Proposed Classification System by Using Artificial Neural Network

59 ABSTRACT The research presented in this paper was aimed to develop a recognition system for microscopic images of human tissues samples. The system should classify different types of tissues (i.e., Breast, Liver and blood cells). In this paper, co-occurrence matrix, run length matrix features combined with developed method to measure the roughness were used to extract a set of textural features in order to perform texture analysis for tissues samples. A feed forward neural network was used to classify different types of tissues according to the extracted feature vectors. For ANN training purpose the back-propagation training algorithm was used. Evaluation tests were carried on 550 tissues images. The test results indicated that the best attained success rate was around 93%. The proposed system was implemented using “visual basic.net” and all tests be done on windows operating system environment.

samples. The system classified breast tissues as malignant or not, or identifying their malignancy types. The multi-scale fractal dimension concept was used to extract a set of textural features for breast tissues samples. The box counting method was used to estimate the multi fractal dimensions. A feed forward neural network was used to classify different types of breast tissues according to the extracted fractal dimension vectors. Evaluation tests were carried on 368 breast tissues images. The test results indicated that the best attained success rate was around 97%. Arunadevi, et al [2] improved classifier for brain tumor tissue characterization. The classifier obtained 98.25% accuracy. They extended the computation of gray level co-occurrence matrix (GLCM) and Run length matrix (RLM) to a threedimensional form for feature extraction. An improved Extreme Learning Machine (ELM) classifier algorithm was explored, for training single hidden layer artificial neural network, integrating an enhanced swarm-based method in optimization of the best parameters (inputweights, bias, norm and hidden neurons), enhancing generalization and conditioning of the algorithm. Padma and Sukanesh [6] classify and segment the brain soft tissues from computed tomography images using the wavelet based dominant gray level run length feature extraction method with Support Vector machine (SVM) classifier. An average accuracy rate of above 98% was obtained using this classification and segmentation algorithm.

2.1.CO-OCCUERRENCE MATRIX
In a statistical texture analysis, texture features were computed on the basis of statistical distribution of pixel intensity at a given position relative to others in a matrix of pixels representing image. Depending on the number of taken pixels in each combination, there is first-order statistics, second-order statistics or higher-order statistics. Feature extraction based on Co-occurrence matrix is the second-order statistics that can be used to analyze image content as a texture. Figure (1) below presents an example about the formation of the Cooccurrence matrix of the gray image (4 levels) image at the distance d = 1 and the direction of 0° [7].

Figure (1): An example of GLCM formation
In this work in addition to the horizontal direction (0º), GLCM can also be formed for the direction of 45º, 90º and 135º. Co-occurrence matrix is a matrix of frequencies at which two pixels, separated by a certain vector, occur in the image. The contents of the GLCM matrix depend on the scan direction and the distance relationship between pixels. By varying the separation distance it allows to capture different texture characteristics, which will reflect important information's about the nature and extent of existing local correlation between pixels (i.e., several values were tested in this work 1,2, and 3). After counting the frequency of each possible transition between pixels values, there is still one step to take before texture measures can be calculated. The measures require that each Co-occurrence matrix cell contain not a count, but rather a probability. It is defined by P(a,b|d,θ) which expresses the probability of the couple pixels at θ direction and d interval. When θ and d is determined, P(a,bd,θ) is showed by P(a,b). Once the normalized Co-occurrence matrix has been created, various features can be computed from it. Haralick and his colleagues extracted 14 features from the Co-occurrence matrix, although in many applications only eight features are widely used, and in this work only these 8 features have been used, they are: Contrast, Energy, Norm Entropy, Homogeneity, Cluster Shade, Cluster Prominence, Inverse Difference Moment, and Maximum Probability, which are obtained using the following equations (1 to 8) respectively [8,9]: having value i and the target pixel having value j.
Contrast is the main diagonal near the moment of inertia, it measures the distribution status of the matrix elements and if there is local changes in number, also, it reflects the image clarity and texture of shadow depth. Therefore, the Contrast feature is a measure of the image contrast or the amount of local variations present in an image. (2) Energy is a gray-scale image texture measure of homogeneity changing, it reflects the distribution of image gray-scale uniformity of weight and texture. Hence it is a suitable measure for detection of disorders in textures.
Provides the pixel pair that is most predominant in the image; the maximum probability is expected high if the occurrence of the most predominant pixel pair is high.

