This document discusses feature extraction techniques for Alzheimer's disease diagnosis using PET scans. It examines voxel intensity, scale space representation through Gaussian pyramids, and a local variance operator to capture local contrast. Voxel intensity provides direct measurement of FDG uptake but neighboring voxels are highly correlated. The scale space representation reduces redundancy through smoothing and subsampling. The local variance operator measures variance within a 3D neighborhood to estimate local contrast, allowing features at different scales by varying neighborhood radius. These feature extraction methods aim to effectively represent relevant information from PET scans to improve computer-aided Alzheimer's diagnosis.
Alzheimer’s disease(AD) is a neurological disease. It affects memory. The livelihood of the people that are
diagnosed with AD. In this paper, we have discussed various imaging modalities, feature selection and
extraction, segmentation and classification techniques.
Multistage Classification of Alzheimer’s DiseaseIJLT EMAS
Alzheimer’s disease is a type of dementia that destroys
memory and other mental functions. During the progression of
the disease certain proteins called plaques and tangles get
deposited in hippocampus which is located in the temporal lobe
of brain. The disease is not a normal part of aging and gets
worsen over time. Medical imaging techniques like Magnetic
Resonance Imaging (MRI), Computed Tomography (CT) and
Positron Emission Tomography (PET) play significant role in the
disease diagnosis. In this paper, we propose a method for
classifying MRI into Normal Control (NC), Mild Cognitive
Impairment (MCI) and Alzheimer’s Disease(AD). An overall
outline of the methodology includes textural feature extraction,
feature reduction process and classification of the images into
various stages. Classification has been performed with three
classifiers namely Support Vector Machine (SVM), Artificial
Neural Network (ANN) and k-Nearest Neighbours (k-NN)
Segmentation of Brain MR Images for Tumor Extraction by Combining Kmeans Clus...CSCJournals
Segmentation of images holds an important position in the area of image processing. It becomes more important while typically dealing with medical images where pre-surgery and post surgery decisions are required for the purpose of initiating and speeding up the recovery process [5] Computer aided detection of abnormal growth of tissues is primarily motivated by the necessity of achieving maximum possible accuracy. Manual segmentation of these abnormal tissues cannot be compared with modern day’s high speed computing machines which enable us to visually observe the volume and location of unwanted tissues. A well known segmentation problem within MRI is the task of labeling voxels according to their tissue type which include White Matter (WM), Grey Matter (GM) , Cerebrospinal Fluid (CSF) and sometimes pathological tissues like tumor etc. This paper describes an efficient method for automatic brain tumor segmentation for the extraction of tumor tissues from MR images. It combines Perona and Malik anisotropic diffusion model for image enhancement and Kmeans clustering technique for grouping tissues belonging to a specific group. The proposed method uses T1, T2 and PD weighted gray level intensity images. The proposed technique produced appreciative results
Alzheimer’s disease(AD) is a neurological disease. It affects memory. The livelihood of the people that are
diagnosed with AD. In this paper, we have discussed various imaging modalities, feature selection and
extraction, segmentation and classification techniques.
Multistage Classification of Alzheimer’s DiseaseIJLT EMAS
Alzheimer’s disease is a type of dementia that destroys
memory and other mental functions. During the progression of
the disease certain proteins called plaques and tangles get
deposited in hippocampus which is located in the temporal lobe
of brain. The disease is not a normal part of aging and gets
worsen over time. Medical imaging techniques like Magnetic
Resonance Imaging (MRI), Computed Tomography (CT) and
Positron Emission Tomography (PET) play significant role in the
disease diagnosis. In this paper, we propose a method for
classifying MRI into Normal Control (NC), Mild Cognitive
Impairment (MCI) and Alzheimer’s Disease(AD). An overall
outline of the methodology includes textural feature extraction,
feature reduction process and classification of the images into
various stages. Classification has been performed with three
classifiers namely Support Vector Machine (SVM), Artificial
Neural Network (ANN) and k-Nearest Neighbours (k-NN)
Segmentation of Brain MR Images for Tumor Extraction by Combining Kmeans Clus...CSCJournals
Segmentation of images holds an important position in the area of image processing. It becomes more important while typically dealing with medical images where pre-surgery and post surgery decisions are required for the purpose of initiating and speeding up the recovery process [5] Computer aided detection of abnormal growth of tissues is primarily motivated by the necessity of achieving maximum possible accuracy. Manual segmentation of these abnormal tissues cannot be compared with modern day’s high speed computing machines which enable us to visually observe the volume and location of unwanted tissues. A well known segmentation problem within MRI is the task of labeling voxels according to their tissue type which include White Matter (WM), Grey Matter (GM) , Cerebrospinal Fluid (CSF) and sometimes pathological tissues like tumor etc. This paper describes an efficient method for automatic brain tumor segmentation for the extraction of tumor tissues from MR images. It combines Perona and Malik anisotropic diffusion model for image enhancement and Kmeans clustering technique for grouping tissues belonging to a specific group. The proposed method uses T1, T2 and PD weighted gray level intensity images. The proposed technique produced appreciative results
MRIIMAGE SEGMENTATION USING LEVEL SET METHOD AND IMPLEMENT AN MEDICAL DIAGNOS...cseij
Image segmentation plays a vital role in image processing over the last few years. The goal of image segmentation is to cluster the pixels into salient image regions i.e., regions corresponding to individual surfaces, objects, or natural parts of objects. In this paper, we propose a medical diagnosis system by using level set method for segmenting the MRI image which investigates a new variational level set algorithm without re- initialization to segment the MRI image and to implement a competent medical diagnosis system by using MATLAB. Here we have used the speed function and the signed distance function of the image in segmentation algorithm. This system consists of thresholding technique, curve evolution technique and an eroding technique. Our proposed system was tested on some MRI Brain images, giving promising results by detecting the normal or abnormal condition specially the existence of tumers. This system will be applied to both simulated and real images with promising results
Glaucoma Disease Diagnosis Using Feed Forward Neural Network ijcisjournal
Glaucoma is an eye disease which damages the optic nerve and or loss of the field of vision which leads to
complete blindness caused by the pressure buildup by the fluid of the eye i.e. the intraocular pressure
(IOP). This optic disorder with a gradual loss of the field of vision leads to progressive and irreversible
blindness, so it should be diagnosed and treated properly at an early stage. In this paper,
thedaubechies(db3) or symlets (sym3)and reverse biorthogonal (rbio3.7) wavelet filters are employed for
obtaining average and energy texture feature which are used to classify glaucoma disease with high
accuracy. The Feed-Forward neural network classifies the glaucoma disease with an accuracy of 96.67%.
In this work, the computational complexity is minimized by reducing the number of filters while retaining
the same accuracy.
Automated Diagnosis of Glaucoma using Haralick Texture FeaturesIOSR Journals
Abstract : Glaucoma is the second leading cause of blindness worldwide. It is a disease in which fluid
pressure in the eye increases continuously, damaging the optic nerve and causing vision loss. Computational
decision support systems for the early detection of glaucoma can help prevent this complication. The retinal
optic nerve fibre layer can be assessed using optical coherence tomography, scanning laser polarimetry, and
Heidelberg retina tomography scanning methods. In this paper, we present a novel method for glaucoma
detection using an Haralick Texture Features from digital fundus images. K Nearest Neighbors (KNN)
classifiers are used to perform supervised classification. Our results demonstrate that the Haralick Texture
Features has Database and classification parts, in Database the image has been loaded and Gray Level Cooccurrence
Matrix (GLCM) and thirteen haralick features are combined to extract the image features, performs
better than the other classifiers and correctly identifies the glaucoma images with an accuracy of more than
98%. The impact of training and testing is also studied to improve results. Our proposed novel features are
clinically significant and can be used to detect glaucoma accurately.
