Search

Purpose: Pneumonia detection is usually done by specialized highly trained specialists through the review of chest radiographs combined with vital signs, laboratory exams, and medical history. Though the presence of pneumonia in images appears in the form of areas of high opacity in the lungs which is difficult to differentiate from opaque areas caused by other conditions. The medical diagnosis of pneumonia is a very costly and time-consuming process. The motivation of this work is to automate the diagnosis process of pneumonia through image processing. In this paper, an automated pneumonia detection system is proposed using improved deep residual networks (ResNet) architectures, which are tested on the dataset of 30,227 DICOM Chest X-rays. Dataset was divided into 80% training and 20% testing, with 20% of the 80% used for training dedicated to validation. Methods: Two residual network models were used (Version 1 & Version 2), and results were also compared with three different CNN models as well as methods found in recent literature. Results: The overall results indicate that the proposed ResNet (Version 2) method achieves higher accuracy than convolution neural networks and other recently proposed methods (Table 5). The proposed ResNet network of a depth of 110 and a batch size of 16 with epochs 80 achieved an average accuracy of 88.67%. Conclusion: An automated method is proposed and implemented in this work for the proper diagnosis of pneumonia using images of the DICOM chest X-rays dataset. The proposed method in this paper outperforms other methods from recent literature.

Background: Detection of brain tumor is a complicated task, which requires specialized skills and interpretation techniques. Accurate brain tumor classification and segmentation from MR images provide an essential choice for medical treatments. Different objects within an MR image have similar size, shape, and density, which makes the tumor classification and segmentation even more complex.Objective: Classification of the brain MR images into tumorous and non-tumorous using deep features and different classifiers to get higher accuracy.Methods: In this study, a novel four-step process is proposed; pre-processing for image enhancement and compression, feature extraction using convolutional neural networks (CNN), classification using the multilayer perceptron and finally, tumor segmentation using enhanced fuzzy cmeans method.Results: The system is tested on 65 cases in four modalities consisting of 40,300 MR Images obtained from the BRATS-2015 dataset. These include images of 26 Low-Grade Glioma (LGG) tumor cases and 39 High-Grade Glioma (HGG) tumor cases. The proposed CNN feature-based classification technique outperforms the existing methods by achieving an average accuracy of 98.77% and a noticeable improvement in the segmentation results are measured.Conclusion: The proposed method for brain MR image classification to detect Glioma Tumor detection can be adopted as it gives better results with high accuracies.

Background: An approach based on QR decomposition, to remove speckle noise from medical ultrasound images, is presented in this paper. Methods: The speckle noisy image is segmented into small overlapping blocks. A global covariance matrix is calculated by averaging the corresponding covariances of the blocks. QR decomposition is applied to the global covariance matrix. To filter out speckle noise, the first subset of orthogonal vectors of the Q matrix is projected onto the signal subspace. The proposed approach is compared with five benchmark techniques; Homomorphic Wavelet Despeckling (HWDS), Speckle Reducing Anisotropic Diffusion (SRAD), Frost, Kuan and Probabilistic Non-Local Mean (PNLM). Results and Conclusion: When applied to different simulated and real ultrasound images, the QR based approach has secured maximum despeckling performance while maintaining optimal resolution and edge detection, and that is regardless of image size or nature of speckle; fine or rough.

Background: MRI which stands for Magnetic Resonance Imaging is commonly used to capture images of internal body organs, functionality and structure. Manual analysis is usually performed by Radiologists on a large set of MR images in order to detect brain tumor. Aims: This research aims to improve automated brain MR image classification and tumor segmentation using phase congruency. Methods: The skull part is removed from brain MR image by applying converging square algorithm and phase congruency based edge detection method. Features are then extracted from the segmented brain portion using discrete wavelet transforms. In order to minimize the extracted feature set, we applied the principal Component Analysis algorithm. The MR images are classified into tumorous and non-tumorous using Multilayer perceptron and compared with other classifiers such as K-Nearest Neighbor, Naïve Bayes, and Support Vector Machines (SVM) along with discrete cosine and discrete cosine transform features. The tumor is segmented using Fuzzy C-mean and reconstructed tumor 3D model to measure the volume, location and shape accurately. Results & Conclusion: Experimental results are obtained by testing the proposed method on a dataset of 19 patients with a total number of 2920 brain MR images. The proposed method achieved an accuracy of 99.43% for classification which is higher as compared to other current studies.