Browsing by Author "Harrar, Khaled"

Now showing 1 - 20 of 32

3D reconstruction method of the proximal femur and shape correction
(IEEE, 2014) Akkoul, S.; Hafiane, A.; Leconge, R.; Harrar, Khaled
The aim of this work is to present a 3D reconstruction method of the proximal femur shape using contours identification from pairs of 2D X-ray radiographs without any prior acknowledge. 3D personalized model was reconstructed following a processing chain of seven different steps. After localization of the 2D contours on the images and the matching points of these contours, a 3D contour is generated using an algorithm based on a mathematical model. Thus, with a reduced number of pairs of images, we reconstruct a 3D points cloud, which enables obtaining a closed 3D surface. The accuracy of our approach was evaluated by comparing the reconstruction result with the 3D CT-scan reconstruction of cadaveric proximal femur. The estimated error shows that it is possible to rebuild the proximal femur shape from a limited number of radiographs
Analyse de texture pour la caractérisation de l'architecture osseuse
(2012) Harrar, Khaled; Hamami, latifa
L'ostéoporose est une maladie caractérisée par une fragilité excessive du squelette, due à une diminution de la masse osseuse et à l'altération de la microarchitecture osseuse, ce qui expose à un risque accru de fractures. L’objectif de cet article est la caractérisation de l’architecture osseuse pour la discrimination de deux groupes de sujets (sains et ostéoporotiques) en utilisant deux méthodes. L’une repose sur l’analyse de texture (la lacunarité) et l’autre sur l’analyse histomorphométrique. Plusieurs paramètres sont calculés, ainsi, un nouvel index d’interconnectivité est développé. L’application des méthodes sur dix images IRM donne des résultats prometteurs et montre que le nouvel index d’interconnectivité est bien corrélé à la lacunarité
The box counting method for evaluate the fractal dimension in radiographic images
(2007) Harrar, Khaled; Hamami, L.
Since the end of the Seventies, following work of the French mathematician Mandelbrot, We are witnessing the true exploitation of the interest expressed by the scientific community for the fractals objects. The aim of this paper is to introduce the fractal theory by the calculation of the fractal dimension in the radiographic images. We implement for that, the box counting method for the segmentation of the images. This method will be presented, as well as a study of the effect of the change of the rang of the box sizes (rmin and rmax) on fractal dimension is carried out
Cancellous bone structure assessment using a new trabecular connectivity
(Elsevier, 2021) Harrar, Khaled
Osteoporosis is a major public health problem; it is characterized by a loss in bone connectivity, which leads to a higher risk of fracture. The objective of this article is to develop a new connectivity parameter for bone microarchitecture characterization and osteoporosis assessment. The purpose is to discriminate 164 subjects composed of 82 healthy patients (HL) and 82 osteoporotic cases (OP). The new connectivity parameter involves several new topological features. The proposed method was compared to a traditional connectivity index, and the results reveal the superiority and the outperformance of the new parameter to discriminate the two groups of subjects with an accuracy (Acc) of 71.95 % and area under curve (AUC) of 80.03 %. Moreover, clinical parameters from patients were involved in this study, and five configurations were constructed, tested, and validated on the data using the k-fold cross-validation (CV) model with several values of k. Furthermore, support vector machine (SVM) was used and various kernels (i.e., linear, quadratic, cubic, and RBF functions) were tested in this study. The objective is to look for the configuration providing the best performance in terms of separation between the two populations. Furthermore, several classifiers (logistic regression, k-nearest neighbors, boosted trees, and naïve Bayes) were tested and a combination of these classifiers was carried out using the stacking ensemble technique to improve the accuracy of the final prediction. Moreover, several studies of state-of-the-art were compared to the proposed method. The results obtained reveal that the 10-fold CV approach combining the new trabecular connectivity index and RBF function of SVM achieved the highest accuracy with Acc = 88.41 %, and AUC = 95.24 %. In addition, the proposed ensemble Meta classifier improved the accuracy of SVM and achieved a high rate with Acc = 95.12 % and AUC = 98.40 % outperforming the existing methods in the literature
Classification of surface defects on steel strip images using convolution neural network and support vector machine
(Springer, 2022) Boudiaf, Adel; Benlahmidi, Said; Harrar, Khaled; Zaghdoudi, Rachid