Keywords: Glaucoma, Haralick Texture features, KNN Classifiers, Feature Extraction
SVM Classification of MRI Brain Images for ComputerAssisted DiagnosisIJECEIAES
Magnetic Resonance Imaging is a powerful technique that helps in the diagnosis of various medical conditions. MRI Image pre-processing followed by detection of brain abnormalities, such as brain tumors, are considered in this work. These images are often corrupted by noise from various sources. The Discrete Wavelet Transforms (DWT) with details thresholding is used for efficient noise removal followed by edge detection and threshold segmentation of the denoised images. Segmented image features are then extracted using morphological operations. These features are finally used to train an improved Support Vector Machine classifier that uses a Gausssian radial basis function kernel. The performance of the classifier is evaluated and the results of the classification show that the proposed scheme accurately distinguishes normal brain images from the abnormal ones and benign lesions from malignant tumours. The accuracy of the classification is shown to be 100% which is superior to the results reported in the literature.
A Novel Approach for Diabetic Retinopthy ClassificationIJERA Editor
Sustainable Diabetic Mellitus may lead to several complications towards patients. One of the complications is
diabetic retinopathy. Diabetic retinopathy is the type of complication towards the retinal and interferes with
patient’s sight. Medical examination toward patients with diabetic retinopathy is observed directly through
retinal images using fundus camera. Diabetic retinopathy is classified into four classes based on severity, which
are: normal, non-proliferative diabetic retinopathy (NPDR), proliferative diabetic retinopathy (PDR), and
macular edema (ME). The aim of this research is to develop a method which can be used to classify the level of
severity of diabetic retinopathy based on patient’s retinal images. Seven texture features were extracted from
retinal images using gray level co-occurence matrix three dimensional method (3D-GLCM). These features are
maximum probability, correlation, contrast, energy, homogeneity, and entropy; subsequently trained using
Levenberg-Marquardt Backpropagation Neural Network (LMBP). This study used 600 data of patient’s retinal
images, consist of 450 data retinal images for training and 150 data retinal images for testing. Based on the result
of this test, the method can classify the severity of diabetic retinopathy with sensitivity of 97.37%, specificity of
75% and accuracy of 91.67%
Extraction of Circle of Willis from 2D Magnetic Resonance AngiogramsIDES Editor
Magnetic resonance angiogram is a way to study
cerebrovascular structures. It helps to obtain information
regarding blood flow in a non-invasive fashion. Magnetic
resonance angiograms are examined basically for detection
of vascular pathologies, neurosurgery planning, and vascular
landmark detection. In certain cases it becomes complicated
for the doctors to assess the cerebral vessels or Circle of Willis
from the two-dimensional (2D) brain magnetic resonance
angiograms. In this paper an attempt has been made to extract
the Circle of Willis from 2D magnetic resonance angiograms,
so as to overcome such difficulties. The proposed method preprocesses
the magnetic resonance angiograms and
subsequently extracts the Circle of Willis. The extraction has
been done by color-based segmentation using K-means
clustering algorithm. As the developed method successfully
extracts the vasculature from the brain magnetic resonance
angiograms, therefore it will help the doctors for diagnosis
and serve as a step in the prevention of stroke. The algorithms
are developed on MATLAB 7.6.0 (R2008a) programming
platform.
International Journal of Computational Engineering Research(IJCER) is an intentional online Journal in English monthly publishing journal. This Journal publish original research work that contributes significantly to further the scientific knowledge in engineering and Technology
SEGMENTATION OF MULTIPLE SCLEROSIS LESION IN BRAIN MR IMAGES USING FUZZY C-MEANSijaia
Magnetic resonance images (MRI) play an important role in supporting and substituting clinical
information in the diagnosis of multiple sclerosis (MS) disease by presenting lesion in brain MR images. In
this paper, an algorithm for MS lesion segmentation from Brain MR Images has been presented. We revisit
the modification of properties of fuzzy -c means algorithms and the canny edge detection. By changing and
reformed fuzzy c-means clustering algorithms, and applying canny contraction principle, a relationship
between MS lesions and edge detection is established. For the special case of FCM, we derive a sufficient
condition and clustering parameters, allowing identification of them as (local) minima of the objective
function.
MULTIPLE SCLEROSIS DIAGNOSIS WITH FUZZY C-MEANScscpconf
Magnetic resonance imaging (MRI) can support and substitute clinical information in the
diagnosis of multiple sclerosis (MS) by presenting lesion. In this paper, we present an algorithm
for MS lesion segmentation. We revisit the modification of properties of fuzzy c means
algorithms and the canny edge detection. Using reformulated fuzzy c means algorithms, apply
canny contraction principle, and establish a relationship between MS lesions and edge
detection. For the special case of FCM, we derive a sufficient condition for fixed lesions,
allowing identification of them as (local) minima of the objective function.
A UGMENT R EALITY IN V OLUMETRIC M EDICAL I MAGING U SING S TEREOSCOPIC...ijcga
This paper is written about augment reality in medi
cine. Medical imaging equipment (CT, PET, MRI) are
produced 3D volumetric data, so using the stereosco
pic 3D display, observer feels depth perception. Th
e
major factors about depth-Convergence, Accommodatio
n, Relative size are tested. Convergence and
Accommodation have affected depth perception but re
lative size is negligibl
Augmented Reality in Volumetric Medical Imaging Using Stereoscopic 3D Display ijcga
This paper is written about augmented reality in medicine. Medical imaging equipment (CT, PET, MRI) are produced 3D volumetric data, so using the stereoscopic 3D display, observer feels depth perception. The major factors about depth-Convergence, Accommodation, Relative size are tested. Convergence and Accommodation have affected depth perception but relative size is negligible.
Image Binarization for the uses of Preprocessing to Detect Brain Abnormality ...Journal For Research
Computerized MR of brain image binarization for the uses of preprocessing of features extraction and brain abnormality identification of brain has been described. Binarization is used as intermediate steps of many MR of brain normal and abnormal tissues detection. One of the main problems of MRI binarization is that many pixels of brain part cannot be correctly binarized due to the extensive black background or the large variation in contrast between background and foreground of MRI. Proposed binarization determines a threshold value using mean, variance, standard deviation and entropy followed by a non-gamut enhancement that can overcome the binarization problem. The proposed binarization technique is extensively tested with a variety of MRI and generates good binarization with improved accuracy and reduced error.
