Mojtaba Mohammadpoor | Gonabad - Academia.edu (original) (raw)

Papers by Mojtaba Mohammadpoor

Research paper thumbnail of An Leukocytes Counting System for Acute Lymphoblastic Leukemia Detection

Shiraz E-Medical Journal, 2017

Background: Today, blood disease is one of the most important causes of human deaths around the w... more Background: Today, blood disease is one of the most important causes of human deaths around the world, therefore, early diagnosis of these diseases are very important. Counting and classification of white blood cells (leukocytes) lead to identification of a variety of blood diseases such as leukemia. The aim of this research is producing a computer algorithm to count the leukocytes in order to help the hematologists detect acute lymphoblastic leukemia (ALL) in an accurate and time-efficient way. Objectives: The purpose of this research is to design and implement an intelligent software system based on image processing algorithms and fuzzy logic to analyze and accurately count blood leukocytes to identify acute lymphoblastic leukemia (ALL). Methods: The proposed image processing system consists of several sections. The first pre-processing is done to remove noise and improve image contrast. In the second step, the image is segmented using improved fuzzy clustering technique (IFCM) and active contour algorithm. In the third step, the image feature extraction and classification is done. The final step determines whether or not the image is ALL, using ANFIS neural network algorithm where its objective function is optimized by genetic algorithm. Results: Using samples of blood leukocyte images taken under the same lighting conditions let us introduce a computer aided diagnosis (CAD) system, which is empowered by fuzzy techniques for detection of all types of acute lymphoblastic cancers by 98% accuracy. Conclusions: A method for the detection and classification of blood leukocytes from the blood microscopic images using image processing techniques and fuzzy logic have been proposed. The results show that the proposed method is able to detect and classify leukocytes in an image with high accuracy.

Research paper thumbnail of A review of clinoptilolite, its photocatalytic, chemical activity, structure and properties: in time of artificial intelligence

Journal of Materials Science

Research paper thumbnail of Detection of Disc Destruction Between Lumbar Vertebrae Using MRI Images

Soft Computing, Apr 10, 2020

Most people experience low back pain at least once in their lifetime. Lumbar disc herniation is o... more Most people experience low back pain at least once in their lifetime. Lumbar disc herniation is one of the major causes of low back pain. Treatment methods for disc herniation are very diverse. Therefore, diagnosis of the exact size of herniation and its location can greatly help specialists to select the best treatment method. In this research, an automated method for diagnosing lumbar disc herniation using 130 MR images is proposed. In the proposed method, using three algorithms, namely region growing, OTSU and active contour, the intervertebral discs and their boundary were precisely separated from the background of the image. In the next step, after extracting the most significant features of the images, they are divided into healthy and unhealthy classes by SVM classifier with 89.9% accuracy. The classification accuracy was compared with classifiers KNN, Ensemble, and decision tree. Finally, it was determined the SVM classifier has the highest accuracy for the classification.

Research paper thumbnail of A Multi-objective Distribution Network Reconfiguration and Optimal Use of Distributed Generation Unites by Harmony Search Algorithm

Lecture Notes in Electrical Engineering, 2018

In this paper, the method of network reconfiguration and simultaneous use of distributed generati... more In this paper, the method of network reconfiguration and simultaneous use of distributed generation resources (DG) in optimal location and capacity is analyzed in order to minimize losses and reach the optimum level of voltage stability and voltage profile. A new method for this purpose is proposed by use of the Harmonic Search Algorithm (HSA). To investigate the effectiveness of the proposed method, the capabilities of the MATLAB software and the DPL language linked to the DIGSILENT application is used. The 33-node distribution network of IEEE standard was selected for investigation. The simulation results shows that by using the proposed method, network losses were minimized and the voltage level and voltage profiles were improved correctly.