2.2.Run Length Matrix
Run-length statistics capture the coarseness of a texture in specified directions. A run is defined as a string of consecutive pixels which have the same intensity along a specific linear orientation. Fine textures tend to contain more short runs with similar intensities, while coarse textures have more long runs with significantly different intensities.
Run length is the number of adjacent pixels that have the same intensity in a particular direction. Run-length matrix is a two-dimensional matrix where each element is the number of elements j with the intensity i, in a given direction. For example, Figure (2.a) below shows a matrix of size 4x4 pixel image with 4 levels. Figure (2.b) the corresponding Run-length matrix in the direction of 0° [7].

2.3.Roughness Feature
Roughness is a measure of the texture of a surface. It is quantified by the vertical deviations of a real surface from its ideal form. If these deviations are large, the surface is rough; if they are small the surface is smooth. Roughness is typically considered to be the indication for the degree of existing high frequency, short wavelength component of a measured surface. The surface of a rough texture presents a high number of asperities. In an image, roughness can be described as a set of quick spatial transitions with varying amplitude.  Also, different window size values were tested (i.e., 5, 7, and 9).

B. Moments Determination
In this stage, the central moments (moments about the mean) were used because they are more interesting than the moments about zero. The expression for the nth order moments about the mean is given by: Where, zi is a random variable indicating intensity, P(zi) is the histogram of the intensity levels in a region, L is the number of possible intensity levels and m is the mean (average) intensity.
The first step towards extracting the feature vectors of the roughness attribute is the determination of histogram of residue which represents the difference between the original image and smoothed image. Since the using of histogram as it is as a feature vector; then the feature size with be high such that it will increase the computation cost of the next steps in the

Neural Network
Artificial neural network models have been studied for many years in the hope of achieving human-like performance in several fields such as speech and image understanding. The networks are composed of many nonlinear computational elements operating in parallel and arranged in patterns reminiscent of biological neural networks. The network nodes, belong to adjacent layers, are connected, and their weights are typically adapted during the training phase to achieve high network performance [11].
Generally, the classification and recognition problems have been solved by traditional mathematical or statistical techniques. However, when there is large amount of data had to be processed, and it has wide dynamic range of variations, then the neural nets are capable to successfully process this data with reasonable amount of calculations.
In this research the multilayer feed forward artificial neural network had been trained to classify different kinds of tissues. For training purpose, the back-propagation algorithm was used. The architecture of the applied neural network consists of four layers: an input layer, two hidden layers and an output layer. The ANN input is 85 extracted texture features.

2.5.Proposed System
The system is composed of the following main processes as shown in Figure (3

):
A) Image loading: The type of image format used in this research is 24BMP format. Then the loaded RGB color image was transformed into gray images using the following equations: Where, R is red, G is green, and B is blue color component, Gr is the gray.
Also, the quantization level was taken (20) to quantize the intensity component, because fewer number of grey levels faster the computation when the statistics are applied.

B)
Feature extraction and analysis: Eighty Five features were extracted, divided into three categories: forty features extracted from Co-occurrence matrix, eight features for each direction (horizontally, vertically, diagonal, inverse diagonal, all directions) and thirty features based on run-length matrix, ten features for each direction (horizontal, vertical and both direction) and fifteen features for roughness measure using five order moment (five for mean, five for median and five for maximum probability).