Brain Tumor Detection using MRI ImagesYogeshIJTSRD
Brain tumor segmentation is a very important task in medical image processing. Early diagnosis of brain tumors plays a crucial role in improving treatment possibilities and increases the survival rate of the patients. For the study of tumor detection and segmentation, MRI Images are very useful in recent years. One of the foremost crucial tasks in any brain tumor detection system is that the detachment of abnormal tissues from normal brain tissues. Because of MRI Images, we will detect the brain tumor. Detection of unusual growth of tissues and blocks of blood within the system is seen in an MRI Imaging. Brain tumor detection using MRI images may be a challenging task due to the brains complex structure.In this paper, we propose an image segmentation method to detect tumors from MRI images using an interface of GUI in MATLAB. The method of distinguishing brain tumors through MRI images is often sorted into four sections of image processing as pre processing, feature extraction, image segmentation, and image classification. During this paper, weve used various algorithms for the partial fulfillment of the necessities to hit the simplest results that may help us to detect brain tumors within the early stage. Deepa Dangwal | Aditya Nautiyal | Dakshita Adhikari | Kapil Joshi "Brain Tumor Detection using MRI Images" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Special Issue | International Conference on Advances in Engineering, Science and Technology - 2021 , May 2021, URL: https://www.ijtsrd.com/papers/ijtsrd42456.pdf Paper URL : https://www.ijtsrd.com/engineering/computer-engineering/42456/brain-tumor-detection-using-mri-images/deepa-dangwal
International Journal of Engineering Research and Applications (IJERA) is an open access online peer reviewed international journal that publishes research and review articles in the fields of Computer Science, Neural Networks, Electrical Engineering, Software Engineering, Information Technology, Mechanical Engineering, Chemical Engineering, Plastic Engineering, Food Technology, Textile Engineering, Nano Technology & science, Power Electronics, Electronics & Communication Engineering, Computational mathematics, Image processing, Civil Engineering, Structural Engineering, Environmental Engineering, VLSI Testing & Low Power VLSI Design etc.
A Review of different method of Medical Image Segmentationijsrd.com
Image Segmentation is a most important task of image analysis. Number of method used for image segmentation. Image segmentation mainly used in different field like medical image analysis, character re-congestion etc. A segmentation method finds the sets that are different structure from each other and completion of segmentation process that cover entire image.
Alzheimer’s detection through neuro imaging and subsequent fusion for clinica...IJECEIAES
In recent years, vast improvement has been observed in the field of medical research. Alzheimer's is the most common cause for dementia. Alzheimer's disease (AD) is a chronic disease with no cure, and it continues to pose a threat to millions of lives worldwide. The main purpose of this study is to detect the presence of AD from magnetic resonance imaging (MRI) scans through neuro imaging and to perform fusion process of both MRI and positron emission tomography (PET) scans of the same patient to obtain a fused image with more detailed information. Detection of AD is done by calculating the gray matter and white matter volumes of the brain and subsequently, a ratio of calculated volume is taken which helps the doctors in deciding whether the patient is affected with or without the disease. Image fusion is carried out after preliminary detection of AD for MRI scan along with PET scan. The main objective is to combine these two images into a single image which contains all the possible information together. The proposed approach yields better results with a peak signal to noise ratio of 60.6 dB, mean square error of 0.0176, entropy of 4.6 and structural similarity index of 0.8.
A modified residual network for detection and classification of Alzheimer’s ...IJECEIAES
Alzheimer's disease (AD) is a brain disease that significantly declines a person's ability to remember and behave normally. By applying several approaches to distinguish between various stages of AD, neuroimaging data has been used to extract different patterns associated with various phases of AD. However, because the brain patterns of older adults and those in different phases are similar, researchers have had difficulty classifying them. In this paper, the 50-layer residual neural network (ResNet) is modified by adding extra convolution layers to make the extracted features more diverse. Besides, the activation function (ReLU) was replaced with (Leaky ReLU) because ReLU takes the negative parts of its input, drops them to zero, and retains the positive parts. These negative inputs may contain useful feature information that could aid in the development of high-level discriminative features. Thus, Leaky ReLU was used instead of ReLU to prevent any potential loss of input information. In order to train the network from scratch without encountering the issue of overfitting, we added a dropout layer before the fully connected layer. The proposed method successfully classified the four stages of AD with an accuracy of 97.49 % and 98 % for precision, recall, and f1-score.
MRIIMAGE SEGMENTATION USING LEVEL SET METHOD AND IMPLEMENT AN MEDICAL DIAGNOS...cseij
Image segmentation plays a vital role in image processing over the last few years. The goal of image segmentation is to cluster the pixels into salient image regions i.e., regions corresponding to individual surfaces, objects, or natural parts of objects. In this paper, we propose a medical diagnosis system by using level set method for segmenting the MRI image which investigates a new variational level set algorithm without re- initialization to segment the MRI image and to implement a competent medical diagnosis system by using MATLAB. Here we have used the speed function and the signed distance function of the image in segmentation algorithm. This system consists of thresholding technique, curve evolution technique and an eroding technique. Our proposed system was tested on some MRI Brain images, giving promising results by detecting the normal or abnormal condition specially the existence of tumers. This system will be applied to both simulated and real images with promising results
Glaucoma Disease Diagnosis Using Feed Forward Neural Network ijcisjournal
Glaucoma is an eye disease which damages the optic nerve and or loss of the field of vision which leads to
complete blindness caused by the pressure buildup by the fluid of the eye i.e. the intraocular pressure
(IOP). This optic disorder with a gradual loss of the field of vision leads to progressive and irreversible
blindness, so it should be diagnosed and treated properly at an early stage. In this paper,
thedaubechies(db3) or symlets (sym3)and reverse biorthogonal (rbio3.7) wavelet filters are employed for
obtaining average and energy texture feature which are used to classify glaucoma disease with high
accuracy. The Feed-Forward neural network classifies the glaucoma disease with an accuracy of 96.67%.
In this work, the computational complexity is minimized by reducing the number of filters while retaining
the same accuracy.
Automated Diagnosis of Glaucoma using Haralick Texture FeaturesIOSR Journals
Abstract : Glaucoma is the second leading cause of blindness worldwide. It is a disease in which fluid
pressure in the eye increases continuously, damaging the optic nerve and causing vision loss. Computational
decision support systems for the early detection of glaucoma can help prevent this complication. The retinal
optic nerve fibre layer can be assessed using optical coherence tomography, scanning laser polarimetry, and
Heidelberg retina tomography scanning methods. In this paper, we present a novel method for glaucoma
detection using an Haralick Texture Features from digital fundus images. K Nearest Neighbors (KNN)
classifiers are used to perform supervised classification. Our results demonstrate that the Haralick Texture
Features has Database and classification parts, in Database the image has been loaded and Gray Level Cooccurrence
Matrix (GLCM) and thirteen haralick features are combined to extract the image features, performs
better than the other classifiers and correctly identifies the glaucoma images with an accuracy of more than
98%. The impact of training and testing is also studied to improve results. Our proposed novel features are
clinically significant and can be used to detect glaucoma accurately.
Keywords: Glaucoma, Haralick Texture features, KNN Classifiers, Feature Extraction
SVM Classification of MRI Brain Images for ComputerAssisted DiagnosisIJECEIAES
Magnetic Resonance Imaging is a powerful technique that helps in the diagnosis of various medical conditions. MRI Image pre-processing followed by detection of brain abnormalities, such as brain tumors, are considered in this work. These images are often corrupted by noise from various sources. The Discrete Wavelet Transforms (DWT) with details thresholding is used for efficient noise removal followed by edge detection and threshold segmentation of the denoised images. Segmented image features are then extracted using morphological operations. These features are finally used to train an improved Support Vector Machine classifier that uses a Gausssian radial basis function kernel. The performance of the classifier is evaluated and the results of the classification show that the proposed scheme accurately distinguishes normal brain images from the abnormal ones and benign lesions from malignant tumours. The accuracy of the classification is shown to be 100% which is superior to the results reported in the literature.
