=Paper=
{{Paper
|id=Vol-2100/paper9
|storemode=property
|title=Ontology-based Automatic Reclassication of
Tissues and Organs in Histological Images
|pdfUrl=https://ceur-ws.org/Vol-2100/paper9.pdf
|volume=Vol-2100
|authors=Claudia Mazo,Maria Trujillo,Enrique Alegre,Liliana Salazar
|dblpUrl=https://dblp.org/rec/conf/amw/MazoTAS18
}}
==Ontology-based Automatic Reclassication of
Tissues and Organs in Histological Images==
https://ceur-ws.org/Vol-2100/paper9.pdf
Ontology-based Automatic Reclassification of
Tissues and Organs in Histological Images
Claudia Mazo1,2,3 , Maria Trujillo1 , Enrique Alegre4 , and Liliana Salazar5
1
Universidad del Valle, Computer and Systems Engineering School, Cali, Colombia
claudia.mazo@correounivalle.edu.co
2
University College Dublin, CeADAR: Centre for Applied Data Analytics Research,
School of Computer Science, Dublin, Ireland
3
OncoMark Limited, Dublin, Ireland
4
Universidad de León, Industrial and Informatics Engineering School, León, Spain
5
Universidad del Valle, Morphology Department, Cali, Colombia
Abstract. Heterogeneous data source produces different types of data
that cannot be treated in the same way. In this paper, two sources of
data are considered: image and human knowledge. The former is rep-
resented using visual descriptors and the latter is represented using an
ontology. The integration of these data sources is used in the automatic
classification of tissues and organs of the human cardiovascular system
together. Firstly, visual descriptors – texture descriptors – are used in
the automatic classification using a cascade Support Vector Machine.
Secondly, obtained classification results are refined using a histological
ontology of the human cardiovascular system to confirm or reclassified.
The final classification results are more precise than the obtained using
only image data, in all cases.
Keywords: Automatic Classification · Histological Ontology · Histology
Images· Image Processing.
1 Introduction
On the one hand, a computer vision problem consists of identifying the funda-
mental tissues and recognising distinctive patterns, formed by spatial structures
among them, in order to infer an organ. The solution of this problem may be
used to reinforce learning processes and translate into better-formed profession-
als without requiring any mayor social or economic investment [3]. On the other
hand, histological knowledge representation is used to solve complex tasks, such
as: support teaching and medical practices or have natural language interactions,
which are challenges. In fact, multiple and heterogeneous data sources produce
different types of representations of data that cannot be treated in the same
way, being an open problem. In this paper, we present a method that combines
the automatic classification based on texture descriptors with the histological
ontology in order to improve the classification results.
Ontologies and taxonomies contain relevant knowledge represented with rich
structural and semantic information. Approaches that use those knowledge in an
automatic classification process are twofold: (i) model the relation between visual
and semantic information [9] [10] and (ii) use those ontologies and taxonomies
in the classification algorithm [6] [8] [1] [7] [2]. We will focus in the second group
to perform the classification process using images and an ontology, after creating
a histological ontology [5].
In this paper, we describe the use of an ontology to refine the results of a
classification based on image data. We propose two different methods: (i) refining
an organs classification and (ii) a classification of epithelial tissue. The proposed
approaches allow to reduce the classification error, in all cases.
2 Method
Our proposal consists of three steps: (1) Input: histological image, along with
histological and expert knowledge are used as input source. (2) Image process-
ing: given a histological image, the block-based classification method, proposed
in [4], is used in the classification of individual blocks. Classified blocks are
concatenated in a way to represent an image. (3) The histological knowledge
representation: the ontology, presented in [5], is used to refine classification re-
sults. We propose two methods, as follows: (3.1) refining organ classification and
(3.2) recognising epithelial tissue. The refining processes are described as follows.
2.1 Refining organ classification
A given image is divided into blocks and each block is classified in one of six
classes. Four classes are considered discriminative classes, whilst two classes are
non-discriminative. Discriminative classes are associated to organs.
Afterwards, classification results are refining by a SPARQL’s query, based on
RDF triples derived from the frequencies per discriminative class. The RDF’s
subject is the organ with higher frequency among the classes. The RDF’s ob-
jects are the remaining classes. RDF triples are built with the predicate hasPres-
enceOf. If the obtained result is empty, then blocks classified in the organ class,
used as object in the RDF triple, should be reclassified. A new label will be de-
cided according to the behaviour of false positives in the classification process.
In other case, the classification is confirmed and the organ class is not modified.
2.2 Refining epithelial tissue classification
The two non-discriminative classes are used for identifying epithelial tissue, us-
ing as input the concatenated blocks forming an image. Firstly, large image
regions, classified as light regions, are selected using a threshold value, selected
heuristically, in order to decide if there is epithelial tissue. Secondly, if a light
region is large enough, there is not loose connective tissue in the neighbourhood.
