A MULTIMEDIA DATABASE MANAGEMENT SYSTEM FOR

MEDICAL DATA

Liana Stanescu, Dumitru Burdescu, Marius Brezovan and Cosmin Stoica

University of Craiova, Faculty of Automation, Computers and Electronics, Romania

Keywords: Database management system, content-based visual query, color and texture features, medical data.

Abstract: The paper presents a relational multimedia database management system for managing visual and

alphanumerical information from the medical domain. The MMDBMS offers numerical and char data types

for alphanumerical information, and Image data type used for storing in an original manner the visual

information. An Image data type stores the image in a binary manner, its type, its dimensions and

information about color and texture that are automatically extracted. This information will be used for

content-based visual query process. The color information is represented by the color histogram quantified

to 166 colors in the HSV color space. The texture information is represented by a vector with 12 values

resulted from the method that uses Gabor filters for texture detection. This DBMS brings up as an element

of originality the visual interface for building content-based image query using color and texture

characteristics and a modified Select command. This MMDBMS, implemented using Java technologies is

platform independent and can be easily used by the medical personnel.

1 INTRODUCTION

It is a fact that the number of images produced and

used in the medical domain is exponentially

increasing due to its development. More than 10,000

images are produced daily in large hospitals. It is

considered that images from cardiology area are the

most numerous, closely followed by the endoscope

images. The use at large scale of the DICOM

standard for image communication allows medical

equipment to create a file containing both

information about a patient (name, diagnostic,

consultation data, doctor’s name etc) and one or

many images (Muller et al, 2004, Muller et al,

2005).

That is why managing and querying these large

collections of images and alphanumerical

information are difficult tasks. The most efficient

way for solving these problems is to have a

multimedia database management system with the

following characteristics (Kalipsiz, 2000,

Khoshafian and Baker, 1996, Lu, 1999):

− support for multimedia data types;

− possibility to manage a large number of

multimedia objects;

− hierarchical storage management;

− conventional database capabilities;

− possibility to retrieve multimedia information.

Concerning the last characteristic, multimedia

data (including visual information) need access

methods much more complex that the simple text

based query, or exact matching queries. The content

based query takes into consideration attributes or

characteristics extracted from multimedia

information. The technique applied for images is

called content-based visual retrieval and was

intensely studied after 1990 (Kalipsiz, 2000,

Khoshafian and Baker, 1996). The easiest way for

implementation the content-based visual query is

using primitive characteristics as color and texture

(Del Bimbo, 2001, Faloutsos, 2005, Smith, 1997).

The directions where content based retrieval is

needed in medical multimedia databases were fully

specified. They are (Muller at al, 2004, Muller et al,

2005): diagnostic aid, medical teaching and medical

research.

In order to manage content based retrieval for

medical image collections a series of applications

that use traditional DBMS (MS SQL Server, My

SQL, Interbase) have been implemented. The

complete solution is provided by Oracle - the Oracle

10g database server and Intermedia tool can manage

all kind of multimedia data, including DICOM files.

This kind of solution involves high costs for buying

the database server and for designing and

371

Stanescu L., Burdescu D., Brezovan M. and Stoica C. (2007).

A MULTIMEDIA DATABASE MANAGEMENT SYSTEM FOR MEDICAL DATA.

In Proceedings of the Second International Conference on Signal Processing and Multimedia Applications, pages 361-364

DOI: 10.5220/0002137203610364

 SciTePress

implementing complex applications for content

based query (Chigrik, 2007, Kratochvil, 2005,

Oracle, 2005).

This paper presents a DBMS based on the

relational model, which is less expensive. The

DBMS is platform independent and can easily

manage medium sized image collections and

alphanumerical information from the medical

domain. It has a visual interface for building content

based retrieval using color and texture

characteristics and can be easily used by any person

working in this area, even if he does not have

advanced knowledge in using the computer.

2 DATA ORGANIZATION IN

THE DATABASE

MANAGEMENT SYSTEM

In this section it is described how the information is

organized in the DBMS.

For each database a new folder is created with

the same name as the name provided by the user.

This folder is in the Database folder which is also

created in the application main folder when the

software is installed. Each table in the database is

represented by a file with “.tbl” extension stored in

corresponding database folder. The file has two

components:

− A header – is created in the design phase

− Data area – is updated when executing

traditional operations of Insert, Update and

Delete

The header structure is made of:

- the number of records in table header (in the

header there will be a record for each column in

table, a record for primary key, and a record for each

external key defined in the table).

- the size of each record from the header (a

header record has information about a column of the

table: name, type, length – in case of char strings; it

can also store information about primary or foreign

key/keys).

- the header records.

The DBMS has three types of data: int, char

(fixed length strings) and image.

For the Image data type, in the data area the

following attributes are stored:

− Image type (bmp, jpg or gif);

− Image height and width

− Number of bytes needed to store the image

− The image in binary;

− 166 integer values, representing the color

histogram.

− 12 integer values, representing texture vector.