A Novel Approach for Diabetic Retinopthy ClassificationIJERA Editor
Sustainable Diabetic Mellitus may lead to several complications towards patients. One of the complications is
diabetic retinopathy. Diabetic retinopathy is the type of complication towards the retinal and interferes with
patient’s sight. Medical examination toward patients with diabetic retinopathy is observed directly through
retinal images using fundus camera. Diabetic retinopathy is classified into four classes based on severity, which
are: normal, non-proliferative diabetic retinopathy (NPDR), proliferative diabetic retinopathy (PDR), and
macular edema (ME). The aim of this research is to develop a method which can be used to classify the level of
severity of diabetic retinopathy based on patient’s retinal images. Seven texture features were extracted from
retinal images using gray level co-occurence matrix three dimensional method (3D-GLCM). These features are
maximum probability, correlation, contrast, energy, homogeneity, and entropy; subsequently trained using
Levenberg-Marquardt Backpropagation Neural Network (LMBP). This study used 600 data of patient’s retinal
images, consist of 450 data retinal images for training and 150 data retinal images for testing. Based on the result
of this test, the method can classify the severity of diabetic retinopathy with sensitivity of 97.37%, specificity of
75% and accuracy of 91.67%
Extraction of Circle of Willis from 2D Magnetic Resonance AngiogramsIDES Editor
Magnetic resonance angiogram is a way to study
cerebrovascular structures. It helps to obtain information
regarding blood flow in a non-invasive fashion. Magnetic
resonance angiograms are examined basically for detection
of vascular pathologies, neurosurgery planning, and vascular
landmark detection. In certain cases it becomes complicated
for the doctors to assess the cerebral vessels or Circle of Willis
from the two-dimensional (2D) brain magnetic resonance
angiograms. In this paper an attempt has been made to extract
the Circle of Willis from 2D magnetic resonance angiograms,
so as to overcome such difficulties. The proposed method preprocesses
the magnetic resonance angiograms and
subsequently extracts the Circle of Willis. The extraction has
been done by color-based segmentation using K-means
clustering algorithm. As the developed method successfully
extracts the vasculature from the brain magnetic resonance
angiograms, therefore it will help the doctors for diagnosis
and serve as a step in the prevention of stroke. The algorithms
are developed on MATLAB 7.6.0 (R2008a) programming
platform.
International Journal of Computational Engineering Research(IJCER) is an intentional online Journal in English monthly publishing journal. This Journal publish original research work that contributes significantly to further the scientific knowledge in engineering and Technology
SEGMENTATION OF MULTIPLE SCLEROSIS LESION IN BRAIN MR IMAGES USING FUZZY C-MEANSijaia
Magnetic resonance images (MRI) play an important role in supporting and substituting clinical
information in the diagnosis of multiple sclerosis (MS) disease by presenting lesion in brain MR images. In
this paper, an algorithm for MS lesion segmentation from Brain MR Images has been presented. We revisit
the modification of properties of fuzzy -c means algorithms and the canny edge detection. By changing and
reformed fuzzy c-means clustering algorithms, and applying canny contraction principle, a relationship
between MS lesions and edge detection is established. For the special case of FCM, we derive a sufficient
condition and clustering parameters, allowing identification of them as (local) minima of the objective
function.
MULTIPLE SCLEROSIS DIAGNOSIS WITH FUZZY C-MEANScscpconf
Magnetic resonance imaging (MRI) can support and substitute clinical information in the
diagnosis of multiple sclerosis (MS) by presenting lesion. In this paper, we present an algorithm
for MS lesion segmentation. We revisit the modification of properties of fuzzy c means
algorithms and the canny edge detection. Using reformulated fuzzy c means algorithms, apply
canny contraction principle, and establish a relationship between MS lesions and edge
detection. For the special case of FCM, we derive a sufficient condition for fixed lesions,
allowing identification of them as (local) minima of the objective function.
A UGMENT R EALITY IN V OLUMETRIC M EDICAL I MAGING U SING S TEREOSCOPIC...ijcga
This paper is written about augment reality in medi
cine. Medical imaging equipment (CT, PET, MRI) are
produced 3D volumetric data, so using the stereosco
pic 3D display, observer feels depth perception. Th
e
major factors about depth-Convergence, Accommodatio
n, Relative size are tested. Convergence and
Accommodation have affected depth perception but re
lative size is negligibl
Augmented Reality in Volumetric Medical Imaging Using Stereoscopic 3D Display ijcga
This paper is written about augmented reality in medicine. Medical imaging equipment (CT, PET, MRI) are produced 3D volumetric data, so using the stereoscopic 3D display, observer feels depth perception. The major factors about depth-Convergence, Accommodation, Relative size are tested. Convergence and Accommodation have affected depth perception but relative size is negligible.
Image Binarization for the uses of Preprocessing to Detect Brain Abnormality ...Journal For Research
Computerized MR of brain image binarization for the uses of preprocessing of features extraction and brain abnormality identification of brain has been described. Binarization is used as intermediate steps of many MR of brain normal and abnormal tissues detection. One of the main problems of MRI binarization is that many pixels of brain part cannot be correctly binarized due to the extensive black background or the large variation in contrast between background and foreground of MRI. Proposed binarization determines a threshold value using mean, variance, standard deviation and entropy followed by a non-gamut enhancement that can overcome the binarization problem. The proposed binarization technique is extensively tested with a variety of MRI and generates good binarization with improved accuracy and reduced error.
Brain Tumor Detection using MRI ImagesYogeshIJTSRD
Brain tumor segmentation is a very important task in medical image processing. Early diagnosis of brain tumors plays a crucial role in improving treatment possibilities and increases the survival rate of the patients. For the study of tumor detection and segmentation, MRI Images are very useful in recent years. One of the foremost crucial tasks in any brain tumor detection system is that the detachment of abnormal tissues from normal brain tissues. Because of MRI Images, we will detect the brain tumor. Detection of unusual growth of tissues and blocks of blood within the system is seen in an MRI Imaging. Brain tumor detection using MRI images may be a challenging task due to the brains complex structure.In this paper, we propose an image segmentation method to detect tumors from MRI images using an interface of GUI in MATLAB. The method of distinguishing brain tumors through MRI images is often sorted into four sections of image processing as pre processing, feature extraction, image segmentation, and image classification. During this paper, weve used various algorithms for the partial fulfillment of the necessities to hit the simplest results that may help us to detect brain tumors within the early stage. Deepa Dangwal | Aditya Nautiyal | Dakshita Adhikari | Kapil Joshi "Brain Tumor Detection using MRI Images" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Special Issue | International Conference on Advances in Engineering, Science and Technology - 2021 , May 2021, URL: https://www.ijtsrd.com/papers/ijtsrd42456.pdf Paper URL : https://www.ijtsrd.com/engineering/computer-engineering/42456/brain-tumor-detection-using-mri-images/deepa-dangwal
International Journal of Engineering Research and Applications (IJERA) is an open access online peer reviewed international journal that publishes research and review articles in the fields of Computer Science, Neural Networks, Electrical Engineering, Software Engineering, Information Technology, Mechanical Engineering, Chemical Engineering, Plastic Engineering, Food Technology, Textile Engineering, Nano Technology & science, Power Electronics, Electronics & Communication Engineering, Computational mathematics, Image processing, Civil Engineering, Structural Engineering, Environmental Engineering, VLSI Testing & Low Power VLSI Design etc.
A Review of different method of Medical Image Segmentationijsrd.com
Image Segmentation is a most important task of image analysis. Number of method used for image segmentation. Image segmentation mainly used in different field like medical image analysis, character re-congestion etc. A segmentation method finds the sets that are different structure from each other and completion of segmentation process that cover entire image.
