An automated image analysis methodology for classifying megakaryocytes in chronic myeloproliferative disorders
Introduction
Megakaryocytes constitute a population of bone marrow resident cells which are responsible for the production of blood platelets. Among bone marrow cells, the mature megakaryocytes have a peculiar morphology, being both large, as well as having a characteristically shape nucleus (see Fig. 1a). In several pathological conditions, either reactive or neoplastic affecting the bone marrow, the normal morphology of the megakaryocytes can be altered (see Fig. 1b). In particular, in the context of Philadelphia-negative chronic myeloproliferative diseases, a group of neoplastic clonal disorders of the bone marrow stem cell comprising chronic idiopathic myelofibrosis (IMF), policytaemia rubra vera (PV) and essential thrombocythaemia (ET), megakaryocytes have been regarded as key cells on which the histopathological diagnosis is mostly based (Michiels and Thiele, 2002, Thiele et al., 2000, Thiele et al., 2001). Actually, an array of histopathological features, including the size and shape of the megakaryocyte cytoplasm and the morphology of the nucleus, has been proposed and can provide enough information, allowing a correct differential diagnosis among different chronic myeloproliferative disorders and reactive pathological conditions (Florena et al., 2004, Gianelli et al., 2006). However, the issue of an effective standardization of such features has been widely debated and still represents a matter of concern (Thiele et al., 2005).
With respect to the previous work by the same authors (Tripodo et al., 2006), this study is mainly focused on methodological problems. We aimed at developing a system which could provide a non-parametric morphological analysis of human megakaryocytes which could assist the pathologist in differentiating normal from pathological megakaryocytes and in discriminating among different myeloproliferative conditions.
The detection of components embedded in histological images is difficult and in many cases their correct classification is very hard due to, for example, closely clustered cells, color variations of the marker and aberrations caused by the optical system of the acquisition device. Beksaç et al. (1997) proposed a semiautomated approach based on a set of features; however some problems need to be solved in the case of overlapping and touching cells. A number of supervised methods allowing the relating of objects together with their characterizing features, but this approach is time-consuming, subjective and error-prone. Coelho et al. (2002) introduced a methodology based both on binary images of retinal cells and on manual segmentation.
Previous contributions on the automatic analysis of cells required a hand-made segmentation. A system to automatically segment and track cells in microscopy sequences has been presented (Yang et al., 2006), but it deals with fluorescent images that do not contain enough details for a reliable classification.
We aimed at describing a non-parametric method to find the best set of features that identify an object which is useful to the problems of classification and query in pictorial databases.
This paper reports on a methodology based on standard and well know algorithms including mathematical morphology (Section 3), wavelet analysis (Section 4) and elliptic Fourier descriptors (Section 5). In spite of the selected algorithms needing some parameters, the proposed methodology has been designed so that the user does not define any parameters. The chosen components can be extracted with a number of algorithms and many their combinations can be arranged to improve the performances. A system for image analysis could give good and robust algorithms and it could be a sufficient solution to the problem. It does not require a user to make his own choice or to find his own efficient combination of algorithms. Moreover, many algorithms are equipped with a set of input parameters that are not easy to define for an inexperienced user. To avoid these problems, our methodology introduces one of the robust sequences of the necessary algorithms to identify the pre-selected components without parameters. Such a transformation was not easy and in some cases it was necessary to include substantial variations of the original technique: an exhaustive study was necessary to normalize the data during a pre-processing step; a collection of parameters was obtained with an auto-calibration procedure in the segmentation phase. Most operations can be performed quickly to find structures with a particular shape, size and orientation (Serra, 1982, Soille, 2003). The use of watershed clustering (which is closely connected to mathematical morphology) has been proposed by Jiang et al. (2006) to locate white blood cells and by Isitor and Thorne (2007) to segment nuclei.
Different steps are required in order to discriminate the megakaryocytes (see Fig. 2). A first phase segments the image to isolate the cytoplasm and its nucleus in a given microphotograph (Section 2). This information is used in the subsequent enhancement of the shape of the cell (Sections 3 Segmentation using mathematical morphology, 4 Segmentation using wavelets) and extraction of the features which return its distinctive signature (Sections 5 Shape analysis using elliptic Fourier descriptors, 6 Feature extraction). The actual classifier is a regression tree procedure applied on the set of these signatures (Section 7). Final conclusions are given in Section 8.
Section snippets
Pre-processing
It is usually hard to take good photographs of biopsies; for instance, though the hematoxylin stain is light purple, its exact intensity depends on the length of the exposure. The periphery of the images is generally out of focus due to high-power magnification and it is easy to perceive separate parts as one object due to the fact that the section is relatively thick. In other words, we are looking through a three-dimensional stripe and the more distant structures have smooth contours, even
Segmentation using mathematical morphology
Mathematical morphology is a branch of digital image analysis which uses concepts of algebra and geometry (Serra, 1982, Soille, 2003). Its theoretical foundations have been well established and already applied to many medical imaging systems to decompose complex shapes into more meaningful representations and separate them from undesirable parts. We have used mathematical morphology to extract or suppress structures of the nucleus with structuring element SE of a priori known shape, size and
Segmentation using wavelets
Wavelet analysis was introduced in the 1930s as a mathematical tool and provides a powerful approach for representing and processing data. Wavelets have been successfully applied to computer vision and medical imaging.
