Normal Pressure Hydrocephalus Detection Using Active Contour Coupled Ensemble Based Classifier
01 Jan 2020-pp 543-553
TL;DR: Experimental results disclosed a significant improvement in case of ensemble classifier in comparison to Support Vector Machine in terms of its performance.
Abstract: The Brain plays an imperative role in the life of human being as it manages the communication between sensory organs and muscles. Consequently, any disease related to brain should be detected at an early stage. Abundant accumulation of cerebrospinal fluid in the ventricle results to a brain disorder termed as normal pressure hydrocephalus (NPH). The current study aims to segment the ventricular part from CT brain scans and then perform classification to differentiate between the normal brain and affected brain having NPH. In the proposed method, firstly few preprocessing steps have been carried out to enhance the quality of the input CT brain image and ventricle region is cropped out. Then active contour model is employed to perform segmentation of the ventricle. Features are extracted from the segmented region and Ensemble classifier is used to classify CT brain scan into two classes namely, normal and NPH. More than hundreds of CT brain scans were analyzed during this study; area of ventricle has been used as a measure of feature extraction. Experimental results disclosed a significant improvement in case of ensemble classifier in comparison to Support Vector Machine in terms of its performance.
Citations
More filters
[...]
TL;DR: Gradual self-improvement of the AI engine narrowing down the fuzzy zone between confident positive and confident negative of diagnosis would be the key achievement of proposed continuous deep learning framework.
Abstract: Health care is a domain of essential services and requires highest accuracy in diagnosis. AI can just aid the doctors achieving the aforementioned target but not replace them. AI always requires doctor’s validation to improve itself. Application of deep learning in health informatics enables us to recognize the target disease at early stage of maturity. The primary section of the proposed piece of work helps us to automatically diagnose tuberculosis (TB) from chest X- Rays in minimal execution time and memory footprint. Next medical media screening and analysis would drive the process of automated archival strategy, tagged to self-generated meta-data. The classical deep learning method needs to be migrated to continuous deep learning framework before the final archival based on (a) collection or acquisition of periodic medical media and (b) feedback from human experts. Gradual self-improvement of the AI engine narrowing down the fuzzy zone between confident positive and confident negative of diagnosis would be the key achievement of proposed continuous deep learning framework.
References
More filters
[...]
TL;DR: A novel method for cell segmentation and identification has been proposed that incorporated marking cells in cuckoo search (CS) algorithm and experimental results established that the Kapur's entropy segmentation method based on the modified CS required the least computational time.
Abstract: Microscopic image analysis is one of the challenging tasks due to the presence of weak correlation and different segments of interest that may lead to ambiguity It is also valuable in foremost meadows of technology and medicine Identification and counting of cells play a vital role in features extraction to diagnose particular diseases precisely Different segments should be identified accurately in order to identify and to count cells in a microscope image Consequently, in the current work, a novel method for cell segmentation and identification has been proposed that incorporated marking cells Thus, a novel method based on cuckoo search after pre-processing step is employed The method is developed and evaluated on light microscope images of rats' hippocampus which used as a sample for the brain cells The proposed method can be applied on the color images directly The proposed approach incorporates the McCulloch's method for levy flight production in cuckoo search (CS) algorithm Several objective functions, namely Otsu's method, Kapur entropy and Tsallis entropy are used for segmentation In the cuckoo search process, the Otsu's between class variance, Kapur's entropy and Tsallis entropy are employed as the objective functions to be optimized Experimental results are validated by different metrics, namely the peak signal to noise ratio (PSNR), mean square error, feature similarity index and CPU running time for all the test cases The experimental results established that the Kapur's entropy segmentation method based on the modified CS required the least computational time compared to Otsu's between-class variance segmentation method and the Tsallis entropy segmentation method Nevertheless, Tsallis entropy method with optimized multi-threshold levels achieved superior performance compared to the other two segmentation methods in terms of the PSNR
71 citations
[...]
TL;DR: An automated image processing based approach for the identification of AD from MRI of the brain that has higher specificity/accuracy values despite the use of smaller feature set as compared to existing approaches is proposed.
Abstract: Structural brain imaging is playing a vital role in identification of changes that occur in brain associated with Alzheimer's disease. This paper proposes an automated image processing based approach for the identification of AD from MRI of the brain. The proposed approach is novel in a sense that it has higher specificity/accuracy values despite the use of smaller feature set as compared to existing approaches. Moreover, the proposed approach is capable of identifying AD patients in early stages. The dataset selected consists of 85 age and gender matched individuals from OASIS database. The features selected are volume of GM, WM, and CSF and size of hippocampus. Three different classification models (SVM, MLP, and J48) are used for identification of patients and controls. In addition, an ensemble of classifiers, based on majority voting, is adopted to overcome the error caused by an independent base classifier. Ten-fold cross validation strategy is applied for the evaluation of our scheme. Moreover, to evaluate the performance of proposed approach, individual features and combination of features are fed to individual classifiers and ensemble based classifier. Using size of left hippocampus as feature, the accuracy achieved with ensemble of classifiers is 93.75%, with 100% specificity and 87.5% sensitivity.