Alzheimer’s detection through neuro imaging and subsequent fusion for clinica...IJECEIAES
In recent years, vast improvement has been observed in the field of medical research. Alzheimer's is the most common cause for dementia. Alzheimer's disease (AD) is a chronic disease with no cure, and it continues to pose a threat to millions of lives worldwide. The main purpose of this study is to detect the presence of AD from magnetic resonance imaging (MRI) scans through neuro imaging and to perform fusion process of both MRI and positron emission tomography (PET) scans of the same patient to obtain a fused image with more detailed information. Detection of AD is done by calculating the gray matter and white matter volumes of the brain and subsequently, a ratio of calculated volume is taken which helps the doctors in deciding whether the patient is affected with or without the disease. Image fusion is carried out after preliminary detection of AD for MRI scan along with PET scan. The main objective is to combine these two images into a single image which contains all the possible information together. The proposed approach yields better results with a peak signal to noise ratio of 60.6 dB, mean square error of 0.0176, entropy of 4.6 and structural similarity index of 0.8.
A modified residual network for detection and classification of Alzheimer’s ...IJECEIAES
Alzheimer's disease (AD) is a brain disease that significantly declines a person's ability to remember and behave normally. By applying several approaches to distinguish between various stages of AD, neuroimaging data has been used to extract different patterns associated with various phases of AD. However, because the brain patterns of older adults and those in different phases are similar, researchers have had difficulty classifying them. In this paper, the 50-layer residual neural network (ResNet) is modified by adding extra convolution layers to make the extracted features more diverse. Besides, the activation function (ReLU) was replaced with (Leaky ReLU) because ReLU takes the negative parts of its input, drops them to zero, and retains the positive parts. These negative inputs may contain useful feature information that could aid in the development of high-level discriminative features. Thus, Leaky ReLU was used instead of ReLU to prevent any potential loss of input information. In order to train the network from scratch without encountering the issue of overfitting, we added a dropout layer before the fully connected layer. The proposed method successfully classified the four stages of AD with an accuracy of 97.49 % and 98 % for precision, recall, and f1-score.
The biomedical profession has gained importance due to the rapid and accurate diagnosis of clinical patients using computer-aided diagnosis (CAD) tools.
The diagnosis and treatment of Alzheimer’s disease (AD) using complementary multimodalities can improve the quality of life and mental state of patients.
In this study, we integrated a lightweight custom convolutional neural network
(CNN) model and nature-inspired optimization techniques to enhance the performance, robustness, and stability of progress detection in AD. A multi-modal
fusion database approach was implemented, including positron emission tomography (PET) and magnetic resonance imaging (MRI) datasets, to create a fused
database. We compared the performance of custom and pre-trained deep learning models with and without optimization and found that employing natureinspired algorithms like the particle swarm optimization algorithm (PSO) algorithm significantly improved system performance. The proposed methodology,
which includes a fused multimodality database and optimization strategy, improved performance metrics such as training, validation, test accuracy, precision, and recall. Furthermore, PSO was found to improve the performance of
pre-trained models by 3-5% and custom models by up to 22%. Combining different medical imaging modalities improved the overall model performance by
2-5%. In conclusion, a customized lightweight CNN model and nature-inspired
optimization techniques can significantly enhance progress detection, leading to
better biomedical research and patient care.
Evaluation of Default Mode Network In Mild Cognitive Impairment and Alzheimer...CSCJournals
Although progressive functional brain network disorders has been one of the indication of Alzheimer's disease, The current research on aging and dementia focus on diagnostics of the cognitive changes of normal aging and Alzheimer Disease (AD), these changes known as Mild Cognitive Impairment (MCI). The default mode network (DMN) is a network of interacting brain regions known to have activity highly correlated with each other and distinct from other networks in the brain, the default mode network is active during passive rest and consists of a set of brain areas that are tightly functionally connected and distinct from other systems within the brain. Anatomically, the DMN includes the posterior cingulated cortex (PCC), dorsal and ventral medial prefrontal cortex, the lateral parietal cortex, and the medial temporal lobes. DMN involves multiple anatomical networks that converge on cortical hubs, such as the PCC, ventral medial prefrontal, and inferior parietal cortices. The aim of this study was to evaluate the default mode network functional connectivity in MCI patients. While no treatments are recommended for MCI currently, Mild Cognitive Impairment is becoming a very important subject for researchers and deserves more recognition and further study, In order to increase the ability to recognize earlier symptoms of Alzheimer's disease.
Hippocampus’s volume calculation on coronal slice’s for strengthening the dia...TELKOMNIKA JOURNAL
Alzheimer’s is one of the most common types of dementia in the world. Although not a contagious disease, this disease has many impacts, especially in socio-economic life. In diagnosing Alzheimer’s and using interview techniques, physical examination methods are also used, namely using an magnetic resonance imaging (MRI) machine to get a clear image of the patient’s brain condition, with a focus on the hippocampus and ventricular area. In this paper, we discuss the calculation of the volume of the hippocampus, especially the coronal slice, to provide information to doctors in making decisions on diagnosing the severity of Alzheimer’s. Using the basis of volume calculations, we made a 3D visualization reconstruction of the coronal hippocampus slice area in order to make it easier for doctors to analyze the condition of the hippocampus area, which in the end will be used as a recommendation in the classification of the severity of Alzheimer’s. Our experimental results show, the lower the severity, the bigger the volume, the more slices, and the longer the counting time.
Convolutional neural networks (CNN) trained using deep learning (DL) have advanced dramatically in recent years. Researchers from a variety of fields have been motivated by the success of CNNs in computer vision to develop better CNN models for use in other visually-rich settings. Successes in image classification and research have been achieved in a wide variety of domains throughout the past year. Among the many popularized image classification techniques, the detection of plant leaf diseases has received extensive research. As a result of the nature of the procedure, image quality is often degraded and distortions are introduced during the capturing of the image. In this study, we look into how various CNN models are affected by distortions. Corn-maze leaf photos from the 4,188-image corn or maize leaf Dataset (split into four categories) are under consideration. To evaluate how well they handle noise and blur, researchers have deployed pre-trained deep CNN models like visual geometry group (VGG), InceptionV3, ResNet50, and EfficientNetB0. Classification accuracy and metrics like as recall and f1-score are used to evaluate CNN performance.
Classification of Abnormalities in Brain MRI Images Using PCA and SVMIJERA Editor
The impact of digital image processing is increasing by the day for its use in the medical and research areas. Medical image classification scheme has been on the increase in order to help physicians and medical practitioners in their evaluation and analysis of diseases. Several classification schemes such as Artificial Neural Network (ANN), Bayes Classification, Support Vector Machine (SVM) and K-Means Nearest Neighbor have been used. In this paper, we evaluate and compared the performance of SVM and PCA by analyzing diseased image of the brain (Alzheimer) and normal (MRI) brain. The results show that Principal Components Analysis outperforms the Support Vector Machine in terms of training time and recognition time.
A Novel Efficient Medical Image Segmentation Methodologyaciijournal
Image segmentation plays a crucial role in many medical applications. The threshold based medical image
segmentation approach is the most common and effective method for medical image segmentation, but it
has some shortcomings such as high complexity, poor real time capability and premature convergence, etc.
To address above issues, an improved evolution strategies is proposed to use for medical image
segmentation, there are 2 populations concurrently during evolution, one focuses on local search in order
to search solutions near optimal solution, and the other population that implemented based on chaotic
theory focuses on global search so as to keep the variety of individuals and jump out from the local
maximum to overcome the problem of premature convergence. The encoding scheme, fitness function, and
evolution operators are also designed. The experimental results validated the effectiveness and efficiency of
the proposed approach.
