Mohd Yusoff Mashor

An automated cervical pre-cancerous diagnostic system

OBJECTIVE This paper proposes to develop an automated diagnostic system for cervical pre-cancerous. METHODS AND DATA SAMPLES: The proposed automated diagnostic system consists of two parts; an automatic feature extraction and an intelligent diagnostic. In the automatic feature extraction, the system automatically extracts four cervical cells features (i.e. nucleus size, nucleus grey level, cytoplasm size and cytoplasm grey level). A new features extraction algorithm called region-growing-based features extraction (RGBFE) is proposed to extract the cervical cells features. The extracted features will then be fed as input data to the intelligent diagnostic part. A new artificial neural network (ANN) architecture called hierarchical hybrid multilayered perceptron (H(2)MLP) network is proposed to predict the cervical pre-cancerous stage into three classes, namely normal, low grade intra-epithelial squamous lesion (LSIL) and high grade intra-epithelial squamous lesion (HSIL). We empirically assess the capability of the proposed diagnostic system using 550 reported cases (211 normal cases, 143 LSIL cases and 196 HSIL cases). RESULTS For evaluation of the automatic feature extraction performance, correlation test approach was used to determine the capability of the RGBFE algorithm as compared to manual extraction by cytotechnologist. The manual extraction of size was recorded in micrometer while the automatic extraction of size was recorded in number of pixels. Region color was recorded in mean of grey level value for both manual and automatic extraction. The results show that the estimated size and mean of grey level have strong linear relationship (correlation test more than 0.8) with those extracted manually by cytotechnologist. Hence, the size of nucleus, size of cytoplasm and grey level of cytoplasm created very strong linear relationship with correlation test more than 0.95 (approaching one). For the intelligent diagnostic, the performance of the H(2)MLP network was compared with three standard ANNs (i.e. multilayered perceptron (MLP), radial basis function (RBF) and hybrid multilayered perceptron (HMLP)). The performance was done based on accuracy, sensitivity, specificity, false negative and false positive. The H(2)MLP network performed the best diagnostic performance as compared to other ANNs. It was able to achieve 97.50% accuracy, 100% specificity and 96.67% sensitivity. The false negative and false positive were 1.33% and 3.00%, respectively. CONCLUSIONS This project has successfully developed an automatic diagnostic system for cervical pre-cancerous. This study has also successfully proposed one image processing technique namely the RGBFE algorithm for automatic feature extraction process and a new ANN architecture namely the H(2)MLP network for better diagnostic performance.

Blood Cell Image Segmentation: A Review

Image processing technique involved five basic components which are image acquisition, image preprocessing, image segmentation, image post-processing and image analysis. The most critical step in image processing is the segmentation of the image. In this paper, we review some of the general segmentation methods that have found application in classification in biomedical-image processing especially in blood cell image processing. Basically, segmentation of the image divides the whole image into some unique disjoint regions. The fact that the segmented image should retain maximum useful information and discard unwanted information makes the whole process critical.

Detection of mycobacterium tuberculosis in Ziehl-Neelsen stained tissue images using Zernike moments and hybrid multilayered perceptron network

Conventional clinical diagnosis of tuberculosis disease such as manual screening by microbiologist are tedious, laborious and time consuming. Therefore, more research has been carried out to develop technologies that able to automate the detection process. This paper presents an automated approach to tuberculosis bacilli detection in tissue section. The proposed approach employs image processing technique and neural network for the segmentation and detection of tuberculosis bacilli. First, images of tuberculosis bacilli in tissue samples are captured using light microscope after stained with Ziehl-Neelsen staining method. Then colour image segmentation using moving k-mean clustering is used to extract tuberculosis bacilli from the tissue image. Two colour spaces, RGB and C-Y colour, were utilised in order to improve the quality of segmentation and robust against various staining condition. Next, geometrical features of Zernike moments are calculated. From these features, the best features that could detect tuberculosis bacilli with higher accuracy were selected using hybrid multilayered perceptron network. Experimental results demonstrate that the proposed method is efficient and accurate to detect the tubercle bacilli in tissue.

Segmentation based approach for detection of malaria parasites using moving k-means clustering

Recent progress based on microscopic imaging has given significant contribution in diagnosis of malaria infection based on blood images. Due to the requirement of prompt and accurate diagnosis of malaria, the current study has proposed an unsupervised colour image segmentation of malaria parasites using moving k-means (MKM) clustering algorithm. It has been applied on malaria images of P. vivax species. The proposed segmentation method provides a basic step for detection of the presence of malaria parasites in thin blood smears. With the aim of obtaining the fully segmented red blood cells infected with malaria parasites, the malaria images will firstly enhanced by using the partial contrast stretching technique. Then, the MKM clustering algorithm has been applied on the saturation and intensity components of HSI (hue, saturation, intensity) colour space for segmenting the infected cell from the background. After that, the segmented images have been processed using median filter and seeded region growing area extraction algorithms for smoothing the image and removing any unwanted regions from the image, respectively. Finally, the holes inside the infected cell are filled by applying region filling based on morphological reconstruction algorithm. The proposed segmentation method has been analyzed using 100 malaria images which consist of the trophozoite and gametocyte stages. Overall, the results indicate that MKM clustering that has been performed on saturation component image has produced the best segmentation performance with segmentation accuracy of 99.49% compared to the intensity component image with segmentation accuracy of 98.89%.