54 citations
[...]
TL;DR: This paper presents an analysis of the mathematical morphological approach with comparison to various other state-of-art techniques for addressing the problems of low contrast in images.
Abstract: Image enhancement is one of the most interesting and visually appealing areas of image processing. It involves operations such as enhancing contrast, reducing noise for improving the quality of the image. This paper presents an analysis of the mathematical morphological approach with comparison to various other state-of-art techniques for addressing the problems of low contrast in images. Histogram equalization (HE) is one of the common methods used for improving contrast in digital images. This method is simple and effective for global contrast enhancement of images but it suffers from some drawbacks. Contrast Limited Adaptive Histogram Equalization (CLAHE) enhances the local contrast of the images without the amplification of the noise. Morphological Contrast enhancement is performed using the white and black top-hat transformation. It can be performed at a single scale or at multiple scales of the structuring element. The structuring element can be of various shapes and sizes.
40 citations
[...]
TL;DR: The variational Bayes inference to brain MRI image segmentation is introduced, and a novel segmentation algorithm based on learning a cohort of local variational Gaussian mixture (LVGM) models is proposed that can segment brain MRI images more effectively and provide more precise distribution of major brain tissues.
Abstract: Measuring the distribution of major brain tissues, including the gray matter, white matter and cerebrospinal fluid (CSF), using magnetic resonance imaging (MRI) has attracted extensive research efforts. Many brain MRI image segmentation methods in the literature are based on the Gaussian mixture model (GMM), which however is not strictly followed due to the intrinsic complex nature of MRI data and may lead to less accurate results. In this paper, we introduce the variational Bayes inference to brain MRI image segmentation, and thus propose a novel segmentation algorithm based on learning a cohort of local variational Gaussian mixture (LVGM) models. By assuming all Gaussian parameters to be random variables, the LVGM model has more flexibility than GMM in characterizing the complexity of brain voxel distributions. To alleviate the impact of bias field, we train each LVGM model on a sampled small data volume and linearly combine the trained models to classify each brain voxel. We also construct a co-registered probabilistic brain atlas for each MRI image to incorporate the prior knowledge about brain anatomy into the segmentation process. The proposed LVGM learning algorithm has been evaluated against five state-of-the-art brain MRI image segmentation methods on both synthetic and clinical data. Our results suggest that the LVGM algorithm can segment brain MRI images more effectively and provide more precise distribution of major brain tissues.
40 citations
[...]
TL;DR: An efficient MRI brain image analysis method to efficiently deal with segmentation and classification process for brain tumour analysis with use of feature extraction methods, so this method can yield the better result of brain tumours diagnosis in advance where this method using in medical fields.
Abstract: Background: Magnetic Resonance Images (MRI) is an important medical diagnosis tool for the detection of tumours in brain as it provides the detailed information associated to the anatomical structures of the brain. MRI images help the radiologist to find the presence of abnormal cell growths or tumours. MRI image analysis plays a vital role in diagnosis of brain tumours in the earlier stages and treatment of diseases. Methods: Therefore, this paper introduces an efficient MRI brain image analysis method, where, the MRI brain images are classified into normal, non cancerous (benign) brain tumour and cancerous (malignant) brain tumour. This proposed method follows four steps, 1. Pre-processing, 2. Segmentation, 3. Textural and shape feature extraction and 4. Classification. In this proposed MRI image analysis using the region based Active Contour Method (ACM) used for segmentation and Artificial Neural Network (ANN) based Levenberg-Marquardt (LM) algorithm used for classification process, which used to efficiently classify the MRI image as normal and Tumourous. Findings: The results revealed that the proposed MRI brain image tumour diagnosis process is accurate, fast and robust. The classifier based MRI brain image processing approach produced the best MRI brain image classification with use of feature extraction and segmentation results, in terms of accuracy. Best overall classification accuracy results were obtained using the given DioCom Images; The performance results proven that there is not sufficient result given to the classification process when it perform separately. With the use of ACM segmentation and feature extraction approaches, the proposed LM classification approach provides better classification accuracy than the existing approach. Application: The proposed MRI image based brain tumour analysis would efficiently deal with segmentation and classification process for brain tumour analysis with use of feature extraction methods, so this method can yield the better result of brain tumour diagnosis in advance where this method using in medical fields.
37 citations
Related Papers (5)
[...]
[...]