International Journal of Engineering Research and Applications (IJERA) is an open access online peer reviewed international journal that publishes research and review articles in the fields of Computer Science, Neural Networks, Electrical Engineering, Software Engineering, Information Technology, Mechanical Engineering, Chemical Engineering, Plastic Engineering, Food Technology, Textile Engineering, Nano Technology & science, Power Electronics, Electronics & Communication Engineering, Computational mathematics, Image processing, Civil Engineering, Structural Engineering, Environmental Engineering, VLSI Testing & Low Power VLSI Design etc.
Brain Image Fusion using DWT and Laplacian Pyramid Approach and Tumor Detecti...INFOGAIN PUBLICATION
Image fusion is the process of combining important information from two or more images into a single image. The resulting image will be more enhanced than any of the input pictures. The idea of combining multiple image modalities to furnish a single, more enhanced image is well established, special fusion methods have been proposed in literature. This paper is based on image fusion using laplacian pyramid and Discreet Wavelet Transform (DWT) methods. This system uses an easy and effective algorithm for multi-focus image fusion which uses fusion rules to create fused image. Subsequently, the fused image is obtained by applying inverse discreet wavelet transform. After fused image is obtained, watershed segmentation algorithm is applied to detect the tumor part in fused image.
Classification and Segmentation of Glaucomatous Image Using Probabilistic Neu...ijsrd.com
The gradual visual field loss and there is a characteristic type of damage to the retinal nerve fiber layer associated with the progression of the disease glaucoma. Texture features within images are actively pursued for accurate and efficient glaucoma classification. Energy distribution over wavelet subband is applied to find these important texture features. In this paper, we investigate the discriminatory potential of wavelet features obtained from the Daubechies (db3), symlets (sym3), and biorthogonal (bio3.3, bio3.5, and bio3.7) wavelet filters. We propose a novel technique to extract energy signatures obtained using 2-D discrete wavelet transform, and subject these signatures to different feature ranking and feature selection strategies. Here my project aims at the use of Probabilistic Neural Network (PNN), Fuzzy C-means (FCM) and K-means helps for the detection of glaucoma disease. For this, fuzzy c-means clustering algorithm and k-means algorithm is used. Fuzzy c-means results faster and reliably good clustering when compare to k-means.
Transcript: Selling digital books in 2024: Insights from industry leaders - T...BookNet Canada
The publishing industry has been selling digital audiobooks and ebooks for over a decade and has found its groove. What’s changed? What has stayed the same? Where do we go from here? Join a group of leading sales peers from across the industry for a conversation about the lessons learned since the popularization of digital books, best practices, digital book supply chain management, and more.
Link to video recording: https://bnctechforum.ca/sessions/selling-digital-books-in-2024-insights-from-industry-leaders/
Presented by BookNet Canada on May 28, 2024, with support from the Department of Canadian Heritage.
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024Albert Hoitingh
In this session I delve into the encryption technology used in Microsoft 365 and Microsoft Purview. Including the concepts of Customer Key and Double Key Encryption.
Accelerate your Kubernetes clusters with Varnish CachingThijs Feryn
A presentation about the usage and availability of Varnish on Kubernetes. This talk explores the capabilities of Varnish caching and shows how to use the Varnish Helm chart to deploy it to Kubernetes.
This presentation was delivered at K8SUG Singapore. See https://feryn.eu/presentations/accelerate-your-kubernetes-clusters-with-varnish-caching-k8sug-singapore-28-2024 for more details.
Generative AI Deep Dive: Advancing from Proof of Concept to ProductionAggregage
Join Maher Hanafi, VP of Engineering at Betterworks, in this new session where he'll share a practical framework to transform Gen AI prototypes into impactful products! He'll delve into the complexities of data collection and management, model selection and optimization, and ensuring security, scalability, and responsible use.
LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...DanBrown980551
Do you want to learn how to model and simulate an electrical network from scratch in under an hour?
Then welcome to this PowSyBl workshop, hosted by Rte, the French Transmission System Operator (TSO)!
During the webinar, you will discover the PowSyBl ecosystem as well as handle and study an electrical network through an interactive Python notebook.
PowSyBl is an open source project hosted by LF Energy, which offers a comprehensive set of features for electrical grid modelling and simulation. Among other advanced features, PowSyBl provides:
- A fully editable and extendable library for grid component modelling;
- Visualization tools to display your network;
- Grid simulation tools, such as power flows, security analyses (with or without remedial actions) and sensitivity analyses;
The framework is mostly written in Java, with a Python binding so that Python developers can access PowSyBl functionalities as well.
What you will learn during the webinar:
- For beginners: discover PowSyBl's functionalities through a quick general presentation and the notebook, without needing any expert coding skills;
- For advanced developers: master the skills to efficiently apply PowSyBl functionalities to your real-world scenarios.
UiPath Test Automation using UiPath Test Suite series, part 3DianaGray10
Welcome to UiPath Test Automation using UiPath Test Suite series part 3. In this session, we will cover desktop automation along with UI automation.
Topics covered:
UI automation Introduction,
UI automation Sample
Desktop automation flow
Pradeep Chinnala, Senior Consultant Automation Developer @WonderBotz and UiPath MVP
Deepak Rai, Automation Practice Lead, Boundaryless Group and UiPath MVP
State of ICS and IoT Cyber Threat Landscape Report 2024 previewPrayukth K V
The IoT and OT threat landscape report has been prepared by the Threat Research Team at Sectrio using data from Sectrio, cyber threat intelligence farming facilities spread across over 85 cities around the world. In addition, Sectrio also runs AI-based advanced threat and payload engagement facilities that serve as sinks to attract and engage sophisticated threat actors, and newer malware including new variants and latent threats that are at an earlier stage of development.
The latest edition of the OT/ICS and IoT security Threat Landscape Report 2024 also covers:
State of global ICS asset and network exposure
Sectoral targets and attacks as well as the cost of ransom
Global APT activity, AI usage, actor and tactic profiles, and implications
Rise in volumes of AI-powered cyberattacks
Major cyber events in 2024
Malware and malicious payload trends
Cyberattack types and targets
Vulnerability exploit attempts on CVEs
Attacks on counties – USA
Expansion of bot farms – how, where, and why
In-depth analysis of the cyber threat landscape across North America, South America, Europe, APAC, and the Middle East
Why are attacks on smart factories rising?
Cyber risk predictions
Axis of attacks – Europe
Systemic attacks in the Middle East
Download the full report from here:
https://sectrio.com/resources/ot-threat-landscape-reports/sectrio-releases-ot-ics-and-iot-security-threat-landscape-report-2024/
Builder.ai Founder Sachin Dev Duggal's Strategic Approach to Create an Innova...Ramesh Iyer
In today's fast-changing business world, Companies that adapt and embrace new ideas often need help to keep up with the competition. However, fostering a culture of innovation takes much work. It takes vision, leadership and willingness to take risks in the right proportion. Sachin Dev Duggal, co-founder of Builder.ai, has perfected the art of this balance, creating a company culture where creativity and growth are nurtured at each stage.
Elevating Tactical DDD Patterns Through Object CalisthenicsDorra BARTAGUIZ
After immersing yourself in the blue book and its red counterpart, attending DDD-focused conferences, and applying tactical patterns, you're left with a crucial question: How do I ensure my design is effective? Tactical patterns within Domain-Driven Design (DDD) serve as guiding principles for creating clear and manageable domain models. However, achieving success with these patterns requires additional guidance. Interestingly, we've observed that a set of constraints initially designed for training purposes remarkably aligns with effective pattern implementation, offering a more ‘mechanical’ approach. Let's explore together how Object Calisthenics can elevate the design of your tactical DDD patterns, offering concrete help for those venturing into DDD for the first time!
