Volume 8 Supplement 1
Proceedings of the 11th European Congress on Telepathology and 5th International Congress on Virtual Microscopy
Automatic measurement of the evolutionary process dynamics of primary of biliary cirrhosis
 Nicola Dioguardi^{1},
 Carlo Russo^{1},
 Emanuela Morenghi^{2},
 Barbara Franceschini^{1},
 Sonia Di Biccari^{1},
 Stefano Musardo^{1} and
 Guido Bosticco^{3}
DOI: 10.1186/174615968S1S16
© Dioguardi et al; licensee BioMed Central Ltd. 2013
Published: 30 September 2013
Background
PBC is a chronic inflammatory autoimmune disease [1] that evolves into cirrhosis via four stages. These are determined by the successive implementation in hepatic tissue of the following events: inflammation, destruction and regeneration of biliary tissue CK7+ and fibrosis. This paper describes a method to study of the dynamics of the disease histologic behaviour [2].
Materials and methods
We studied 58 liver biopsy samples, from the archives of Departments of Pathological Anatomy of Istituto Clinico Humanitas IRCCS, Rozzano, Milan, Italy and the Department of Gastroenterology of Milan’s Ospedale Policlinico IRCCS, of Caucasian subjects with PBC in various stages. Each procedure was performed in accordance with the guidelines of the Ethics Committees of the hospitals involved and diagnoses were defined by the expert pathologists of these same hospitals.
Three consecutive 23 µm thick sections were obtained from each sample: the first was used to identify inflammatory cells; the second was stained with Sirius red to visualize collagen deposits and the third was used to visualize the biliary tree with CK7 antibody.
The histological metrization method [3] was developed in the Laboratory for the Study of Metric Methods in Medicine of the Istituto Clinico Humanitas (Rozzano, Milan, Italy). The prototype of this computerized device called HM and its controlling software were entirely designed and developed in our laboratory. The Metrizer is a totally computerdriven machine that automates the focusing of the microscope and the exclusion of Glisson’s capsule from the computation of fibrotic islets. It performs reproducible metric measurements from digitalized images of the entire histological section, giving results within a few minutes. This method measures all notable liver structures bared of any property not involved in the measurement.
Statistical analysis
The data are given as numbers and percentages or mean or median values and ranges where appropriate. All variables were logtransformed in order to approximate a Gaussian distribution and normalized to a (0,1) range to allow for comparison between variables. Thus the final data are expressed in percentage of the interval between 0 and 1. All of the analyses were performed with Stata10 [http://www.stata.com].
Results and discussion
This method identifies a line that connects the centers of outermost cells that are distanced ≤20 μm. This line is arbitrarily considered the separator of the inflammatory cells within a cluster from the mononuclear cells dispersed in the surrounding hepatic tissue.
The inflammatory basin, which increases constantly due to the autoimmune process, is the set of areas in a section that are covered by inflammatory cell clusters. The area of the basin covered by clusterresident cells is called the pure inflammatory space, which varies with the number (density) of cells present in the clusters.
The intrahepatic biliary duct area (Figure 1B) was metrically measured. During the course of PBC, the biliary tree duct status is determined by two components: autoimmune destruction and regeneration of intrahepatic CK7+ ductular segment.
The collagen islets forming fibrosis (Figure 1C) were measured in linear meters; the result was corrected by the fractal dimension [5] to include details of the irregularity of their shapes The fractal dimension was obtained by means of the box counting method because the objects to be measured were “truncated fractals” [6], the fractal dimension was used as a dilation factor rather than an exponent [7] Three classes of islet magnitude were arbitrarily identified: area from 10 and 10^{3} μm^{2}, from 10^{3} to 10^{4} μm^{2} and over 10^{4} μm^{2}.
Summary of all of the metric data obtained from the structural measurements and the data used for staging purposes. Minimum, median, and maximum values of tissue parameters are given in % of total histologic section area of biopsy specimen.
Tissue parameters  Minimum  Median  Maximum 

