Sami Ur Rahman

Combining short-axis and long-axis cardiac MR images by applying a super-resolution reconstruction algorithm

In cardiac MR images the slice thickness is normally greater than the pixel size within the slices. In general, better segmentation and analysis results can be expected for isotropic high-resolution (HR) data sets. If two orthogonal data sets, e. g. short-axis (SA) and long-axis (LA) volumes are combined, an increase in resolution can be obtained. In this work we employ a super-resolution reconstruction (SRR) algorithm for computing high-resolution data sets from two orthogonal SA and LA volumes. In contrast to a simple averaging of both data in the overlapping region, we apply a maximum a posteriori approach. There, an observation model is employed for estimating an HR image that best reproduces the two low-resolution input data sets. For testing the SRR approach, we use clinical MRI data with an in-plane resolution of 1.5 mm×1.5 mm and a slice thickness of 8 mm. We show that the results obtained with our approach are superior to currently used averaging techniques. Due to the fact that the heart deforms over the cardiac cycle, we investigate further, how the replacement of a rigid registration by a deformable registration as preprocessing step improves the quality of the final HR image data. We conclude that image quality is dramatically enhanced by applying an SRR technique especially for cardiac MR images where the resolution in slice-selection direction is about five times lower than within the slices.

Towards Patient Specific Catheter Selection: Computation of Aortic Geometry Based on Fused MRI Data

In coronary angiography, a catheters tip has to be directed through the aorta towards the ostium - the region where the coronary arteries arise. Due to the anatomical variation in different humans, there is no common catheter which can be used for all patients. Thus, in a trial and error procedure cardiologists find a catheter that fits to the patients anatomy. To replace this time consuming approach by providing a computer aided planning tool to be used prior to the intervention is the focus of our work. First of all, it is necessary for such a system to derive geometrical parameters for the patients aorta as well as for the different available catheters. Based thereon, the best fitting catheter can be selected. In this paper, we discuss the first step: the computation of geometrical parameters from the patients image data. Due to the setting defined by our clinical partner, two MRI data sets are acquired and should be used for the computation. This requires a specific image processing pipeline which we present here and which has to our knowledge not been proposed so far. Furthermore, we show first results obtained for real clinical data sets and discuss the subsequent steps for the development of the catheter selection tool.

Detection of the Top Anemic Diseases in Blood Smear Images Using Image Quantization Followed by Ensemble of Classifiers

Anemia is a condition caused due to the deficiency of Red Blood Cells (RBCs) and hemoglobin in blood. It is an indication to a specific disorder in the human body. Different types of anemic diseases infect the shapes of Red Blood Cells in different ways and the infected cells form various geometric shapes, such as elongated ellipse, triangular shapes, cut circles, boundary interruption in ellipse or circle etc. Leveraging these shapes the type of anemia can easily be identified. We have used various boundary based shape descriptors like shape signatures and color profiles as features for the infected RBCs recognition. The algorithm is followed by preprocessing steps like color channel separation, segmentation through quantization, feature extraction and finally classification of Red Blood Cells and the diseases associated with them. We have achieved 92 % accuracy and the proposed method is cost effective and easy to use.

An experimental evaluation of different face recognition algorithms using closed circuit Television images

Bio-metric based recognition is rapidly replacing the non-biometric based recognition systems. Face recognition is one of the most important bio-metric based recognition technique. Different algorithms exist for face recognition that works well on high resolution digital images. The aim of this work is to check whether these existing face recognition algorithms work on low quality Closed Circuit Television images? We have considered three such algorithms. These algorithms are Artificial Neural Network, Principal Component Analysis and Single Value Decomposition. We have experimentally evaluated these for Closed Circuit Television images. Our experiments show that using the CCTV images the accuracy rate of ANN for face recognition is 85%, for PCA the accuracy rate is 75% and for SVD the accuracy rate is 65%. The experiments also show that the accuracy of all the three algorithms reduces when the number of input faces increases.

Automated Marker Augmentation and Path Discovery in Indoor Navigation for Visually Impaired

The past two decades have seen abundance of applications of Augmented Reality (AR), from gaming to medical, engineering, and academic fields. Certain work has been done to employ AR techniques for assisting the blind and visually impaired people to navigate in large indoor environments. This research contributes to the existing solutions by providing a viable technique, using merely a mobile phone camera and fiducial markers. The markers are detected and connected to generate a floor plan with the help of our proposed automatic path generation algorithm. Similarly, path augmentation algorithm efficiently populate the generated path with auditory and textual information. The proposed solution also provides a way to edit an already stored path when we need to extend the floor plan for inclusion of additional paths. An android application is developed to implement these algorithms. Time benchmarking the system shows effective results in automatic path generation, path augmentation, and path extension processes.

Swarm Based-Artificial Neural System for Human Health Data Classification

Cancer is an important medical disorder, which is not a single disease but a cluster more than 200 different serious medical complications. The accurate prediction in patients with early-stage cancer is of significant importance to reduce the mortality rate of those patients. Therefore, biologically inspired approaches that are motivated by the natural behaviors of swarms are used in this work. They are robust, easy to implement, and has few setting parameters. However, one disadvantage is that they are of slow convergence, which is due to the poor exploration and exploitation processes. To overcome this deficiency of the traditional algorithm, we propose the Global Guided Artificial Bee Colony (GGABC) and Hybrid Guided Artificial Bee Colony (HGABC) algorithms; which employ new hybrid population-based meta-heuristic approaches, simulated by the foraging behaviors of global best and guided honey bees. In this chapter, GGABC and HGABC algorithms are proposed in order to determine whether patients have cancer or not. The simulation results of the proposed approaches were compared with other algorithms including ABC, Guided ABC, and Global ABC. The classification results of cancer by GGABC and HGABC models are highly accurate in contrast to the results given by the benchmarked algorithms.

Enhancing super-resolution reconstructed image quality in 3D MR images using simulated annealing

Super-resolution reconstruction (SRR) algorithms are used for getting high-resolution (HR) data from low-resolution observations. In Maximum a posteriori (MAP) based SRR the observation model is employed for estimating a HR image that best reproduces the two low-resolution input data sets. The parameters of the prior play a significant role in the MAP based SRR. This work concentrates on the investigation of the influence of one such parameter, called temperature, on the reconstructed 3D MR images. The existing approaches on SRR in 3D MR images use a constant value for this parameter. We use a cooling schedule similar to simulated annealing for computing the value of the temperature parameter at each iteration of the SRR. We have used 3D MR cardiac data sets in our experiments and have shown that the iterative computation of the temperature which resembles simulated annealing delivers better results.

Simulation based patient-specific optimal catheter selection for right coronary angiography

Selecting the best catheter prior to coronary angiography significantly reduces the exposure time to radiation as well as the risk of artery punctures and internal bleeding. In this paper we describe a simulation based technique for selecting an optimal catheter for right coronary angiography using the Simulation Open Framework Architecture (SOFA). We simulate different catheters in a patient-specific arteries model, obtain final placement of different catheters and suggest an optimally placed catheter. The patient-specific arteries model is computed from the patient image data acquired prior to the intervention and the catheters are modeled using Finite Element Method (FEM).