Then, blocks on the neighbourhood are verified not to be labelled as loose con-
nective. Thirdly, RDF triples are built using frequencies of discriminative and
non-discriminative classes. The RDF’s subject is the organ with higher frequency
among the discriminative classes; the object is TejidoEpitelialRevestimiento or
EpithelialLining; and the predicates are someValuesFrom and subClassOf. The
possible results are: (i) the coating epithelial or the lining epithelial is in the
subject. In this case, blocks surrounding light region and between muscle region
are labelled as the epithelial tissue. (ii) an empty result means that it is highly
probably non-presence of epithelial tissue. This result confirm the classification
and the blocks are not modified.
3 Evaluation and Results
Tissue samples from organs were stained with Hematoxylin and Eosin. Initially,
1500 blocks belonging to five histological images of different organs and persons
acquired at 10× objective were manually labelled. Img-He and Img-He1 repre-
sent the heart images; Img-MA represents the muscular artery image; Img-EA
represents the elastic artery images; and Img-LV represents the large vein image.
We left this dataset publicly available at http://biscar.univalle.edu.co/datasets.
Algorithms were implemented in C++, using the CImg library, SPARQL and
Protégé in a computer of 8-cores and 8Gb of RAM.
Results of the refining classification process were evaluated based on the True
Positive (TP) and the False Positive (FP). Figure 1(a) contains the total of TP
and FP obtained using the initial and the refined classifications, by organ. It can
be observed that the TP is increased after the refining. The highest increment in
the TP is obtained with the elastic artery. Figure 1(b) contains a graphical rep-
resentation of the TP and the FP by blocks obtained after identifying epithelial
tissue in the set of test images. The reclassified blocks correctly classified be-
tween 0 to 7 per image and the recognition of epithelial tissue process increases
TP rate classification between 0% and 2.333% according to the area which con-
tain epithelial tissue. It is important to highlight that epithelial tissue regions
occupy a smaller proportion in histological images, for this reason rates of im-
provement are highly variable between images and less than 3%. Additionally,
the behaviour after the epithelial recognition process corresponds to increasing
TP in images with epithelial tissue.
Fig. 1. Results (a) Block diagram of the TP and the FP of the initial and the re-
fined classifications, by organ. (b) TP and FP blocks with automatic classification and
improvement process.
4 Conclusions
Our refinement proposal enables us to obtain more consistent information and
to reduce margins of error and uncertainty through corroboration and verifica-
tion, by comparing analysis of different data sources separately. Besides, more
information of histological knowledge as a system, a composition, but also as
structures, relations, regions, layers, sectors, tissues and cells is obtained or in-
ferred from an image. Epithelial tissue is identifies in 10× magnification im-
ages that cannot be done manually. As future work, automatic identification of
micro-circulation organs using macro-circulation identified in this work could be
proposed.
5 Acknowledgment
This project has received funding from the EI and from the European Union’s
Horizon 2020 research and innovation programme under the Marie Slodowska-
Curie grant agreement No 713654
References
1. Abdollahpour, Z., Samani, Z.R., Moghaddam, M.E.: Image classification using on-
tology based improved visual words. In: 2015 23rd Iranian Conference on Electrical
Engineering. pp. 694–698 (May 2015)
2. Breen, C., Khan, L., Kumar, A., Wang, L.: Ontology-based image classification
using neural networks (2002)
3. Izet, M.: E-learning as new method of medical education. Acta Informatica Medica
16(2), 102–117 (2008)
4. Mazo, C., Alegre, E., Trujillo, M.: Classification of cardiovascular tissues using
lbp based descriptors and a cascade svm. Computer Methods and Programs in
Biomedicine 147, 1 – 10 (2017)
5. Mazo, C., Salazar, L., Corcho, Ó., Trujillo, M.P., Alegre, E.: A histological ontology
of the human cardiovascular system. J. Biomedical Semantics 8(1), 47:1–47:15
(2017)
6. Othmani, A., Meziat, C., Loménie, N.: Ontology-driven image analysis for
histopathological images. In: Advances in Visual Computing - 6th International
Symposium, ISVC 2010, Las Vegas, NV, USA, November 29-December 1, 2010.
Proceedings, Part I. pp. 1–12 (2010)
7. Paulson, P., Hohimer, R., Doucette, P., Harvey, W., Seedahmed, G., Petrie, G.,
Martucci, L.: A methodology for integrating images and text for object identifica-
tion. In: Prospecting for Geospatial Information Integration. pp. 1057–1066 (2006)
8. Smith, B., Arabandi, S., Brochhausen, M., Calhoun, M., Ciccarese, P., Doyle, S.,
Gurcan, M.: Biomedical imaging ontologies: A survey and proposal for future work.
Journal of Pathology Informatics 6(37) (2015)
9. Wu, Z., Jiang, S., Li, L., Cui, P., Huang, Q., Gao, W.: Vicept: link visual features
to concepts for large-scale image understanding. In: Proceedings of the 18th Inter-
national Conference on Multimedia 2010, Firenze, Italy, October 25-29, 2010. pp.
711–714 (2010)
10. Yang, J., Jiang, Y.G., Hauptmann, A.G., Ngo, C.W.: Evaluating bag-of-visual-
words representations in scene classification. In: Proceedings of the International
Workshop on Workshop on Multimedia Information Retrieval. pp. 197–206. MIR
’07, ACM (2007)