Research paper thumbnail of Using Support Vector Machines as an Intelligent Algorithm for Detecting Seizures from EEG Signals

The Neuroscience Journal of Shefaye Khatam, 2021

Research paper thumbnail of FULL LENGTH ARTICLE OPEN ACCESS Recognition of Persian Handwritten Numbers using LBP - HOG Desc riptor

Recognition of handwritten numbers in any language is one of the most important issues attracted ... more Recognition of handwritten numbers in any language is one of the most important issues attracted many researchers and this is because of the increasing importance of the issue in today's world and in applications such as automatic detection of mail addresses or identification of numbers of bank checks. Given the great interclass diversities and much extra - class similarities between some Persian handwritten numbers, solving the Persian handwritten numbers recognition problem is difficult. In this paper, a new method for recognizing Persian handwritten numbers using a combination of HOG and LBP descriptors is provided. The proposed descriptor enjoys significant advantages of which the important one is the recording of information and features relat ing to the image (by descriptor LBP)and yet the feature extraction of the image edges (by descriptor HOG). Another advantage of the proposed descriptor is the very small length of the feature vector and the fast calculations. To evalu...

Research paper thumbnail of Recognition of Persian Handwritten Numbers using LBP-HOG Descriptor

Recognition of handwritten numbers in any language is one of the most important issues attracted ... more Recognition of handwritten numbers in any language is one of the most important issues attracted many researchers and this is because of the increasing importance of the issue in today's world and in applications such as automatic detection of mail addresses or identification of numbers of bank checks. Given the great interclass diversities and much extra-class similarities between some Persian handwritten numbers, solving the Persian handwritten numbers recognition problem is difficult. In this paper, a new method for recognizing Persian handwritten numbers using a combination of HOG and LBP descriptors is provided. The proposed descriptor enjoys significant advantages of which the important one is the recording of information and features relating to the image (by descriptor LBP)and yet the feature extraction of the image edges (by descriptor HOG). Another advantage of the proposed descriptor is the very small length of the feature vector and the fast calculations. To evaluate...

Research paper thumbnail of A deep learning algorithm to detect coronavirus (COVID-19) disease using CT images

PeerJ Computer Science, 2021

Background COVID-19 pandemic imposed a lockdown situation to the world these past months. Researc... more Background COVID-19 pandemic imposed a lockdown situation to the world these past months. Researchers and scientists around the globe faced serious efforts from its detection to its treatment. Methods Pathogenic laboratory testing is the gold standard but it is time-consuming. Lung CT-scans and X-rays are other common methods applied by researchers to detect COVID-19 positive cases. In this paper, we propose a deep learning neural network-based model as an alternative fast screening method that can be used for detecting the COVID-19 cases by analyzing CT-scans. Results Applying the proposed method on a publicly available dataset collected of positive and negative cases showed its ability on distinguishing them by analyzing each individual CT image. The effect of different parameters on the performance of the proposed model was studied and tabulated. By selecting random train and test images, the overall accuracy and ROC-AUC of the proposed model can easily exceed 95% and 90%, respec...

Research paper thumbnail of An Efficient ECC-based Authentication and Key Agreement Protocol

Public-key cryptography is commonly used to authenticate communicating entities in some networks.... more Public-key cryptography is commonly used to authenticate communicating entities in some networks. One of the key tools in this way is to use the elliptic curves cryptography (ECC) which is relatively lightweight due to its shorter key size compared to the conventional River-Shamir-Adleman (RSA) method. This paper is proposing an efficient protocol by analysing two variants of ECC-based wireless authentication protocol, namely, Aydos-Savas-Koc's wireless authentication protocol (ASK-WAP) and user authentication protocol (UAP) from various security aspects and communication concerns. We show that although UAP is able to address some of ASK-WAP vulnerabilities, it is confined to one-way communication where the authentication can only be initialized by users and not the server. In light of their limitations, we suggest several possible improvements to both ASK-WAP and UAP. The proposed solutions focus on applying encryption methods to the transmitted keys and enabling two-way commun...