Quality control of the surfaces of rolled products has received wide attention due to the crucial role that these products play in the manufacture of various car bodies, planes, ships, and trains. The process of quality control has undergone remarkable development. Previously, it was based on the human eye and characterized by slowness, fatigue, and error. To overcome these problems, nowadays the quality control is based mainly on computer vision. In this context, we propose in this work to develop an intelligent recognition system of surface defects for hot-rolled steel strips images using modified AlexNet convolution neural network and support vector machine model. Furthermore, we conducted a study on the effect of layers selection on classification accuracy. We have trained and tested our classification model using a public database of Northeastern University composed of 1800 images of defects. The results showed that our classifier model can be used easily for effective screening of surface defects for hot-rolled steel strips with very a high classification accuracy up to 99.7%, using only 7% of the total extracted features for each image with activations on the fully connected layer “FC7.” In addition, we addressed through this research a comparative study between the proposed classification model and the well-known modern classification models. This study highlighted the efficiency and effectiveness of our proposed model for the classification of surface defects
Combining GLCM with LBP features for knee osteoarthritis prediction: data from the osteoarthritis initiative
(EAI, 2021) Harrar, Khaled; Messaoudene, Khadidja; Ammar, Mohammed
Knee osteoarthritis is a chronic disease that can make a person more susceptible to develop health complications. It is a significant cause of disability among adults. In advanced stages, people can die from these complications. OBJECTIVES: This paper introduces a quick and effective approach to classify knee X-ray images using LogitBoost and wavelet-based Gray Level Co-occurrence Matrix (GLCM) and Local Binary Patterns (LBP) to increase image classification accuracy and minimize training and testing time. METHODS: The proposed technique involves image enhancement followed by Haar wavelet transformation. GLCM and LBP were extracted from the transformed image and these attributes were used to differentiate the radiographs into two groups of patients composed of 100 normal subjects (KL 0) and 100 pathological cases with osteoarthritis (KL 2). The validation of the classification was carried out using the K-fold cross-validation technique with k = 10. RESULTS: The results revealed that the GLCM provided an accuracy of 77 % and the LBP approach achieved an accuracy of 82.5 %. Moreover, the combination of the two techniques LBP-GLCM improved the accuracy of the prediction with the LogitBoost model (91.16 %). Compared to other classifiers (SVM, logistic regression, and decision tree), the LogitBoost provided a low root mean square error (RMSE) of 27.5 %. CONCLUSION: In addition, the proposed method was compared to the state-of-the-art and revealed the highest accuracy in the prediction of KOA, outperforming the methods existing in the literature
A comparative study between convolutional and multilayer perceptron neural networks classification models
(2019) Bachiri, Mohamed Elssaleh; Harrar, Khaled
Image classification plays an important role in image processing, computer vision, and machine learning. This paper deals with image classification using deep learning. For this, a conventional neural network (CNN) and multilayer perceptron neural network (MLP) models were used for the classification. The two models were implemented on the MNIST dataset which was used at 100% and half of capacity, The models were trained with fixed and flexible number of epochs in two runs. CNN provided an accuracy of 98,43% with a loss of 4,44%, where MLP reached 92,80% of classification with a loss of 25,87%. Indeed, for each model, variables as number of filters, size, and activation functions were discussed. The CNN demonstrated a good performance providing high accuracy for image and also proved to be a better candidate for data applications.
Computerized diagnosis of knee osteoarthritis from x-ray images using combined texture features: Data from the osteoarthritis initiative
(Wiley-Blackwell, 2024) Messaoudene, Khadidja; Harrar, Khaled
The prevalence of knee osteoarthritis (KOA) cases has witnessed a significant increase on a global scale in recent years, emphasizing the need for automated diagnostic computer systems to aid in early-stage osteoarthritis (OA) diagnosis. The accurate characterization of knee KOA stages through feature extraction poses significant research challenges due to the complexity of identifying relevant attributes. In this study, the development of a KOA diagnostic system is presented, leveraging a combination of Gabor, and Tamura parameters using the Canonical Correlation Analysis algorithm. Two feature selection algorithms, namely Principal Component Analysis and Relief, were employed for KOA classification. Furthermore, various classifiers, including K-Nearest Neighbors, AdaBoost, Bagging, and Random Forest, were used to assess the proposed feature extraction approach. The diagnostic system was assessed using a dataset comprising 688 x-ray images sourced from the OA initiative (OAI) dataset, consisting of 344 images from healthy subjects (Grade 0) and 344 images from pathological patients (Grade 2). To mitigate overfitting, a 10-fold cross-validation method was utilized. The experimental results indicate that the combination of Tamura and Gabor parameters with the Random Forest classifier achieved remarkable performance in KOA diagnosis, yielding an accuracy of 94.59%, and an area under the curve of 98.3%. Notably, the combined Gabor and Tamura models exhibited superior performance compared to individual models, as well as existing techniques reported in the literature.
