Abstract: Transepiphyseal tumor resection is a common surgical procedure in patients with malignant bone tumors. The aim of this study is to develop and validate a computer-assisted method for selecting the most appropriate allograft from a cadaver bone bank. Fifty tibiae and femora were 3D reconstructed from computed tomography (CT) images. A transepiphyseal resection was applied to all of them in a virtual environment. A tool was developed and evaluated that compares each metaphyseal piece against all other bones in the data bank. This is done through a template matching process, where the template is extracted from the contralateral healthy bone of the same patient. The method was validated using surface distance metrics and statistical tests comparing it against manual methods. The developed algorithm was able to accurately detect the bone segment that best matches the patient's anatomy. The automatic method showed improvement over the manual counterpart. The proposed method also substantially reduced computation time when compared to state-of-the-art methods as well as the manual selection. Our findings suggest that the accuracy, robustness, and speed of the developed method are suitable for clinical trials and that it can be readily applied for preoperative allograft selection.
Abstract: In this paper we present a new population-based method for the design of bone xation plates. Standard pre-contoured plates are designed based on the mean shape of a certain population. We propose a computational process to design implants while reducing the amount of required intra-operative shaping, thus reducing the mechanical stresses applied to the plate. A bending and torsion model was used to measure and minimize the necessary intra-operative deformation. The method was applied and validated on a population of 200 femurs that was further augmented with a statistical shape model. The obtained results showed substantial reduction in the bending and torsion needed to shape the new design into any bone in the population when compared to the standard mean-based plates.
Abstract: This article describes the design and development process of an eye tracking-based computer system that benefits from the intact ocular motor control of the completely locked-in patient to provide him or her with an alternative means of communication. A completely locked-in patient is an individual who has lost all types of motor control and communication ability with people in his or her environment. This developed system uses a head-mounted web camera to capture real-time images of the patient’s eye. These images are then passed to a program, developed using Matlab®, which processes them and computes the coordinates of the pupil position. The program then sends commands to an interactive JAVA™-based interface, which provides the patient with a matrix of pictograms representing the most essential daily communication activities. When a pictogram is activated (clicked), the system plays back an audible statement, recorded in any language, reflecting the desired activity. Ten healthy adult volunteers, free from any musculoskeletal or neurological disorders, participated in the validation of the system. Validation results revealed a system accuracy of 96.11 ± 5.58 % and repeatability of 94.44 ± 2.51%. The rehabilitative system developed in this project offers the locked-in patient, of any social class, the ability of simple yet effective communication. The advantages of this system over existing systems are low cost, low processing power, ease of operation, little training requirements, minimal disturbance to the patient, and ease of customization to any mother tongue.
Abstract: The work described herein focuses on numerically simulating the blood flow and contrast agent distribution in patient-specific representations of human coronary arteries obtained through segmentation of computed tomography angiography (CTA) images. The goal of the project is to research the results of the simulations to support and enhance the outcome of a currently available automatic atherosclerotic plaque detection algorithm. Atherosclerosis is a progressive disease that develops in the circulatory system and that could cause fatal complications such as blockage or burst of a certain blood vessel. Several simulation approaches involving the solution of steady and pulsatile blood flow as well as the concentration of contrast agent have been implemented and tested. The simulations consist of solving the Navier-Stokes equations in order to reconstruct a steady-state or time-varying velocity field, and consequently solving the transport or advection-diffusion equation to obtain a concentration field representing the predicted distribution of contrast agent within the lumen of the artery. Information extracted from two clinical techniques, namely the bolus tracking and the test bolus technique, have been employed in order to mimic the patientspecific temporal variation of the contrast agent concentration at the inlet of the coronary tree. In many cases, analysis of the results showed good agreement between the simulation outcome and the intensity distribution in the CT images. However, in some other cases, they were weakly comparable. At false positive locations where comparisons were successful, the predicted concentration of contrast agent was higher than its surroundings by 14.91% ± 9.55% (mean ± SD), whereas at normal locations the difference between the highest and lowest concentration values in the same cross-section was 3.30% ± 1.65% (mean ± SD). These findings can be used to eliminate false positives by injecting the knowledge acquired from the CFD simulations into an automatic plaque detection algorithm.
Abstract: This report describes the design and development process of a pattern analysis software that automatically classifies atherosclerosis lesions within the coronary artery into one of two classes, namely, hard and mixed plaques. Atherosclerosis is a disease that hits the circulatory system and with time, could cause fatal complications such as blockage or burst of a certain blood vessel. The datasets used in this work are dual-source computed tomography angiography (CTA) images that were examined in advance by an expert observer who indicated the spatial localization as well as the diagnosed type of all present plaques. A total of 16 different features have been extracted and used in the classification stage. Computation of all 16 features has been done through intensity-based approaches, i.e., based on the intensity distribution of voxels within a certain volume of interest. Two pattern classification algorithms have been used, in particular, the k-Nearest Neighbor (k-NN) and the Support Vector Machine (SVM) classifiers. Different parameters have been tested for each of the algorithms, thus generating a thorough comparison and validation of the results. The initial configuration of the data consisted of 30 training scans containing 194 plaques, 43 out of which were mixed lesions. Whereas the testing set consisted of 50 scans containing 324 plaques. The mixed plaques among the testing set counted a total of 77. With the optimal parameters, the k-NN classifier was able to achieve an accuracy of 79.32% on the expense of seven false positives, whereas the SVM produced an accuracy of 79.01% wile resulting in six false positive outputs.
Abstract: This report describes the design and development process of an eye tracking-based computer system that benefits from the intact ocular motor control of the completely locked-in patient to provide him or her with an alternative means of communication. A completely locked-in patient is an individual who has lost all types of motor control and communication ability with people in his or her environment. The developed system uses a head-mounted web camera to capture real-time images of the patient’s eye. These images are then passed to a program, developed using MATLAB®, which processes them and computes the coordinates of the pupil position. The program then sends commands to an interactive JAVA™-based interface, which provides the patient with a matrix of pictograms representing the most essential daily communication activities. When a pictogram is activated (clicked), the system plays back an audible statement, recorded in any language, reflecting the desired activity. A 2 × 3 pictorial matrix has provided optimal results and thus was adopted in this application. The rehabilitative system developed in this project, in both its hardware and software components, is intended to minimize the physical and psychological irritations to the patient. The system offers the locked-in patient, of any social class, the ability of simple yet effective communication. The advantages of this system over existing systems are low cost, low processing power, ease of operation, little training requirements, minimal disturbance to the patient, and ease of customization to any mother tongue.