Research paper thumbnail of A ground based circular synthetic aperture radar

2013 14th International Radar Symposium (IRS), 2013

Detecting on-the-ground objects is a subject of interest for some applications. Typical example i... more Detecting on-the-ground objects is a subject of interest for some applications. Typical example is foreign object detection on the airport runway. In response to this demand, a ground-based Circular Synthetic Aperture Radar (CSAR) system is proposed and explained in the paper. In the proposed CSAR, the antennas represent a circular movement trajectory. Wideband Linear Frequency (LFM) chirps were used for transmission. A simulation model for CSAR, based on the Doppler Effect between the radar and object is developed in this paper. In addition, a processing method for object detection using correlation between image data produced by simulation and experimental data is developed. The resultant of the simulated model at each point, which represents the object's behavior in an ideal and clutter-free environment, is used as a template for object detection. Simulation and experimental results demonstrate that the proposed method is well suited in detecting small objects at different po...

Research paper thumbnail of Designing a Hybrid Clustering Routing Algorithm based on Cellular Learning Automata for Optimizing Lifetime of Wireless Sensor Networks

Advances in Computer Science : an International Journal, 2015

One of the most important factors in wireless sensor networks is energy consumption, hence the li... more One of the most important factors in wireless sensor networks is energy consumption, hence the lifetime of these networks are strongly depending on remaining energy in the nodes. According to sensors placement farness and wireless communication between them, it is necessary to optimally consume the energy in these networks. In this study a hybrid approach is proposed by mixing two existing protocols, namely flat multi-hop routing and hierarchical multi-hop routing. Also by using Cellular Learning Automata (CLA) as clustering technique, the energy in the network will be managed and finally the lifetime of nodes will be increased. Mathematical simulation and analysis show a good performance of clustered hybrid model for energy saving that in compare with multi hop routing algorithm and hierarchical routing in non-clustered and clustered conditions, the lifetime increasing are %10.39, %27.36 and %5.57, %23.83 respectively.

Research paper thumbnail of A Hierarchical Classification Method for Breast Tumor Detection

Iranian Journal of Medical Physics, 2016

Introduction Breast cancer is the second cause of mortality among women. Early detection of it ca... more Introduction Breast cancer is the second cause of mortality among women. Early detection of it can enhance the chance of survival. Screening systems such as mammography cannot perfectly differentiate between patients and healthy individuals. Computer-aided diagnosis can help physicians make a more accurate diagnosis. Materials and Methods Regarding the importance of separating normal and abnormal cases in screening systems, a hierarchical classification system is defined in this paper. The proposed system is including two Adaptive Boosting (AdaBoost) classifiers, the first classifier separates the candidate images into two groups of normal and abnormal. The second classifier is applied on the abnormal group of the previous stage and divides them into benign and malignant categories. The proposed algorithm is evaluated by applying it on publicly available Mammographic Image Analysis Society (MIAS) dataset. 288 images of the database are used, including 208 normal and 80 abnormal im...

Research paper thumbnail of An Intelligent Algorithm For Detecting Minerals In Microscopic Sections Using Machine Learning

Detecting the size and shape of minerals is very important for collecting information on minerals... more Detecting the size and shape of minerals is very important for collecting information on minerals and the texture of rocks for classification and naming. Therefore, it is necessary to study the size and shape of minerals. By combining image processing and intelligent pattern recognition techniques, successful results can be obtained through detecting minerals in sections as well as their size and shape, especially in thin-section images that have reduced the third dimensional effect. In this research, a method is proposed for segmentation of thin sections and finding train minerals in them. In this regard, user input is used to select some portions of the studied mineral. These pieces of data are used as seed points to learn the neural network. Support Vector Machine (SVM) is used as a strong classifier algorithm to find the mineral samples in the whole image. The combination of colours and mineral features is used to train the SVM in order to find the minerals with high precision. ...

