-scale computer vision algorithms perform fast analysis of a histopathological WSI. It relies on application-driven for high-resolution and generic for low-resolution image analysis algorithms embedded in a multi-scale framework to rapidly identify the high power fields of interest used by the pathologist to assess a global grading. GPU technologies as well speed up the global time-efficiency of the system. Sparse coding and dynamic sampling constitute the keystone of our approach. These methods are implemented within a computer-aided breast biopsy analysis application based on histopathology images and designed in collaboration with a pathology department. The current ground truth slides correspond to about 36,000 high magnification (40×) high power fields. The processing time to achieve automatic WSI analysis is on a par with the pathologists performance (about ten minutes a WSI), which constitutes by itself a major contribution of the proposed methodology.

A cognitive virtual microscopic framework for knowlege-based exploration of large microscopic images in breast cancer histopathology

Histopathological examination is a powerful method for prognosis of major diseases such as breast cancer. Analysis of medical images largely remains the work of human experts. Current virtual microscope systems are mainly an emulation of real microscopes with annotation and some image analysis capabilities. However, the lack of effective knowledge management prevents such systems from being computer-aided prognosis platforms. The cognitive virtual microscopic framework, through an extended modeling and use of medical knowledge, has the capacity to analyse histopathological images and to perform grading of breast cancer, providing pathologists with a robust and traceable second opinion.

An Exploration Scheme for Large Images: Application to Breast Cancer Grading

Most research works focus on pattern recognition within a small sample images but strategies for running efficiently these algorithms over large images are rarely if ever specifically considered. In particular, the new generation of satellite and microscopic images are acquired at a very high resolution and a very high daily rate. We propose an efficient, generic strategy to explore large images by combining computational geometry tools with a local signal measure of relevance in a dynamic sampling framework. An application to breast cancer grading from huge histopathological images illustrates the benefit of such a general strategy for new major applications in the field of microscopy.

How to involve structural modeling for cartographic object recognition tasks in high-resolution satellite images?

With the new generation of satellite systems, very high resolution satellite images will be available daily at a high delivery rate. The exploitation of such a huge amount of data will be made possible by the design of high performance analysis algorithms for decision making systems. In particular, the detection and recognition of complex man-made objects is a new challenge coming with this new level of resolution. In this study, we develop a system that recognizes such structured and compact objects like bridges or roundabouts. The original contribution of this work is the use of structural shape attributes in an appearance-based statistical learning method framework leading to valuable recognition and false alarm rates. This hybrid structural/statistical approach aims to construct an intermediate step between the low-level image characteristics and high-level semantic concepts.

Region-based segmentation on depth images from a 3D reference surface for tree species recognition

The aim of the work presented in this paper is to develop a method for the automatic identification of tree species using Terrestrial Light Detection and Ranging (T-LiDAR) data. The approach that we propose analyses depth images built from 3D point clouds corresponding to a 30 cm segment of the tree trunk in order to extract characteristic shape features used for classifying the different tree species using the Random Forest classifier. We will present the method used to transform the 3D point cloud to a depth image and the region based segmentation method used to segment the depth images before shape features are computed on the segmented images. Our approach has been evaluated using two datasets acquired in two different French forests with different terrain characteristics. The results obtained are very encouraging and promising.

Cognitive virtual microscopy: a cognition-driven visual explorer for histopathology – the MICO ANR TecSan 2010 initiative

Within the last decade, histopathology became widely accepted as a powerful exam for diagnosis and prognosis in mainstream diseases such as breast cancer. Currently, analysis of medical images in histopathology largely remains the work of human experts. For pathologists, this consists of hundreds of slides examined daily. Such a tedious manual work is often inconsistent and subjective. The recent cognitive microscope – MICO - ANR TecSan project aims at radically modifying the medical practices by proposing a new cognitive medical imaging environment able to improve reliability of decision-making and prognosis assistance in histopathology. Our goal is to design a generic, open-ended, semantic digital histology platform including a cognitive dimension. MICO combines visual perception, pervasive exploration of whole slide images, context (including uncertainties) modeling, cognitive vision and quality of experience to reinforce a visual diagnosis assistance following an approach centered on the user behavior. http://ipal.i2r.a-star.edu.sg/project_MICO.htm

Advances in Bio-Imaging: From Physics to Signal Understanding Issues

Advances in Imaging Devices and Image processing stem from cross-fertilization between many fields of research such as Chemistry, Physics, Mathematics and Computer Sciences. This BioImaging Community feel the urge to integrate more intensively its various results, discoveries and innovation into ready to use tools that can address all the new exciting challenges that Life Scientists (Biologists, Medical doctors, ...) keep providing, almost on a daily basis. Devising innovative chemical probes, for example, is an archetypal goal in which image quality improvement must be driven by the physics of acquisition, the image processing and analysis algorithms and the chemical skills in order to design an optimal bioprobe. This book offers an overview of the current advances in many research fields related to bioimaging and highlights the current limitations that would need to be addressed in the next decade to design fully integrated BioImaging Device.

Neural stem cell tracking with phase contrast video microscopy

Tracking and segmenting objects for video surveillance is a well known field of research and very efficient methods exist. Usually embedded in traffic surveillance camera, these processes are not necessary adapted for biological surveillance context. In stem cell study, the design of a framework to monitor cell development in real time improves the stem cell analysis and biological understanding. In this purpose, we propose to test the Σ - ▵ motion filter, normally developed for security and surveillance camera, in order to track neural stem cells and their evolution over time, based on phase contrast image sequences. The motion filter is based on the difference between the current frame and a reference image of the background and uses a recursive spatio-temporal morphological operator called hybrid reconstruction to compensate for ghost and trace usually occurring with those kinds of methods.

Chapter 4 - Point Set Analysis

Abstract Dealing with imaging issues usually entails handling digital radiometric images. However, visual data can be efficiently handled as geometric point sets either due to the nature of the acquisition device or the intrinsic redundancy within large amounts of radiometric data. Most research works about geometric structures are related to computer graphics and image synthesis; meshes as graph representations have been involved only in a few image analysis issues to date. Yet, much room remains for completing the visual analysis tools as most image analysis algorithms are designed ro radiometric data distributed over a regular grid. We propose to extend the standard image analysis toolbox to unstructured point sets usually connected via mesh structures such as Delaunay triangulations. A particular focus on mathematical morphology sheds light on the potential applications of these ideas. More specifically, applications to digital microscopy imaging issues are discussed and preliminary results are presented.