Le nuove frontiere dell'AI nell'RPA con UiPath Autopilot™UiPathCommunity
In questo evento online gratuito, organizzato dalla Community Italiana di UiPath, potrai esplorare le nuove funzionalità di Autopilot, il tool che integra l'Intelligenza Artificiale nei processi di sviluppo e utilizzo delle Automazioni.
📕 Vedremo insieme alcuni esempi dell'utilizzo di Autopilot in diversi tool della Suite UiPath:
Autopilot per Studio Web
Autopilot per Studio
Autopilot per Apps
Clipboard AI
GenAI applicata alla Document Understanding
👨🏫👨💻 Speakers:
Stefano Negro, UiPath MVPx3, RPA Tech Lead @ BSP Consultant
Flavio Martinelli, UiPath MVP 2023, Technical Account Manager @UiPath
Andrei Tasca, RPA Solutions Team Lead @NTT Data
Observability Concepts EVERY Developer Should Know -- DeveloperWeek Europe.pdfPaige Cruz
Monitoring and observability aren’t traditionally found in software curriculums and many of us cobble this knowledge together from whatever vendor or ecosystem we were first introduced to and whatever is a part of your current company’s observability stack.
While the dev and ops silo continues to crumble….many organizations still relegate monitoring & observability as the purview of ops, infra and SRE teams. This is a mistake - achieving a highly observable system requires collaboration up and down the stack.
I, a former op, would like to extend an invitation to all application developers to join the observability party will share these foundational concepts to build on:
Observability Concepts EVERY Developer Should Know -- DeveloperWeek Europe.pdf
D04472327
1. IOSR Journal of Engineering (IOSRJEN) www.iosrjen.org
ISSN (e): 2250-3021, ISSN (p): 2278-8719
Vol. 4, Issue 4 (April. 2014), ||V7|| PP 23-27
International organization of Scientific Research 23 | P a g e
Feature Extraction for Alzheimer’s Disease
Sangam Mhatre
(Department of Instrumentation Engineering, R.A.I.T, Navi Mumbai)
Sangam9838@gmail.com
Abstract: - Alzheimers disease (AD) is a progressive and degenerative disease that affects brain cells, and its
early diagnosis has been essential for appropriate intervention by health professionals. Noninvasive in vivo
neuroimaging techniques such as magnetic resonance imaging (MRI) and positron emission tomography (PET)
are commonly used to diagnose and monitor the progression of the disease and the effect of treatment. In this
regard, the problem of developing computer aided diagnosis (CAD) tools to distinguish images with AD from
those of normal brains. Computer Aided Diagnosis is applied to the field of medical image diagnosis. It can
improve the accuracy and accordance of the diagnosis result. According to the analysis of the features of the
images information, we get the result. If the features extracted are carefully chosen it is expected that the
features set will extract the relevant information from the input data in order to perform the desired task using
this reduced representation instead of the full size input. The vast majority of 3D brain image-based computer
aided diagnosis methods implemented so far relied simply on voxel intensity, as feature. Classification is
accomplished through Support Vector Machines, after an automatic feature selection step.
Keywords: - Alzheimer’s disease, feature extraction, feature transformation, voxel intensity
I. INTRODUCTION
Alzheimer’s disease (AD), named after the German physician Alois Alzheimer, is a condition defined
by progressive dementia and the abundant presence in the brain of characteristic neuropathological structures.
The earliest symptom is generally memory loss, followed by further functional and cognitive decline, such that
patients become gradually less able to perform even basic tasks. There is currently no disease-modifying therapy
for AD however, symptomatic treatments can help patients to maintain mental function and manage the
behavioural symptoms. Ongoing clinical trials are focused on the development of new treatments, including
those aimed at lowering the risk of developing the disease or delaying its onset and progression . As illustrated
in Figure 1.1, changes associated with AD are thought to start occurring many years before the onset of clinical
symptoms. Any disease-modifying or causal therapy would therefore likely be of greatest benefit to
asymptomatic individuals at high risk of developing AD, so-called pre-symptomatic patients. A diagnosis of AD
is made according to consensus such as the NINCDS-ADRDA Alzheimers Criteria , which provide guidelines
for the classification of patients as having definite, probable, or possible AD. A diagnosis of definite AD
requires that neuropathological findings be confirmed by a direct analysis of brain tissue samples, which may be
obtained either at autopsy or from a brain biopsy. A delay of one year in both disease onset and progression
would reduce the number of AD cases in 2050 by an estimated 10% . The early identification of presymptomatic
patients is therefore important to allow the recruitment of appropriate participants for clinical trials. If a
successful disease-modifying therapy for AD were to be developed, early identification would become even
more important to allow targeting of patients for whom the treatment may be most effective.
Figure 1.1: An illustrative timeline of AD progression.
2. Feature Extraction for Alzheimer’s Disease
International organization of Scientific Research 24 | P a g e
II. FEATURE SELECTION AND DIMENSIONALITY REDUCTION
Dimensionality reduction is one additional component common to most CAD systems both for the ones
that use the whole brain and for those that use ROIs. The grounds for this step are linked, once again, to the high
dimensionality, low sample size problem. To get a rough idea of the gap between the number of features and the
sample size, in the whole brain based systems, the number of voxels easily exceeded tens or even hundreds of
thousands, and in the ROIs based systems, this number, despite being smaller, reached a few hundreds in the
simplest setting found in the literature. On the other hand, the cardinality of datasets available for study was
usually smaller than 200.
Distinct approaches have been tested regarding this problem, including methods that study linear
combinations of the original variables like Principal Component Analysis (PCA), Linear Discriminant Analysis
(LDA) or Nonnegative Matrix Factorization (NMF), and feature selection procedures, more specifically ranking
algorithms that assign one measure of relevance to each feature to select the most important ones. From the
measures of relevance found in the literature, one can highlight the mutual information , correlation coefficients,
the Fisher Discriminant Ratio (FDR) and the absolute value of the two-sample t-test statistic.
III. FEATURE EXTRACTION AND FEATURE TRANSFORMATON
The main objective of the current paper is to build and study a system for the computer-aided diagnosis
of Alzheimer’s disease, using three-dimensional images produced by the FDG-PET neuroimaging technique.
For the purpose of feature extraction different approaches were studied. The first uses voxels intensities (VI)
which are the features obtained directly from the FDG-PET scan with no further processing and the scale-space
of the FDG-PET images was also considered.
Voxel Intensity
Voxel intensity features are obtained directly from the PET scan and its value V (x, y, z) is a direct
measure of the FDG uptake detected in a certain voxel. The image database used in the present work, which was
retrieved from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database, had already undergone a
preprocessing stage, resulting in a co-registered and normalized set of images with identical dimensions,
specifically, 128 by 128 by 60. The domain of V (x, y, z), denoted by B, can be stated as follows:
(3.1)
Only one more preprocessing step was carried out on the original images before the feature selection
phase. Its aim was to ignore all voxels that lie outside the brain, reducing substantially the dimensionality of the
input patterns. To build a binary mask M(x, y, z), where every position inside the brain is set to true or otherwise
set to false. First, an average brain was calculated using the whole VI database and then, the subsequent volume
was thresholded at 5% of the maximum value. The threshold was determined empirically so that the brain mask
would adapt correctly to the brain. The output of this preprocessing step is illustrated in Figure 3.1.
Figure 3.1: Binary mask of the brain. On the left, one example of an axial cut of an arbitrary patient. On the right, the same example but with
all voxels outside the brain removed.