Research paper thumbnail of A Novel Method for Persian Handwritten Digit Recognition Using Support Vector Machine

Handwritten digit recognition has got a special role in different applications in the field of di... more Handwritten digit recognition has got a special role in different applications in the field of digital recognition including; handwritten address detection, check, and document. Persian handwritten digits classification has been facing difficulties due to different handwritten styles, inter-class similarities, and intra-class differences. In this paper, a novel method for detecting Persian handwritten digits is presented. In the proposed method, a combination of Histogram of Oriented Gradients (HOG), 4-side profiles of the digit image, and some horizontal and vertical samples was used and the dimension of the feature vector was reduced using Principal Component Analysis (PCA). The proposed method applied to the HODA database, and Support Vector Machine (SVM) was used in the classification step. Results revealed that the detection accuracy of such method has 99% accuracy with an adequate rate due to existing unacceptable samples in the database, therefore, the proposed method could ...

Research paper thumbnail of Introducing an intelligent algorithm for extraction of sand dunes from Landsat satellite imagery in terrestrial and coastal environments

Journal of Coastal Conservation

Research paper thumbnail of An Intelligent Technique for Grape Fanleaf Virus Detection

International Journal of Interactive Multimedia and Artificial Intelligence

Grapevine Fanleaf Virus (GFLV) is one of the most important viral diseases of grapes, which can d... more Grapevine Fanleaf Virus (GFLV) is one of the most important viral diseases of grapes, which can damage up to 85% of the crop, if not treated at the right time. The aim of this study is to identify infected leaves with GFLV using artificial intelligent methods using an accessible database. To do this, some pictures are taken from infected and healthy leaves of grapes and labeled by technical specialists using conventional laboratory methods. In order to provide an intelligent method for distinguishing infected leaves from healthy ones, the area of unhealthy parts of each leaf is highlighted using Fuzzy C-mean Algorithm (FCM), and then the percentages of the first two segments area are fed to a Support Vector Machines (SVM). To increase the diagnostic reliability of the system, K-fold cross validation method with k = 3 and k =5 is applied. After applying the proposed method over all images using K-fold validation technique, average confusion matrix is extracted to show the True Positive, True Negative, False Positive and False Negative percentages of classification. The results show that specificity, as the ability of the algorithm to really detect healthy images, is 100%, and sensitivity, as the ability of the algorithm to correctly detect infected images is around 97.3%. The average accuracy of the system is around 98.6%. The results imply the ability of the proposed method compared to previous methods.

Research paper thumbnail of The development of an artificial neural network – genetic algorithm model (ANN-GA) for the adsorption and photocatalysis of methylene blue on a novel sulfur–nitrogen co-doped Fe2O3 nanostructure surface

RSC Advances

In this research an S-N doped Fe2O3 nanostructure is synthesized and its adsorption ability and p... more In this research an S-N doped Fe2O3 nanostructure is synthesized and its adsorption ability and photocatalytic activity were evaluated. The optimum experimental conditions were obtained and an ANN-GA model was proposed for predicting experimental values.

Research paper thumbnail of A Bayesian Regularized Artificial Neural Network for Simultaneous Determination of Loratadine, Naproxen and Diclofenac in wastewaters

Current Pharmaceutical Analysis

Background:: Simultaneous determination of medication components in pharmaceutical samples using ... more Background:: Simultaneous determination of medication components in pharmaceutical samples using ordinary methods have some difficulties and therefore these determinations usually were made by expensive methods and instruments. Chemometric methods are an effective way to analyze several components simultaneously. Objective:: In this paper a novel approach based on Bayesian regularized artificial neural network is developed for determination of Loratadine, Naproxen and Diclofenac in water using UV-Vis spectroscopy. Method:: A dataset is collected by performing several chemical experiments and recording the UV-Vis spectra and actual constituent values. The effect of different number of neuron in hidden layer was analyzed based on final mean square error, and the optimum number was selected. Principle Component Analysis (PCA) was also applied on the data. Other back-propagation methods, such as Levenberg-Marquardt, scaled conjugate gradient and resilient backpropagation are tested. Res...