Detection of knee osteoarthritis based on wavelet and random forest model
(2021) Messaoudene, Khadidja; Harrar, Khaled
The most recurrent kind of osteoarthritis is Knee osteoarthritis (KOA). Doctors encounter difficulties for a precise diagnosis through its features and to the naked eye. In this paper, we propose a new approach for the classification of KOA by combining the discrete wavelet decomposition (DWT) and random forest classifier from knee X-ray images. A total of 50 images from patients suffering or not from osteoarthritis were used in this study. The suggested technique includes image enhancement using the Gaussian filter followed by Haar wavelet transform. Five texture features namely, contrast, entropy, correlation, energy, and homogeneity were extracted from the transformed image, and these attributes were used to differentiate the radiographs into two groups: normal (KL 0) or affected with osteoarthritis (KL2). Four classifiers including random forest, SVM, RNN, and Naïve Bayes were tested and compared. The results obtained reveal that random forest achieved the highest performance in terms of accuracy (ACC = 88%) on X-Ray images of the Osteoarthritis Initiative (OAI) dataset.
Electronique des impulsions destiné aux étudiants licence 3éme année electronique
(Université M'hamed Bougara de Boumerdès : Faculté de technologie, 2021) Harrar, Khaled
A Fast and Efficient Approach for Image Compression Using Curvelet Transform
(Springer, 2018) Inouri, Lynda; Azni, Mohamed; Khireddine, Abdelkrim; Harrar, Khaled; Tighidet, Soraya
In this paper a novel image compression technique using features of wavelet and curvelet transforms is proposed to improve efficiency and compression performance. Indeed, the curvelet transform is one of the recently developed multiscale transforms which is especially designed to represent efficiently curves and edges in an image. In the proposed method, the compression algorithm involves the Haar wavelet transform to decompose the image into four frequency sub-bands. The lowest frequency sub-band coefficients are processed using Set Partitioning In Hierarchical Trees (SPIHT) encoding. Meanwhile, Fast Discrete Curvelet Transform (FDCT) is applied to the remaining frequency sub-bands. The FDCT output coefficients are then quantized according to the sub-band they belong to. The lowest frequency FDCT output coefficients are quantized using Differential Pulse Code Modulation, the medium frequency coefficients are processed using SPIHT, whereas the high frequency coefficients are removed. Experimental results demonstrate that our method provides high performance for edge detection compared to existing techniques particularly for images with abrupt changes. In addition, this new image coding and decoding approach is powerful in terms of computation time. Moreover, the proposed method reveals significant improvement in compression ratio and decoded peak-signal-to-noise-ratio.
Feature extraction using CNN for peripheral blood cells recognition
(European Alliance for Innovation, 2022) Ammar, Mohammed; Daho, Mostafa El Habib; Harrar, Khaled; Laidi, Amel
INTRODUCTION: The diagnosis of hematological diseases is based on the morphological differentiation of the peripheral blood cell types. OBJECTIVES: In this work, a hybrid model based on CNN features extraction and machine learning classifiers were proposed to improve peripheral blood cell image classification. METHODS: At first, a CNN model composed of four convolution layers and three fully connected layers was proposed. Second, the features from the deeper layers of the CNN classifier were extracted. Third, several models were trained and tested on the data. Moreover, a combination of CNN with traditional machine learning classifiers was carried out. This includes CNN_KNN, CNN_SVM (Linear), CNN_SVM (RBF), and CNN_AdaboostM1. The proposed methods were validated on two datasets. We have used a public dataset containing 12444 images with four types of leukocytes to find the best optimizer function(eosinophil, lymphocyte, monocyte, and neutrophil images). The second dataset contains 17,092 images divided into eight groups: lymphocytes, neutrophils, monocytes). the second public dataset was used to find the best combination of CNN and the machine learning algorithms. the dataset containing 17,092 images: lymphocytes, neutrophils, monocytes, eosinophils, basophils, immature granulocytes, erythroblasts, and platelets. RESULTS: The results reveal that CNN combined with AdaBoost decision tree classifier provided the best performance in terms of cells recognition with an accuracy of 88.8%, demonstrating the performance of the proposed approach. CONCLUSION: The obtained results show that the proposed system can be used in clinical practice
Fractal analysis of bone radiographs correlated with histomorphometry
(2011) Harrar, Khaled; Hamami, Latifa; Jennane, Rachid
