=Paper=
{{Paper
|id=Vol-1175/CLEF2009wn-ImageCLEF-KovalevEt2009
|storemode=property
|title=United Institute of Informatics - Problems at CLEF 2009 Medical Image Retrieval Task
|pdfUrl=https://ceur-ws.org/Vol-1175/CLEF2009wn-ImageCLEF-KovalevEt2009.pdf
|volume=Vol-1175
|dblpUrl=https://dblp.org/rec/conf/clef/KovalevSDP09
}}
==United Institute of Informatics - Problems at CLEF 2009 Medical Image Retrieval Task==
https://ceur-ws.org/Vol-1175/CLEF2009wn-ImageCLEF-KovalevEt2009.pdf
United Institute Of Informatics Problems at
CLEF2009: Medical Image Retrieval Task
Vassili Kovalev, Igor Safonov, Alexander Dmitruk, Alexander Prus
United Institute of Informatics Problems, Surganova st. 6, Minsk, 220012, Belarus
vassili.kovalev@gmail.com, igorsafonov@gmail.com
dmitruk@newman.bas-net.by, alex prus@tut.by
Abstract
This paper describes methods and results archived by our research group at the
Cross Language Evaluation Forum (CLEF’2009). In our work we concentrated on the
medical image retrieval task. All the attention was given to the retrieval based on nine
visual topics only. No textual information was considered.
Our goal was to develop and comparatively assess image descriptors for content
based retrieval on the large medical image database. In addition to results of official
relevance judgment, time- and computational efficiency of the algorithms has also been
of our interest.
An approach based on generalized co-occurrence matrices was used. The following
elementary image characteristics have been utilized: signal intensity, absolute value of
the intensity gradient and the angle between the gradient vectors in the pixel pairs.
Categories and Subject Descriptors
H.3 [Information Storage and Retrieval]: H.3.1 Content Analysis and Indexing; H.3.3 Infor-
mation Search and Retrieval; H.3.4 Systems and Software; H.3.7 Digital Libraries
Keywords
Medical imaging, CBIR, Generalized co-occurrence matrix, Distance measure
1 Introduction
This paper describes basic methods and results archived during the medical image retrieval com-
petition of CLEF 2009 [1]. Since we are working with visual information only, most of attention
is given to generalized co-occurrence matrices as image descriptors.
2 Methods
Image retrieval procedure consists of two steps. Initially, image content is represented by global
features, i.e. descriptors are calculated. Next, the descriptors are compared to a sample image in
order to find most similar ones.
2.1 Image descriptors
Generalized co-occurrence matrices [2] extend classical Haralick matrices [3] first introduced in
1973. Co-occurrence matrix for gray level image can be a matrix of relative frequencies P (i, k, d)
with which the pixels with gray levels i, k occur in the image by distance d apart (Fig. 1).
Figure 1: Pair of pixels with gray levels i, k at distance d.
As both variables take continuous values, the construction of such a histogram will require
their quantization. Depending on the range of relative distances considered and the number of
bins used, this step could be quite computationally intensive. However, it is not necessary to
consider a large range of distances. Depending on the task we want to perform, we may restrict
ourselves to only a small range of. If this distance value is small, the histogram will contain local
information concerning the described shape or texture. If the value is large, global information is
conveyed.
Taking into account the extensive experience on the medical image analysis, classification
and recognition using these descriptors, the following elementary image characteristics have been
utilized:
• brightness (signal intensity);
• absolute value of the intensity gradient;
• the angle between the gradient vectors in the pixel pairs.
As it was shown experimentally [4, 5], together with the information on their variation in space
these elementary image characteristics may describe the image structure properly.
A multidimensional co-occurrence matrix of a general form that combines all of above charac-
teristics and relationships looks as follows:
W = kw (I(i), I(k), G(i), G(k), A(i, k), d(i, k))k ,
where
i = (xi , yi ), k = (xk , yk ) is an arbitrarily pair of pixels,
d(i, k) is the distance between them,
I(i), I(k) is the intensity of the pixels,
G(i), G(k) is the absolute values of brightness gradients,
A(i, k) is the angle between the directions of the gradients.