Research paper thumbnail of Congestion Window Scaling Method to Optimize Delay in TCP/IP

Wireless Personal Communications

Research paper thumbnail of Improvement of MRI Brain Image Segmentation Using Fuzzy Unsupervised Learning

Iranian Journal of Radiology

Research paper thumbnail of An Leukocytes Counting System for Acute Lymphoblastic Leukemia Detection

Shiraz E-Medical Journal, 2017

Background: Today, blood disease is one of the most important causes of human deaths around the w... more Background: Today, blood disease is one of the most important causes of human deaths around the world, therefore, early diagnosis of these diseases are very important. Counting and classification of white blood cells (leukocytes) lead to identification of a variety of blood diseases such as leukemia. The aim of this research is producing a computer algorithm to count the leukocytes in order to help the hematologists detect acute lymphoblastic leukemia (ALL) in an accurate and time-efficient way. Objectives: The purpose of this research is to design and implement an intelligent software system based on image processing algorithms and fuzzy logic to analyze and accurately count blood leukocytes to identify acute lymphoblastic leukemia (ALL). Methods: The proposed image processing system consists of several sections. The first pre-processing is done to remove noise and improve image contrast. In the second step, the image is segmented using improved fuzzy clustering technique (IFCM) and active contour algorithm. In the third step, the image feature extraction and classification is done. The final step determines whether or not the image is ALL, using ANFIS neural network algorithm where its objective function is optimized by genetic algorithm. Results: Using samples of blood leukocyte images taken under the same lighting conditions let us introduce a computer aided diagnosis (CAD) system, which is empowered by fuzzy techniques for detection of all types of acute lymphoblastic cancers by 98% accuracy. Conclusions: A method for the detection and classification of blood leukocytes from the blood microscopic images using image processing techniques and fuzzy logic have been proposed. The results show that the proposed method is able to detect and classify leukocytes in an image with high accuracy.

Research paper thumbnail of A review of clinoptilolite, its photocatalytic, chemical activity, structure and properties: in time of artificial intelligence

Journal of Materials Science

Research paper thumbnail of Detection of Disc Destruction Between Lumbar Vertebrae Using MRI Images

Soft Computing, Apr 10, 2020

Most people experience low back pain at least once in their lifetime. Lumbar disc herniation is o... more Most people experience low back pain at least once in their lifetime. Lumbar disc herniation is one of the major causes of low back pain. Treatment methods for disc herniation are very diverse. Therefore, diagnosis of the exact size of herniation and its location can greatly help specialists to select the best treatment method. In this research, an automated method for diagnosing lumbar disc herniation using 130 MR images is proposed. In the proposed method, using three algorithms, namely region growing, OTSU and active contour, the intervertebral discs and their boundary were precisely separated from the background of the image. In the next step, after extracting the most significant features of the images, they are divided into healthy and unhealthy classes by SVM classifier with 89.9% accuracy. The classification accuracy was compared with classifiers KNN, Ensemble, and decision tree. Finally, it was determined the SVM classifier has the highest accuracy for the classification.

Research paper thumbnail of A Multi-objective Distribution Network Reconfiguration and Optimal Use of Distributed Generation Unites by Harmony Search Algorithm

Lecture Notes in Electrical Engineering, 2018

In this paper, the method of network reconfiguration and simultaneous use of distributed generati... more In this paper, the method of network reconfiguration and simultaneous use of distributed generation resources (DG) in optimal location and capacity is analyzed in order to minimize losses and reach the optimum level of voltage stability and voltage profile. A new method for this purpose is proposed by use of the Harmonic Search Algorithm (HSA). To investigate the effectiveness of the proposed method, the capabilities of the MATLAB software and the DPL language linked to the DIGSILENT application is used. The 33-node distribution network of IEEE standard was selected for investigation. The simulation results shows that by using the proposed method, network losses were minimized and the voltage level and voltage profiles were improved correctly.

