Since the beginning of the COST B11 action there was a need for a computer program for texture feature calculation. Its development started in October 1998 during scientific mission of COST B11 Polish group to German Cancer Research Centre (GCRC) in Heidelberg. The name of the program MaZda is an acronym derived from 'Macierz Zdarzeń' that is Polish counterpart of the English term 'co-occurrence matrix'. Piotr Szczypiński, Marcin Kociołek and Michał Strzelecki wrote MaZda source code.
The program code was written basically in C++ language, while some feature reduction procedures were written using Delphi. IT was compiled using C++ Builder Inprise package. It works under Microsoft Windows 9x/NT/2k/XP operating systems.
MaZda (ver. 3.08, May 2002) implements procedures which allow:
- Loading image files in most popular MR scanner standards (such as Siemens, Picker, Brucker and others). Some of these procedures where extracted from NMRWin program (developed in GCRC by Michael Fridlinger), then modified and implemented in MaZda. MaZda can also load images in the form of Windows Bitmap, DICOM and unformatted grey-scale image files (raw images) with pixels intensity encoded with 8 or 16 bits. Additionally, there is an option for reading details of image acquisition protocol extracted from the image information header.
- Image normalisation. There are three options: "default" (analysis is made for original image); "+/-3 sigma" (image mean m value and standard deviation sigma is computed, then analysis is performed for grey scale range between [m-3sigma, m+3sigma]; "1%-99%" (grey scale range between 1% and 99% of cumulated image histogram is taken into consideration during analysis).
- Defining regions of interest (ROI), then analysis is performed within these regions. Up to 16 regions of any shape can be defined, they can be also edited, loaded and saved as disk files. Additionally, a histogram of defined ROI may be visualised and stored.
- Image analysis, which is computation of texture feature values within defined ROIs. The feature set (almost 300 parameters) is divided into following groups: histogram-, co-occurrence matrix-, run-length matrix-, gradient, autoregressive model- and Haar wavelet derived features. Detailed description of feature definitions can be found in user's manual of MaZda (written by Prof. A. Materka).
- Computation of feature maps that represent distribution of a given feature over an analysed image. There is possible to save/load feature maps into a special floating point file format or into Windows Bitmap file.
- Displaying image analysis reports, saving and loading reports into disk files.
- Feature reduction and selection in order to find a small subset of features that allows minimum error classification of analysed image textures. This is performed by means of two criterions: Fisher coefficient maximisation and minimisation of probability of classification error. Selected features can be transferred to B11 program for further processing and/or classification.
- Image analysis automation by means of text scripts containing MaZda language commands. Scripts allow loading analysed images and their ROI files, running the analysis and saving report files on disk.
Number of features computed by MaZda still increases. New feature groups are added according to suggestions of COST B11 project members. Also, new procedures for data processing are appended. MaZda in version 3.08 computes practically all the image texture features described in world scientific literature.
<< Back | B11 program >>