A series of methods frequently used in the

medical domain are also implemented for the Image

data type: rotating, zooming, pseudo-colors, the

similarity distance between two images, a thumbnail

representation, etc. The Image data type is in

compliance with the SQL/MM standard (SQLMM,

2001).

There are described below the methods used for

extracting color and texture information from an

image and the reason why they where chosen.

The color is the visual feature that is

immediately perceived on an image. The color space

used for representing color information in an image

has a great importance in content-based image

query, so this direction of research was intensely

studied (Del Bimbo, 2001).

There is no color system that it is universally

used, because the notion of color can be modeled

and interpreted in different ways (Gevers, 2004).

There were created several color spaces, for

different purposes (Gevers, 2004). The color

systems were studied taking into consideration

different criteria imposed by content-based visual

query (Gevers and Smeulders, 1999): the

independence of the imaging device; perceptual

uniformity; linear transformation; intuitive for user;

robust against imaging conditions (invariant to a

change in viewing direction, invariant to a change in

object geometry, invariant to a change in direction

and intensity of the illumination and invariant to a

change in the spectral power distribution of the

illumination).

It was proved that the HSV color system has the

following properties (Gevers, 2004): it is close to the

human perception of colors; it is intuitive; it is

invariant to illumination intensity and camera

direction. The studies made on nature and medical

images have shown that in the case of the HSV,

RGB, l1l2l3 and CieLuv color systems, the HSV

color space produces the best results in content

based retrieval (Gevers and Smeulders, 1999,

Gevers, 2004, Smith, 1997, Stanescu et al, 2006).

The operation of color system quantization is

needed in order to reduce the number of colors used

in content-based visual query. The quantization of

the HSV color space to 166 colors, solution

proposed by J.R. Smith, is the idea used in this

MMDBMS (Smith, 1997), having as result the color

histogram which is memorized together with the

image in the data area of the file.

Together with color, texture is a powerful

characteristic of an image, existent in nature and

medical images, where a disease can be indicated by

changes in the color and texture of a tissue. A series

of methods have been studied to extract texture

features (Del Bimbo, 2001), but there is not a certain

SIGMAP 2007 - International Conference on Signal Processing and Multimedia Applications

372

method that can be considered the most appropriate,

this depending on the application and the type of

images taken into account.

Among the most representative methods of

texture detection is the one that uses Gabor filters.

This is why it is used in this MMDBMS for

determining the texture vector.

Starting from the representation of the HSV

color space, the color can be represented in complex

(Palm et al, 2000, Zhang et al, 2000).

The affix of any point from the cone base can be

computed as:

= S (cos H + i sin H)

(1)

Therefore, the saturation is interpreted as the

magnitude and the hue as the phase of the complex

value b; the value channel is not included. The

advantages of this representation of complex color

are: the simplicity due to the fact that the color is

now a scalar and not a vector and the combination

between channels is done before filtering.

So, the color can be represented in complex

(Palm et al, 2000, Zhang et al, 2000):

y)iH(x,

ey)S(x,y)b(x, ⋅=

(2)

The computation of the Gabor characteristics for

the image represented in the HS-complex space is

similar to the one for the monochromatic Gabor

characteristics, because the combination of color

channels is done before filtering:

(3)

The Gabor characteristics vector is created using

the value

computed for 3 scales and 4

orientations:

f = (C

0,0

, C

0,1

, … C

2,3

)

(4)

These 12 characteristics are also stored in the

Image type field.

3 CONTENT-BASED VISUAL

QUERY

The presented multimedia database management

system offers the possibility to build the content-

based visual query, in an easy manner, at the image

level. The elements of the window which permit

content-based retrieval are:

− Similar With – opens the window for choosing

the query image

− Select – permits to choose the field (or fields)

that will be presented in the results of the

query

− From – it is one of the tables in database, that

will be used for the query

− Where – the image type column used for

content-based image query

− Features – it is chosen the characteristic used

for content based visual query – color, texture

or a combination of them

− Threshold – it is chosen a threshold of

accepted similarity between query image and

target image. An image with a similarity under

that threshold will not be added into the

resulted query images

− Maximum images – specify the maximum

number of images returned by the query

The similarity between the texture characteristics

of the query image Q and the target image T is

defined by the metric (Palm et al, 2000, Zhang et al,

2000):

The intersection of the histograms is used for

computing the similarity between the query image Q

and the target image T for color feature (Smith,

1997). If both distances are used in the query, the

total distance is arithmetical average between the

distances.

When building the query, it is actually built a

modified SQL Select command, adapted for content-

based image query. This command has the following

structure:

Select patients.diagnosis, patients.img

From Patients where Patients.img

Similar with Query Image (method:

color, max.images 5)

This modified Select command specifies that the

results are obtained from Patients table, taking into

consideration the values from diagnosis field, the

images similar with the query image for color

characteristic, and there will be 5 resulting images.

In the resulting set it is also presented the distance of

the dissimilarity between query image and target

image. In fact this modified command is very

suggestive for the users (medical personnel).