Research paper thumbnail of Using Support Vector Machines as an Intelligent Algorithm for Detecting Seizures from EEG Signals

The Neuroscience Journal of Shefaye Khatam, 2021

Research paper thumbnail of FULL LENGTH ARTICLE OPEN ACCESS Recognition of Persian Handwritten Numbers using LBP - HOG Desc riptor

Recognition of handwritten numbers in any language is one of the most important issues attracted ... more Recognition of handwritten numbers in any language is one of the most important issues attracted many researchers and this is because of the increasing importance of the issue in today's world and in applications such as automatic detection of mail addresses or identification of numbers of bank checks. Given the great interclass diversities and much extra - class similarities between some Persian handwritten numbers, solving the Persian handwritten numbers recognition problem is difficult. In this paper, a new method for recognizing Persian handwritten numbers using a combination of HOG and LBP descriptors is provided. The proposed descriptor enjoys significant advantages of which the important one is the recording of information and features relat ing to the image (by descriptor LBP)and yet the feature extraction of the image edges (by descriptor HOG). Another advantage of the proposed descriptor is the very small length of the feature vector and the fast calculations. To evalu...

Research paper thumbnail of Recognition of Persian Handwritten Numbers using LBP-HOG Descriptor

Recognition of handwritten numbers in any language is one of the most important issues attracted ... more Recognition of handwritten numbers in any language is one of the most important issues attracted many researchers and this is because of the increasing importance of the issue in today's world and in applications such as automatic detection of mail addresses or identification of numbers of bank checks. Given the great interclass diversities and much extra-class similarities between some Persian handwritten numbers, solving the Persian handwritten numbers recognition problem is difficult. In this paper, a new method for recognizing Persian handwritten numbers using a combination of HOG and LBP descriptors is provided. The proposed descriptor enjoys significant advantages of which the important one is the recording of information and features relating to the image (by descriptor LBP)and yet the feature extraction of the image edges (by descriptor HOG). Another advantage of the proposed descriptor is the very small length of the feature vector and the fast calculations. To evaluate...

Research paper thumbnail of A deep learning algorithm to detect coronavirus (COVID-19) disease using CT images

PeerJ Computer Science, 2021

Background COVID-19 pandemic imposed a lockdown situation to the world these past months. Researc... more Background COVID-19 pandemic imposed a lockdown situation to the world these past months. Researchers and scientists around the globe faced serious efforts from its detection to its treatment. Methods Pathogenic laboratory testing is the gold standard but it is time-consuming. Lung CT-scans and X-rays are other common methods applied by researchers to detect COVID-19 positive cases. In this paper, we propose a deep learning neural network-based model as an alternative fast screening method that can be used for detecting the COVID-19 cases by analyzing CT-scans. Results Applying the proposed method on a publicly available dataset collected of positive and negative cases showed its ability on distinguishing them by analyzing each individual CT image. The effect of different parameters on the performance of the proposed model was studied and tabulated. By selecting random train and test images, the overall accuracy and ROC-AUC of the proposed model can easily exceed 95% and 90%, respec...

Research paper thumbnail of An Efficient ECC-based Authentication and Key Agreement Protocol

Public-key cryptography is commonly used to authenticate communicating entities in some networks.... more Public-key cryptography is commonly used to authenticate communicating entities in some networks. One of the key tools in this way is to use the elliptic curves cryptography (ECC) which is relatively lightweight due to its shorter key size compared to the conventional River-Shamir-Adleman (RSA) method. This paper is proposing an efficient protocol by analysing two variants of ECC-based wireless authentication protocol, namely, Aydos-Savas-Koc's wireless authentication protocol (ASK-WAP) and user authentication protocol (UAP) from various security aspects and communication concerns. We show that although UAP is able to address some of ASK-WAP vulnerabilities, it is confined to one-way communication where the authentication can only be initialized by users and not the server. In light of their limitations, we suggest several possible improvements to both ASK-WAP and UAP. The proposed solutions focus on applying encryption methods to the transmitted keys and enabling two-way commun...