Figure (3): The Diagram of Proposed system
In the established system, the number of input nodes is set equal to the size of extracted texture features (i.e., 85). Two hidden layer were used, the number of hidden nodes were varied to find out the best smallest number of hidden nodes required to get best classification rate. Also, the best value of learning rate was investigated during the learning phase, taking into consideration that this parameter has significant effect on the training time and accuracy.
The number of output nodes was taken 2, to represent the tissue class index in binary form.
The sigmoid function was taken: In general, to train the ANN many of the available data should be used, although it is not necessary to use them all. From the available training data a sufficient number of patterns are needed to be included in the training data set. The remaining data (i.e., 200) can be used to test the network to verify that the network can perform the desired mappings on the input vectors; which they have never been encountered during training.

3.RESULTS AND DISCUSSION
The data sets used in this study are sets of medical images taken from different sources.
The first group of images used in this research has been collected from well-known medical atlas, and the second group images were taken from Kirkuk Educational Hospital. The second group was collected by capturing pre-diagnosed images using a digital camera connected to the microscope. The images are in true color (RGB) images and of varying sizes. Table (

Table (1): Examples of image classes
The main stages of the established system are: feature extraction and recognition using artificial neural network. The feature extraction unit has two parameters, namely; Cooccurrence jump step and roughness window size. The parameters of this stage have considerable effects on the discriminating power of extracted feature vector. In the recognition unit the artificial neural network has several parameters, namely; learning rate, number of hidden layer, and number of hidden nodes each one plays important role to achieve good recognition rate.

The Effect of Co-occurrence Matrix Jump Step
When using Co-occurrence matrix not only one jump step is adopted, several jump steps are tested to analyze each area. It is important to find the suitable value for it. In this work, the tested values are (1, 2, and 3). The assignment of jump step value is very important to get more accurate texture analysis for the image. Table (2) shows the effect of using different jump steps on system success rate.

A. The Effect of Roughness Window Size
In this set of tests the system success rate was determined for different window sizes to estimate the image surface roughness. Table III illustrates the effect of window size parameter on system success rate.

B. The Effect of Number of Hidden Layers
When using ANN, not only single architecture is adopted, several architectures were tested. In this research project, the effect of number of hidden layers had been investigated to define the effect of number of hidden layers on system efficiency and ANN learning time.
When using single hidden layer, the best attained system success rate was (85%), while for multi hidden layers neural network the attained efficiency was increased. Also, the effect of number of hidden layers on learning time was tested, taking into consideration when adding new additional hidden layer additional time is required to train the neural network, see Table   (4) which illustrates the effect of this parameter on system success rate and learning time.

D. The Effect of Learning Rate
One of the important parameters that affect the accuracy of multi-layer feed forward network is the learning rate; it is used to control the rate of weights adjustments. If the value of learning rate is too small then the learning process takes longer time; and if it is too large then the learning rate may disrupt all previous knowledge. There is no analytical method for finding the optimal learning rate; it is usually optimized empirically, just by trying different values. Table (6) shows the effect of learning rate on system success rate and learning time. were tested, the test results indicated that less value give higher success rate and long learning time in comparison with those obtained when using larger values.

4.CONCLUSIONS
The use of 85 features extracted from Co-occurrence, run length matrixes, and roughness measure can be utilized to describe the textural content of various tissues.
A new idea based on taking advantage from using the histogram of residue between the original image and smoothed image to be used as indicator for roughness existence in an image. Using developed method in roughness feature extraction for textured images lead to more accurate results when combined with other traditional methods (i.e., Co-occurrence and run length) to overcome the weakness of these methods.
The established system gave better success rate (93%), when Co-occurrence jump taken 1, roughness window size taken 5, value of ANN hidden layers is set 2, the number of input nodes was 85, value of ANN first hidden nodes equal 65, value of ANN second hidden nodes equal 35, value of learning rate is set 0.2, the number of output nodes was 2, and the time required for training the neural network was 63 minutes.