The color images were converted into gray images using:
I = 0.3 × R + 0.59 × G + 0.11 × B
2.2 Similarity measures
The histograms/co-occurrence matrices that represent different images have to be compared and
their difference be expressed by a single number if possible. This can be achieved with the help of
a metric defined to measure the ”distance” between two histograms.
As a similarity measure of the classes (topics), the following distances were tested:
• canberra metrics
n
X |xi − yi |
dcanberra (X, Y ) = ,
i=1
xi + yi
• manhattan metrics
n
X
dmanhattan (X, Y ) = |xi − yi |
i=1
• and absolute value of correlation coefficient
E(XY ) − E(X)E(Y )
dcorrelation (X, Y ) = p p ,
E(X 2 ) − E 2 (X)E(Y 2 ) − E 2 (Y )
where E is the expected value operator, X = (x1 , x2 , . . . , xn ) and Y = (y1 , y2 , . . . , yn ) are feature
vectors.
Some visual topics could contain several images as a query, so we tried two types of distance
measures to the group of images:
dcentroid (X, Y ) = mean(xi )
and minimum distances
dmin (X, Y ) = min(xi ).
Additionally, color control was used during the retrieval: if the query contained color/gray
images, corresponding images were moved at the top of the results.
3 Experiments
Initially, due to descriptors’ normalization, original images were scaled down:
• proportional rescaling to 320 pixels width;
• rescaling to 150 × 150 pixels;
• rescaling to 300 × 300 pixels.
After that different types of descriptors were calculated and stored into binary files (Table 1).
In our experiments we have chosen 5–7 bins for every dimension of the co-occurrence matrix.
Average time for generation one image descriptor was about half of a second.
Descriptor Number of elements Database of descriptors
GGAD 1944 570Kb
GGGD 2907 852Kb
IIGGAD 54432 16Gb
Table 1: Types of descriptors used in our experiments
Various types of descriptors and metrics in combination with image rescaling, produced 42 runs.
In our opinion, the best retrieval results on such a common database were achieved with
IIGGAD descriptors and minimum distance on manhattan metrics (Fig. 2). Thus only two runs
were submitted (Table 2)
Figure 2: A snapshot of the retrieval result (top 15 results, visual topic # 5)
Run title Description
UIIPMinsk 45 8 1244619573674 300×300 images, IIGGAD, minimum manhattan
UIIPMinsk 45 8 1244704201053 150×150 images, IIGGAD, minimum manhattan
Table 2: Two runs submitted by UIIPMinsk group
4 Results and discussion
As it was our first experience at this competition, we erroneously did not run mixed and semantic
topics. Thus, only 9 out of 25 topics were submitted. According to a general number of relevant
results our two runs can by ranked 3rd and 6th respectively out of 13 visual runs.
Generalized co-occurrence matrices are intended to find visually similar images. As some topics
can by semantically the same but visually absolutely different, employment of textual information
has more potential. Despite all this, generalized co-occurrence matrices showed very promising
results as similarity measures of medical images.
Acknowledgments
This work was partially supported by the ISTC grant B-1682.
References
[1] Henning Müller, Jayashree Kalpathy-Cramer, Ivan Eggel, Steven Bedrick, Saı̈d Radhouani,
Brian Bakke, Charles Kahn Jr., William Hersh, Overview of the CLEF 2009 medical image
retrieval track, CLEF working notes 2009, Corfu, Greece, 2009.
[2] Kovalev V. and Petrou M. Multidimensional Co-occurrence Matrices for Object Recognition
and Matching, Graphical Models and Image Processing, Vol. 58, No. 3, May, pp. 187-197,
1996.
[3] Haralick R.M., Shanmugan K.S., Dinstein I. Textural features for image classification. IEEE
Transactions on Systems. Man and Cybernetics. SMC-3 (6) (1973). pp. 610-621.
[4] Kovalev V.A. and Kruggel F. Retrieving 3D MRI Brain Images, In: Information Retrieval and
Exploration in Large Medical Image Collections (VISIM’01/MICCAI’01), Oct 2001, Utrecht,
The Netherlands, Smeulders A.W.M. (Ed.), pp. 53-56, 2001.
[5] Kovalev V. and Volmer S. Color Co-occurrence Descriptors for Querying-by-Example, Int.
Conf. on Multimedia Modelling, Oct. 12-15, Lausanne, Switzerland, IEEE Comp. Society
Press, pp. 32-38, 1998.