Research paper thumbnail of A ground based circular synthetic aperture radar

2013 14th International Radar Symposium (IRS), 2013

Detecting on-the-ground objects is a subject of interest for some applications. Typical example i... more Detecting on-the-ground objects is a subject of interest for some applications. Typical example is foreign object detection on the airport runway. In response to this demand, a ground-based Circular Synthetic Aperture Radar (CSAR) system is proposed and explained in the paper. In the proposed CSAR, the antennas represent a circular movement trajectory. Wideband Linear Frequency (LFM) chirps were used for transmission. A simulation model for CSAR, based on the Doppler Effect between the radar and object is developed in this paper. In addition, a processing method for object detection using correlation between image data produced by simulation and experimental data is developed. The resultant of the simulated model at each point, which represents the object's behavior in an ideal and clutter-free environment, is used as a template for object detection. Simulation and experimental results demonstrate that the proposed method is well suited in detecting small objects at different po...

Research paper thumbnail of Designing a Hybrid Clustering Routing Algorithm based on Cellular Learning Automata for Optimizing Lifetime of Wireless Sensor Networks

Advances in Computer Science : an International Journal, 2015

One of the most important factors in wireless sensor networks is energy consumption, hence the li... more One of the most important factors in wireless sensor networks is energy consumption, hence the lifetime of these networks are strongly depending on remaining energy in the nodes. According to sensors placement farness and wireless communication between them, it is necessary to optimally consume the energy in these networks. In this study a hybrid approach is proposed by mixing two existing protocols, namely flat multi-hop routing and hierarchical multi-hop routing. Also by using Cellular Learning Automata (CLA) as clustering technique, the energy in the network will be managed and finally the lifetime of nodes will be increased. Mathematical simulation and analysis show a good performance of clustered hybrid model for energy saving that in compare with multi hop routing algorithm and hierarchical routing in non-clustered and clustered conditions, the lifetime increasing are %10.39, %27.36 and %5.57, %23.83 respectively.

Research paper thumbnail of A Hierarchical Classification Method for Breast Tumor Detection

Iranian Journal of Medical Physics, 2016

Introduction Breast cancer is the second cause of mortality among women. Early detection of it ca... more Introduction Breast cancer is the second cause of mortality among women. Early detection of it can enhance the chance of survival. Screening systems such as mammography cannot perfectly differentiate between patients and healthy individuals. Computer-aided diagnosis can help physicians make a more accurate diagnosis. Materials and Methods Regarding the importance of separating normal and abnormal cases in screening systems, a hierarchical classification system is defined in this paper. The proposed system is including two Adaptive Boosting (AdaBoost) classifiers, the first classifier separates the candidate images into two groups of normal and abnormal. The second classifier is applied on the abnormal group of the previous stage and divides them into benign and malignant categories. The proposed algorithm is evaluated by applying it on publicly available Mammographic Image Analysis Society (MIAS) dataset. 288 images of the database are used, including 208 normal and 80 abnormal im...

Research paper thumbnail of An Intelligent Algorithm For Detecting Minerals In Microscopic Sections Using Machine Learning

Detecting the size and shape of minerals is very important for collecting information on minerals... more Detecting the size and shape of minerals is very important for collecting information on minerals and the texture of rocks for classification and naming. Therefore, it is necessary to study the size and shape of minerals. By combining image processing and intelligent pattern recognition techniques, successful results can be obtained through detecting minerals in sections as well as their size and shape, especially in thin-section images that have reduced the third dimensional effect. In this research, a method is proposed for segmentation of thin sections and finding train minerals in them. In this regard, user input is used to select some portions of the studied mineral. These pieces of data are used as seed points to learn the neural network. Support Vector Machine (SVM) is used as a strong classifier algorithm to find the mineral samples in the whole image. The combination of colours and mineral features is used to train the SVM in order to find the minerals with high precision. ...