4 EXPERIMENTS

The MMDBMS was tested using a system with the

following characteristics: AMD Athlon 3000+

processor, 1 GB RAM Memory, 2x150Gb RAID

HDD, Windows XP Professional operating system.

Some of the preliminary results are presented in

table 1. In this phase it was measured the time for

)(d ,),( T)(Q,D

ffwhereTQd −==

∑

(5)

v)}))(u,

Mv){P(u,

FFT((

ϕϕ

⋅

−

∑

A MULTIMEDIA DATABASE MANAGEMENT SYSTEM FOR MEDICAL DATA

373

displaying records (there is only one field of Image

type in the records), and the execution time of the

content-based visual query taking into consideration

the color feature, the texture feature and their

combination. The execution time for these three

types of queries is good. The time necessary to

display all the records is higher, because the kernel

must display the binary image and all attached

information. One of the solutions that can be used in

order to reduce the time needed for displaying the

images, is to organize the display function in pages

(for example 100 records on each page). The

indexing solutions that will be implemented will

improve significantly the presented values.

Table 1: The experimental results.

Number

records

Display

time

Query

time -colo

Query

time -

texture

Query

time -

both

100 4.516 0.350 0.328 0.450

1000 15.281 0.891 0.750 1.225

10000 25.563 1.500 1.480 2.673

20000 73.042 4.750 4.515 8.428

5 CONCLUSIONS

The paper presents a database management system

for managing and querying visual information from

an important domain - the medical one. To

efficiently manage the images, the Image data type

is used. It stores, in an original manner, both the

image and the visual characteristics that are

automatically extracted from it (color and texture

vectors). The HSV color space, quantified to 166

colors is used for representing color information. For

detecting texture the method based on Gabor filters

is used.

The functions of the MMDBMS are:

creating/deleting databases and tables, creating

constrains (primary key, referential integrity), text

based querying and content-based visual querying

using two characteristics (color and texture). The

MMDBMS implements a new type of Select

command, adapted for this complex type of

querying.

REFERENCES

Chigrik, A.: SQL Server 2000 vs Oracle 9i,

(2007)http://www.mssqlcity.com/Articles/Compare/sq

l_server_vs_oracle.htm.

Del Bimbo, A.: Visual Information Retrieval, Morgan

Kaufmann Publishers. San Francisco USA (2001).

Faloutsos, C.: Searching Multimedia Databases by

Content. Springer (2005).

Gevers, T., Smeulders, W.M.: Color-based object

recognition. Pattern Recognition. (1999) 32, 453-464.

Gevers, T.: Image Search Engines: An Overview.

Emerging Topics in Computer Vision. Prentice Hall

(2004).

Kalipsiz, O.: Multimedia databases. Proceedings of the

IEEE International Conference on Information

Visualization (2000).

Khoshafian, S., Baker, A.B.: Multimedia and Imaging

Databases. Morgan Kaufmann Publishers, Inc. San

Francisco California (1996).

Kratochvil, M.: The Move to Store Images In the Database

(2005).

http://www.oracle.com/technology/products/intermedia/pd

f/why_images_in_database.pdf.

Lu, G.: Multimedia Database Management Systems.

Artech House Publishers (1999).

Muller, H., Michoux, N., Bandon, D., Geissbuhler, A.: A

Review of Content_based Image Retrieval Systems in

Medical Application – Clinical Benefits and Future

Directions. Int J Med Inform. (2004)73.

Muller, H., Rosset, A., Garcia, A., Vallee, J.P.,

Geissbuhler, A.: Benefits of Content-based Visual

Data Access in Radiology. Radio Graphics. (2005)

25,849-858.

Oracle, Oracle InterMedia: Managing Multimedia Content

(2005).

http://www.oracle.com/technology/products/intermedia/pd

f/10gr2_collateral/imedia_twp_10gr2.pdf.

Palm, C., Keysers, D., Lehmann, T., Spitzer, K.: Gabor

Filtering of Complex Hue/Saturation Images For

Color Texture Classification. Proc. 5

Joint

Conference on Onformation Science (JCIS2000) 2.

Atlantic City, USA (2000) 45-49.

Smith, J.R.: Integrated Spatial and Feature Image Systems:

Retrieval, Compression and Analysis. Ph.D. thesis,

Graduate School of Arts and Sciences. Columbia

University (1997).

SQL/MM Part 5 Still Image (2001)

http://www.wiscorp.com/2CD1R1-05-stillimage-2001-

12.pdf.

Stanescu, L., Burdescu, D.D., Ion, A., Brezovan, M.:

Content-Based Image Query on Color Feature in the

Image Databases Obtained from DICOM Files.

International Multi-Conference on Computing in the

Global Information Technology. Bucharest. Romania

(2006).

Zhang, D., Wong, A., Infrawan, M., Lu, G.: Content-

Based Image Retrieval Using Gabor Texture Features.

In: The First IEEE Pacific-Rim Conference on

Multimedia. Sydney (2000) 392-395.

SIGMAP 2007 - International Conference on Signal Processing and Multimedia Applications

374