Area of the inflammatory basin  0.026  1.005  6.444 
Area covered by resident cells  0.002  0.636  5.295 
Area of intrahepatic biliary ducts  0.003  0.114  0.882 
Small areas (1010^{3} μm)  0  0.009  0.105 
Mediumsized areas (10^{3}10^{4} μm)  0.001  0.006  0.033 
Large areas (>10^{4} μm)  0  0  0 
Area of fibrosis  0.30  1.456  24.4 
Small areas (1010^{3} μm)  0.008  0.092  0.521 
Mediumsized areas (10^{3}10^{4} μm)  0.001  0.026  0.203 
Large areas (>10^{4} μm)  0.01  0.075  1.005 
Neoangiogenesis  0.030  0.586  3.474 
Hemopoietic stem cells  0  0.006  0? 
Tissue disorder (TI)  0.072  0.274  0.643 
In order to approach the PBC behavioural dynamics [9], the first key was to set into the interval (0,1) the logarithmic transformation and normalization of the measures of CK7+; tissue and fibrosis, taken as the most representative structural elements with which to metrically construct the liver state portrait that defines the grading and staging of the semiquantitative subjective methodologies.
A second key is the transformation of portrait scalars into a single vector to reduce the multiplicity of elements of the liver section into a dotlike geometrical figure. This translation into vectorial arithmetic in classic dynamics is crucial for the construction of the dotlike geometrical operator, called dynamic particola that represents the whole system.
Furthermore, we plotted the value of each particula on the y axis (ranging from 0 to 1) versus the disease timing, identified with all particulae, on the x axis (Figure 2C). Next, we identified the tertiles on this curve that discriminate three phases. This term used in our new method is different from the ‘stages’ characterising semiquantitative static descriptions. As a result, the three phases are extrapolated by increasing amounts of particolae formed by fibrosis and CK7+ tissue. In particular, phases 1, 2, and 3 represent early, intermediate, and final disease to feature mean particula values of 0.2075, 0.4722, and 0.7386, respectively.
The inflammation was excluded from the constitutive components of the trajectory describing the course of PBC process, for its reversibility. This different behaviour is due to the entropy produced in the interior of the inflammation process that is not transferred across its boundaries into the surrounding environment [10].
Conclusions
The method we constructed, with its technology, strongly reduced the computational time and improved the liver tissue structure recognition and description. As these tools allowed this first study of PBC behaviour in terms of physics of dynamics, this paper supports the hypothesis that the long periods of cessation in the history of dynamic knowledge were due to the very higher rapid theoretic development toward facilitating computation devices [11, 12]. The technology introduced into our methodology facilitated the study of PBC behaviour by the following points:
1) Exclusion of the inflammatory infiltrates from dynamic study, as they do not produce topic stable entropy deposits. Collagen interstitial deposition generating fibrosis is hepatocellular necrosis dependent.
2) Standardization of a strictly objective evaluation producing scalars by metrizing the images of the histological structures of the liver section.
3) Metrical measurement also of the smallest dispersed islets of fibrosis, and tiny CK7+ biliary ducts normally undetected with optical microscope.
4) Transformation of scalars into vectors leading to vectorial PBC histological section portraits. A homogeneity was created with this operation that maintains the concepts of mixture and identity of the mixed elements in the sum of CK7+ biliary ducts and fibrosis vectors that define the particola, geometric figure which graphically will trace the process trajectory.
5) Description of PBC evolution by the cumulative curve of the particolae, each representing an actual state of the processes and taking this planar curve as the ideal finite trajectory α–ω) of entire PBC process.
6) Definition of past and future percentage of the disease course with its particola on the trajectory.
7) The score of PBC evolution into three phases on its trajectory and describe the ideal course of the process based on mathematical measurements.
Distribution in the three stadia of the dynamic trajectory of the patients’list classified according to Scheuer’s classification.
Scheuer classification  

Phase  1  2  3  4  
I  19 (33%)  11 (58%)  7 (37%)  1 (5%)  0 
II  19 (33%)  11 (58%)  1 (5%)  6 (32%)  1 (5%) 
III  20 (34%)  9 (45%)  2 (10%)  5 (25%)  4 (20%) 
List of abbreviations
 PBC:

Primary biliary cirrhosis
 IRCCS:

Institute for Scientific Research Hospitalisation and Care
 HM:

Histologic Metrizer
 TI:

Tectonic Index
Declarations
Acknowledgements
This study was completed with the financial support of the Fondazione Michele Rodriguez, Istituto Clinico Humanitas IRCCS, and Attilio and Livia Pagani.
The entire group is extremely grateful to Rosalind Roberts for her numerous translations of the manuscripts.
Authors’ Affiliations
References
 Rubin E, Schüffner F, Popper H: Primary biliary cirrhosis. Chronic nonsuppurative destructive cholangitis. Am J Pathol. 1965, 46: 387407.PubMed CentralPubMedGoogle Scholar
 Dioguardi N: hypothesis for a new method to measure the dynamic patterns of tissue injury. Med Hypotheses. 2011, 77 (6): 10227. 10.1016/j.mehy.2011.08.039.View ArticlePubMedGoogle Scholar
 Dioguardi N, Grizzi F, Fiamengo B, Russo C: Metrically measuring liver biopsy: a chronic hepatitis B and C computeraided morphologic description. World J Gastroenterol. 2008, 14: 73357344. 10.3748/wjg.14.7335.PubMed CentralView ArticlePubMedGoogle Scholar
 Delaunay BN: Sur la sphère vide. 1934, Math and Nat Sci Division, Bulletin of the Academy of Sciences of USSR, 6: 793Google Scholar
 Dioguardi N, Franceschini B, Aletti G, Russo C, Grizzi F: Fractal dimension rectified meter for quantification of liver fibrosis and other irregular microscopic objects. Anal Quant Cytol Histol. 2003, 25: 312320.PubMedGoogle Scholar
 Rigaut JP, SchoëvaërtBrossault D, Downs AM, Landini G: Asymptotic fractals in the context of greyscale images. J Microsc. 1998, 189: 5763. 10.1046/j.13652818.1998.00284.x.View ArticlePubMedGoogle Scholar
 Dioguardi N, Grizzi F, Bossi P, Roncalli M: Fractal and spectral dimension analysis of liver fibrosis in needle biopsy specimens. Anal. Quant. Cytol. Histol. 1999, 21: 262266.PubMedGoogle Scholar
 Hurst HE: Long term storage capacity of reservoirs. Trans Amer Soc Civ Engrs. 1951, 116: 770808.Google Scholar
 Abraham R, Shaw R: Dynamics. The geometry of behaviour. Part 2. Periodic behaviour. 1985, Aerial Press, Inc Santa Cruz, California, third printingGoogle Scholar
 From being to becoming: Time and complexity in the physical sciences. Edited by: Prigogine I. 1980, San Francisco, CA, WH freeman and Co
 Samburski S: The physical world of the Greeks. 1956, Routlerge and Kegan Paul LondonGoogle Scholar
 Russel B: Storia della filosophia occidentale. 1910, TEA, Seventh essayGoogle Scholar
Copyright
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.