Research paper thumbnail of A Novel Method for Persian Handwritten Digit Recognition Using Support Vector Machine

Handwritten digit recognition has got a special role in different applications in the field of di... more Handwritten digit recognition has got a special role in different applications in the field of digital recognition including; handwritten address detection, check, and document. Persian handwritten digits classification has been facing difficulties due to different handwritten styles, inter-class similarities, and intra-class differences. In this paper, a novel method for detecting Persian handwritten digits is presented. In the proposed method, a combination of Histogram of Oriented Gradients (HOG), 4-side profiles of the digit image, and some horizontal and vertical samples was used and the dimension of the feature vector was reduced using Principal Component Analysis (PCA). The proposed method applied to the HODA database, and Support Vector Machine (SVM) was used in the classification step. Results revealed that the detection accuracy of such method has 99% accuracy with an adequate rate due to existing unacceptable samples in the database, therefore, the proposed method could ...

Research paper thumbnail of Introducing an intelligent algorithm for extraction of sand dunes from Landsat satellite imagery in terrestrial and coastal environments

Journal of Coastal Conservation

Research paper thumbnail of An Intelligent Technique for Grape Fanleaf Virus Detection

International Journal of Interactive Multimedia and Artificial Intelligence

Grapevine Fanleaf Virus (GFLV) is one of the most important viral diseases of grapes, which can d... more Grapevine Fanleaf Virus (GFLV) is one of the most important viral diseases of grapes, which can damage up to 85% of the crop, if not treated at the right time. The aim of this study is to identify infected leaves with GFLV using artificial intelligent methods using an accessible database. To do this, some pictures are taken from infected and healthy leaves of grapes and labeled by technical specialists using conventional laboratory methods. In order to provide an intelligent method for distinguishing infected leaves from healthy ones, the area of unhealthy parts of each leaf is highlighted using Fuzzy C-mean Algorithm (FCM), and then the percentages of the first two segments area are fed to a Support Vector Machines (SVM). To increase the diagnostic reliability of the system, K-fold cross validation method with k = 3 and k =5 is applied. After applying the proposed method over all images using K-fold validation technique, average confusion matrix is extracted to show the True Positive, True Negative, False Positive and False Negative percentages of classification. The results show that specificity, as the ability of the algorithm to really detect healthy images, is 100%, and sensitivity, as the ability of the algorithm to correctly detect infected images is around 97.3%. The average accuracy of the system is around 98.6%. The results imply the ability of the proposed method compared to previous methods.

Research paper thumbnail of The development of an artificial neural network – genetic algorithm model (ANN-GA) for the adsorption and photocatalysis of methylene blue on a novel sulfur–nitrogen co-doped Fe2O3 nanostructure surface

RSC Advances

In this research an S-N doped Fe2O3 nanostructure is synthesized and its adsorption ability and p... more In this research an S-N doped Fe2O3 nanostructure is synthesized and its adsorption ability and photocatalytic activity were evaluated. The optimum experimental conditions were obtained and an ANN-GA model was proposed for predicting experimental values.

Research paper thumbnail of A Bayesian Regularized Artificial Neural Network for Simultaneous Determination of Loratadine, Naproxen and Diclofenac in wastewaters

Current Pharmaceutical Analysis

Background:: Simultaneous determination of medication components in pharmaceutical samples using ... more Background:: Simultaneous determination of medication components in pharmaceutical samples using ordinary methods have some difficulties and therefore these determinations usually were made by expensive methods and instruments. Chemometric methods are an effective way to analyze several components simultaneously. Objective:: In this paper a novel approach based on Bayesian regularized artificial neural network is developed for determination of Loratadine, Naproxen and Diclofenac in water using UV-Vis spectroscopy. Method:: A dataset is collected by performing several chemical experiments and recording the UV-Vis spectra and actual constituent values. The effect of different number of neuron in hidden layer was analyzed based on final mean square error, and the optimum number was selected. Principle Component Analysis (PCA) was also applied on the data. Other back-propagation methods, such as Levenberg-Marquardt, scaled conjugate gradient and resilient backpropagation are tested. Res...

Research paper thumbnail of Congestion Window Scaling Method to Optimize Delay in TCP/IP

Wireless Personal Communications

Research paper thumbnail of Improvement of MRI Brain Image Segmentation Using Fuzzy Unsupervised Learning

Iranian Journal of Radiology