The shape of the cell, so far obtained, is normally well defined, but sometimes the edge of its nucleus must be improved; we have obtained good results by using the wavelet transform (Mallat, 1989, Mallat, 1996) to highlight structures of the nucleus with different sizes. The
Shape analysis using elliptic Fourier descriptors
Many biological forms, at different structural levels (organism, organs, etc.) have landmarks which are characteristic of specific anatomical shapes. For such shapes a formal statistical approach which considers the spatial arrangement of such landmarks is available. Unfortunately, these are lacking in cells and nuclei. Moreover, cytological observations are only partial, two-dimensional prospects of three-dimensional structures viewed through different planes of section (tissue slices) or
Feature extraction
Morphometry gives quantitative measures of structures with different levels of detail (Ohshima et al., 1995). Megakaryocyte morphology is usually normal or well-preserved in reactive conditions, with a round-to-oval shape, normal size of the cytoplasm and multi-lobulated nucleus. In cases of chronic myeloproliferative disorders, the cells often appear enlarged in size with over-abundant cytoplasm, irregular shape and hypo- or hyper-lobulated nucleus.
Starting from the segmented photographs,
Results
All the microphotographs were collected by a pathologist in order to select megakaryocytes containing microscopic fields only. The image database was entirely based on cases of essential thrombocythemia, idiopathic myelofibrosis and reactive thrombocytosis diagnosed between December 2002 and January 2005 at the Dipartimento di Patologia Umana, Università degli Studi di Palermo, and at the Collegium Medicum, Uniwersytetu Jagiellon`skiego of Krakòw. The biopsies had been performed for the purpose
Discussion and final remarks
In general, the methodologies to classify items need a number of parameters: in some cases they are visible and can be set by the user, in other cases they are embedded into the method. The proposed methodology grows from exhaustive and continuous discussions with experts: those interviews have highlighted the essential characteristics to discriminate between normal and pathological megakaryocytes. The combinations of such parameters can give rise to many boundless solutions; this has driven
Acknowledgements
We wish to thank Professor Zbigniew Rudzki of the Collegium Medicum, Uniwersytetu Jagiellon`skiego of Krakòw for his kind contribution in providing part of the image database. We are also grateful to Professor Emanuele Trucco of Heriot-Watt University, Edinburgh, for his useful suggestions during the drafting of this paper. Many thanks go the anonymous reviewers for their helpful and comprehensive comments. This work has been partially supported by a grant provided by the University of Palermo.
References (35)
- et al.
Shape-based features for cat ganglion retinal cells classification real-time imaging
Special Issue on Imaging in Bioinformatics
(2002) - et al.
Comparison between nuclear chromatin patterns of digitalized images of cells of the mammalian testicular and renal tissues: an imaging segmentation study
Computerized Medical Imaging and Graphics
(2007) Methods for gamma invariant colour image processing
Image and Vision Computing
(2003)- et al.
Elliptic Fourier feature of a closed contour
Computer Graphics and Image Processing
(1982) - et al.
On the validity of fractal dimension measurements in image analysis
Journal of Visual Communication and Image Representation
(1996) Practical Statistics for Medical Research
(1999)- et al.
An artificial intelligent diagnostic system on differential recognition of hematopoietic cells from microscopic images
Cytometry (Communications in Clinical Cytometry)
(1997) - et al.
Intelligent Data Analysis. An Introduction
(2003) An introduction to wavelets
(1992)- Cohen, A., Kovasevic, J., 1996. Wavelets: the mathematical background. Proceedings of the IEEE 8(4),...
Value of bone marrow biopsy in the diagnosis of essential thrombocythemia
Haematologica
Essential thrombocythemia or chronic idiopathic myelofibrosis? A single-center study based on hematopoietic bone marrow histology
Leukemia & Lymphoma
Digital Image Processing
An introduction to wavelets
IEEE Computational Science and Engineering
Decomposition of Hardy functions into square integrable wavelets of constant shape
SIAM Journal of Mathematical Analysis
Cited by (32)
Assessment of parathyroid gland cellularity by digital slide analysis
2022, Annals of Diagnostic PathologyCitation Excerpt :Previous studies reported that cellular features of H&E staining, nuclear and cytoplasmic texture, nuclear shape and nuclear/cytoplasmic ratio carry prognostic significance [6]. These studies included, for instance, breast cancer [7], myeloproliferative disorders [8] and neuroblastoma cases [9]. In this study we aimed to explore, for the first time, computerized application in the evaluation of parathyroid tissue pathology.
Introduction to digital image analysis in whole-slide imaging: A white paper from the digital pathology association
2019, Journal of Pathology InformaticsCartoon filter via adaptive abstraction
2016, Journal of Visual Communication and Image RepresentationA non-parametric segmentation methodology for oral videocapillaroscopic images
2014, Computer Methods and Programs in BiomedicineMolecular Correlates of Morphometric Subtypes in Glioblastoma Multiforme
2013, Computational Systems Biology: From Molecular Mechanisms to Disease: Second EditionAnalysis of Histopathological images: An Overview
2022, Proceedings of International Conference on Computing, Communication, Security and Intelligent Systems, IC3SIS 2022