Nucleus segmentation technique for acute Leukemia

Leukemia is a disease that affects blood forming cells in the body. Early detection of the disease is necessary for proper treatment management. Abnormal white blood cells or blasts play important role for hematologists in their diagnostic process. Digital image processing technique could help them in their analysis and diagnosis by enhancing the visibility of the interested features of the WBC. In this paper, a global contrast stretching (GCS) and segmentation based on HSI (Hue, Saturation, Intensity) color space will be used to improve the image quality. Image enhancement is very important to increase the visual aspect of blast cells. The results show that the proposed image enhancement procedure is useful to extract the nucleus region in WBC images sample by using the same threshold value, for both ALL and AML images.

Improving colour image segmentation on acute myelogenous leukaemia images using contrast enhancement techniques

Contrast enhancement and image segmentation play an important process in most medical image analysis tasks. One of the main tasks is the analyzing of white blood cells (WBC) where the WBC composition reveals important diagnostic information of a patient. This paper presents a two phase methodology in order to obtain a fully segmented abnormal white blood cell (blast) and nucleus in acute leukaemia images. In the first phase, the three contrast enhancement techniques which are partial contrast, bright stretching and dark stretching were used to improve the image quality. Contrast enhancement techniques enhanced the area of interest of acute leukaemia for easing the segmentation process. In the second phase, image segmentation based on HSI (Hue, Saturation, Intensity) colour space is proposed. The proposed technique helps to improve the image visibility and has successfully segmented the acute leukaemia images into two main components: blast and nucleus. The combination between contrast enhancements and image segmentation has good effect on improving the accuracy of segmentation. Hence, information gain from the resultant images would become useful for haematologists to further analysis the types of acute leukaemia.

Unsupervised colour segmentation of white blood cell for acute leukaemia images

Colour image segmentation has becoming more popular for computer vision due to its important process in most medical analysis tasks. One of the main tasks is the segmentation of white blood cell (WBC) where the WBC composition reveals important diagnostic information of a patient. In this paper, the combination between linear contrast technique and colour segmentation based on HSI (Hue, Saturation, Intensity) colour space were used in order to obtain a fully segmented abnormal WBC and nucleus of acute leukaemia images. The unsupervised segmentation technique namely k-means clustering algorithm is used to ease the segmentation process. By implementing the proposed segmentation technique, the fully segmented WBC which consists of cytoplasm and nucleus regions can be achieved by using the combination of linear contrast technique and segmentation based on H component image. Meanwhile, the fully segmented nucleus can be obtained by applying the segmentation based on S component image. The combinations between linear contrast technique and segmentation based on HSI colour space have produced a better effect on improving the accuracy of WBC segmentation with segmentation accuracies of 99.02% and 99.05% for segmented WBC and nucleus, respectively.

Comparison of acute leukemia Image segmentation using HSI and RGB color space

The Image segmentation plays an important role in computer vision and image processing areas. In this paper, the use of color segmentation for segmenting acute leukemia images is proposed. The segmentation technique segments each leukemia image into two regions: blast and background. In our approach, the segmentation is based on HSI and RGB color space. The performance comparison between the segmentation algorithms based on HSI and RGB color space is carried out to choose a better color image segmentation for blast detection. The results show that the proposed segmentation technique based on HSI has successfully segmented the acute leukemia images while preserving significant features and removing background noise.

Colour image enhancement techniques for acute Leukaemia blood cell morphological features

Image enhancement plays an important role in computer vision and image processing. In this paper, image enhancement was used to eliminate the background noise and improve the image quality for the purpose of determining the focal areas such as nucleus, Auer rode and nucleoli in acute leukaemia images. In this paper, the use of three contrast enhancement techniques for colour images using RGB components is proposed. The three contrast enhancement techniques are partial contrast, bright stretching and dark stretching. The contrast enhancement techniques enhance the morphological features of acute leukaemia images to ease the leukaemia classification between Acute Lymphoblastic Leukaemia (ALL) and Acute Myelogenous Leukaemia (AML). The results show that partial contrast is the best technique that helps to improve the image visibility while preserving the significant features of acute leukaemia images. Hence, the resultant images would become useful to Hematologists for further analysis of acute leukaemia.

Performance comparison for MLP networks using various back propagation algorithms for breast cancer diagnosis

This paper represents the performance comparison of the Multilayered Perceptron (MLP) networks using various back propagation (BP) algorithms for breast cancer diagnosis. The training algorithms used are gradient descent with momentum and adaptive learning, resilient back propagation, Quasi-Newton and Levenberg-Marquardt. The performances of these four algorithms are compared with the standard steepest descent back propagation algorithm. The current study investigates and compares the accuracy, sensitivity, specificity, false negative and false positive results of the selected four algorithms to train MLP networks. The Papinicolou image of breast cancer cells were captured via an image analyzer and thirteen morphological features were extracted to numerical scores. The feature scores are used as data sets to train the MLP network. The MLP network using the Levenberg-Marquardt algorithm displays the best performance for all the five measurement criterias (accuracy, specificity, sensitivity, true positive and true negative) at a lower number of hidden nodes.