Although the procedure just described could be seen as a feature selection routine, it was presented
here because it is constructed based on the VI features and should be used as a preprocessing step, not as the
only feature selection operation.
3. Feature Extraction for Alzheimer’s Disease
International organization of Scientific Research 25 | P a g e
Scale-Space Expansion
A common characteristic of images is that neighboring pixels are highly correlated and this remains
true for the VI features presented in the previous section. As a consequence, a great deal of information present
in the original volume is redundant, which can reduce the system’s performance due to the curse of
dimensionality. In order to overcome this probable source of performance degradation, a Gaussian pyramid
representation of the scale-space of the brain volumes was studied. This pyramid provides representations of an
image, in this case of a volume, at different scales and resolutions.
A low-pass pyramid is generated by the repetition of two steps. The first one smooths the volume with
an appropriate filter, followed by a subsampling step usually by a factor of two in each direction. More formally,
the pyramid is recursively defined as follows:
(3.2)
where Vl represents the level l of the pyramid and w(m, n, o) is a weighting function, also known as “generating
kernel”. In this definition of the scale-space, it should be noted that both steps are merged in equation (3.2), the
width of the generating kernel was set to five (m, n and o range from -2 up to 2) and the subsampling factor was
set to two. A more general definition was not used in order to ease the presentation. On the other hand, the
domain of each layer, denoted by B0, B1, B2 and so on, can be defined as follows:
(3.3)
where l = 1, 2, . . ., the notation stands for the ceiling of a number and xM,l−1, yM,l−1 and zM,l−1 are the
domain upper limits of x, y and z coordinates in the previous level. Usually, the generating kernel is constructed
so that three properties hold:
• Separability: w(m, n, o) = w(m) · w(n) · w(o);
• Symmetry: w(−m) = w(m), w(−n) = w(n) and w(−o) = w(o);
• Each node at level l should contribute the same total weight to nodes at level l + 1;
The generating kernel used in work is the one where w(m) = w(n) = w(o) = 1/16 [1 4 6 4 1], which
resembles the Gaussian function and thus gives rise to the Gaussian pyramid’s name. Figure 3.2 shows the
generation of the first three levels of the Gaussian pyramid.
Figure 3.2: Generation of three levels of the Gaussian pyramid, which are illustrated in the images on the left. The images on the right show
the output of the intermediate smoothing step. Note that, although only one slice of each brain is depicted, both smoothing and subsampling
steps take place in the three-dimensional volume.
4. Feature Extraction for Alzheimer’s Disease
International organization of Scientific Research 26 | P a g e
Local Variance
Although VI is the most evident feature to use, since AD is characterized by a diminished brain
metabolism and this feature measures that same information, other attributes of the volume produced by the PET
scan might also contain discriminative information. In this section, a transformation of the original volume that
captures its local contrast will be presented. The image total variance is one of the many definitions of contrast,
known as RMS contrast. However to measure local contrast, one needs to consider the RMS’ local counterpart.
In fact, areas with low contrast are fundamentally flat, having therefore low variance, while areas near corners
or edges have higher contrast and also higher local variance.
In the present work, the 3D nature of the biomarker that is being used for the CAD of AD demands the
usage of the variance over a 3D neighborhood, which can be simply defined as the variance of P equidistant
sample points xp = (xp, yp, zp) with voxel intensities Vp that lie on a sphere with a predefined radius R and
centered at a given point xc = (xc, yc, zc) (Figure 3.3). This definition of neighbor set has one main advantage: it
allows for the extraction of features at different scales by varying the radius R. The operator VARP,R can
therefore be defined as:
(3.4)
Hence, if one varies the center xc, the local contrast of each voxel’s neighborhood can be computed.
Figure 3.3: Neighbor sets for four different numbers of sampling points. Each neighbor point (red) lies on a sphere and is at the same
distance to its closest samples. To be more precise, the equidistant property only holds completely accurate for the cases P = 8 and P = 12,
while for the cases P = 24 and P = 98 an approximation is used.
Since the voxel intensities in use are sampled at specific coordinates on the sphere, i.e., most samples
do not belong to the VI domain, B, an interpolated value of Vp must often be calculated. In this case, trilinear
interpolation was applied. Figure 3.4 shows a transformation of an input brain volume based on the operator
VAR24,1. Note also that one can and should use the binary mask M(x, y, z) to reduce the number of features.
Despite the simple formulation of this operator, equidistant sampling on the sphere has no exact solution for
most number of sampling points, and the general task is known as Fejes Toth’s problem. Nevertheless, there are
some numerical approximations available. It is stressed that the exact position of the sampling points is not
crucial for this type of feature.
Figure 3.4: Transformation of an input brain volume by the local variance operator based on a neighbor set of 24 samples located on a
sphere of radius 1. After the transformation, voxels outside the brain were removed using the brain mask M described in section 3.2. Only
one axial cut is depicted for visualization purposes.
5. Feature Extraction for Alzheimer’s Disease
International organization of Scientific Research 27 | P a g e
IV. CONCLUSION
Regarding the VI features from the FDG-PET scan, the scale-space of the brain images was studied,
allowing for a reduction of the number of features. Note that the number of voxels is reduced by a factor of eight
(two in each space direction) in each level. The dimensionality reduction achieved by the pyramid
representation of the scale-space has three main objectives. First, there is the possibility of improving the
system’s performance by alleviating the small sample size problem. Second, it reduces the time consumed at the
training stage and third, it allows studying how much data could be discarded without jeopardizing the system’s
performance. A measure of local contrast, LVAR, was also introduced as the sample variance computed on a
given 3D neighborhood. Since this type of feature estimates the variance of the image intensity on a given
sphere for each position of the PET scan, the number of neighbors P has to be set high enough so that good
estimates of the true variance can be computed.
REFERENCES:
[1] J. Ramrez D. Salas-Gonzalez M.M. Lpez F. Segovia I.A. Illn, J.M. Grriz. “18F-FDG PET imaging analysis
for computer aided Alzheimers diagnosis”. Information Sciences, 181(11):903–916, 2010 Elsevier.
[2] A. Lassl D. Salas-Gonzalez E.W. Lang C. G. Puntonet I. AlvarezM. Lpez J. M. Grriz, J. Ramirez and M.
Gmez-Rio. “Automatic computer aided diagnosis tool using component-based SVM”. Nuclear Science
Symposium Conference Record (NSS08), IEEE, 16(13):4392–4395, 2008.
[3] J. M. Grriz J. Ramrez D. Salas-Gonzlez I. lvarez P. Padilla, M. Lpez. “NMF-SVM Based CAD Tool
Applied to Functional Brain Images for the Diagnosis of Alzheimers Disease”. IEEE TRANSACTIONS
ON MEDICAL IMAGING,FEBRUARY 2012, 31(2):1967–1976, 2012.
[4] M. Silveira and J. Marques. “Boosting Alzheimer disease diagnosis using PET images, in Pattern
Recognition (ICPR10)”,. Proceedings of the 2010 20th International Conference on. IEEE Computer
Society,, 16(13):2556–2559,, 2010.
[5] S. Eberl M. Fulham Y. Xia, L. Wen and D. Feng. “Genetic algorithm-based PCA eigenvector selection and
weighting for automated identification of dementia using FDG-PET imaging”. Engineering in Medicine and
Biology Society, 2008. EMBS 2008. 30th Annual International Conference of the IEEE, 16(13):4812–4815,
2008.
[6] UCLA. Laboratory of neuro imaging. http://adni.loni.ucla.edu/about/, December 1,2012.