The bone fragility in osteoporosis is multifactorial and complex. At present, fracture risk prediction in the individual patient relies chiefly on bone mineral density (BMD) measurements. However, many lines of evidence indicate that the decreased bone strength…
The fractal dimension correlated to the bone mineral density
(ACM Digital Library, 2008) Harrar, Khaled; Hamami, Latifa
Osteoporosis is a condition of decreased bone mass. This leads to fragile bones which are at an increased risk for fractures, more often, it affects postmenopausal women. In this paper we propose a study of osteoporosis with the fractal dimension. After an introduction to the theory and fractal dimension, we use the box counting method for the segmentation of radiographic images, the study of the influence of range size boxes on the fractal dimension will be investigated, and the correlation between a reference dimension and bone mineral density. Other imaging techniques will be given in order to see the results of the application of the method on these types of images
A hybrid LBP-HOG model and naive Bayes classifier for knee osteoarthritis detection: data from the osteoarthritis initiative
(Springer, 2022) Messaoudene, Khadidja; Harrar, Khaled
Knee OsteoArthritis (KOA) is a disease characterized by a degeneration of cartilage and the underlying bone. It does not evolve uniformly; it can stay silent for a long time and can quickly intensify for several months or weeks. For this reason, it is necessary to develop an automatic system for diagnosis and reduce the subjectivity in the detection of the disease. In this paper, we present a method for detecting knee osteoarthritis based on the combination of histograms of oriented gradient (HOG) and local binary pattern (LBP). Four classifiers including KNN, SVM, Adaboost, and Naïve Bayes were tested and compared for the prediction of the illness. A total of 620 X-Ray images were analyzed, composed of 310 images from healthy subjects (Grade 0), and 310 images from pathological patients (Grade 2). The results obtained reveal that Naïve Bayes achieved the highest performance in terms of accuracy (ACC = 91%) on the Osteoarthritis Initiative (OAI) dataset. The fusion of HOG and LBP features in KOA classification outperforms the use of either feature alone and the existing methods in the literature
An interconnectivity index for osteoporosis assessment using X-Ray images
(2013) Harrar, Khaled; Hamami, L.
Oriented fractal analysis for improved bone microarchitecture characterization
(Elsevier, 2017) Harrar, Khaled; Jennane, Rachid; Zaouchi, Karima; Janvier, Thomas; Toumi, Hechmi; Lespessailles, Eric
Osteoporosis assessment using Multilayer Perceptron neural networks
(IEEE, 2012) Harrar, Khaled; Hamami, L.; Akkoul, S.; Lespessailles, E.
The objective of this paper is to investigate the effectiveness of a Multilayer Perceptron (MLP) to discriminate subjects with and without osteoporosis using a set of five parameters characterizing the quality of the bone structure. These parameters include Age, Bone mineral content (BMC), Bone mineral density (BMD), fractal Hurst exponent (Hmean) and coocurrence texture feature (CoEn). The purpose of the study is to detect the potential usefulness of the combination of different features to increase the classification rate of 2 populations composed of osteporotic patients and control subjects. k-fold Cross Validation (CV) was used in order to assess the accuracy and reliability of the neural network validation. Compared to other methods MLP-based analysis provides an accurate and reliable platform for osteoporosis prediction. Moreover, the results show that the combination of the five features provides better performance in terms of discrimination of the subjects
Piecewise whittle estimator for trabecular bone radiograph characterization
(Elsevier, 2013) Harrar, Khaled; Hamami, Latifa; Lespessailles, Eric; Jennane, Rachid
Quantification de la Porosité par Analyse des Images Osseuses pour la Détection de l'Ostéoporose
(2010) Harrar, Khaled; Hamami, L.
L'ostéoporose est une maladie caractérisée par la raréfaction de la masse osseuse et la détérioration de la micro-architecture du tissu osseux, qui entraînent une fragilité osseuse accrue et, par conséquent, une augmentation du risque de fracture. L'objectif de cet article est de quantifier la porosité des images radiographiques, afin de pouvoir détecter l'ostéoporose. Deux méthodes sont utilisées pour l'analyse des images osseuses, la lacunarité et le star volume. La première méthode est basée sur les fractals et la seconde sur l'évaluation de l'espace médullaire. Les résultats montrent une corrélation entre les paramètres architecturaux calculés et les taux de densités minérales osseuses (DMO), ce dernier paramètre qui est le plus utilisé en routine clinique pour la détection de l'ostéoporose