Panagiotis Kantartzis

Development of a neonate lung reconstruction algorithm using a wavelet AMG and estimated boundary form

Objective, non-invasive measures of lung maturity and development, oxygen requirements and lung function, suitable for use in small, unsedated infants, are urgently required to define the nature and severity of persisting lung disease, and to identify risk factors for developing chronic lung problems. Disorders of lung growth, maturation and control of breathing are among the most important problems faced by the neonatologists. At present, no system for continuous monitoring of neonate lung function to reduce the risk of chronic lung disease in infancy in intensive care units exists. We are in the process of developing a new integrated electrical impedance tomography (EIT) system based on wearable technology to integrate measures of the boundary diameter from the boundary form for neonates into the reconstruction algorithm. In principle, this approach could provide a reduction of image artefacts in the reconstructed image associated with incorrect boundary form assumptions. In this paper, we investigate the required accuracy of the boundary form that would be suitable to minimize artefacts in the reconstruction for neonate lung function. The number of data points needed to create the required boundary form is automatically determined using genetic algorithms. The approach presented in this paper is to assist quality of the reconstruction using different approximations to the ideal boundary form. We also investigate the use of a wavelet algebraic multi-grid (WAMG) preconditioner to reduce the reconstruction computation requirements. Results are presented that demonstrate a full 3D model is required to minimize artefact in the reconstructed image and the implementation of a WAMG for EIT.

Improving fusion with optimal weight selection in Face Recognition

Face recognition has a large number of applications, including security/counterterrorism, person identification, Internet communications, E-commerce, and computer entertainment. Although research in automatic face recognition has been conducted since the 1960s, there exist research challenges in its practical application in the terms of performance accuracy, which deteriorates significantly with changes in illumination, pose, expression and occlusions. However, these inherent limitations can be potentially alleviated by fusing biometric information based on multiple facial features. Following this vision, the work presented here offers three contributions. Firstly, we present a Face Recognition System, where diverse biometrics features such as total face, eyes, nose, mouth, etc are extracted from the face image. Secondly, we analyse a number of approaches for combining the aforementioned information at matching score level. Thirdly, we proposed a new approach, based on a recently proposed optimisation technique, the Bees Algorithm, to determine the optimal weight parameters to enhance the performance of the fusion system. Experiments on the CASIA and ORL face databases indicate that the proposed method achieves consistently high recognition rates, compared to traditional FR approaches, such as the Eigenfaces, Fisherfaces, and D-LDA methods.

Stimulation and measurement patterns versus prior information for fast 3D EIT: A breast screening case study

Imposing prior information is a typical strategy in inverse problems in return for a stable numerical algorithm. For a given imaging system configuration, Picards stability condition could be deployed as a practical measure of the performance of the system against various priors and noise contaminated measurements. Herein, we make extensive use of this measure to quantify the performance of impedance imaging systems for various injection patterns. In effect, we numerically demonstrate that by varying electrode distributions and numbers, little improvement, if any, in the performance of the impedance imaging system is recorded. In contrast, by using groups of electrodes in the 3D current injection process, a step increase in performance is obtained. Numerical results on a female breast phantom reveal that the performance measure of the imaging system is 15% for a conventional combination of stimulation and prior information, 61% for groups of electrodes and the same prior and 97% for groups of electrodes and a more accurate prior. Finally, since a smaller number of electrodes is involved in the measurement process, a smaller number of measurements is acquired. However, no compromise in the quality of the reconstructed images is observed.

Reconstruction algorithms to monitor neonate lung function

Disorders of lung growth, maturation and control of breathing are among the most important problems faced by the neonatologist. Objective, non-invasive measures of lung maturity and development, oxygen requirements and lung function, suitable for use in small, unsedated infants, are urgently required to define the nature and severity of persisting lung disease, and to identify risk factors for developing chronic lung problems. At present, no system for continuous monitoring of neonate lung function to reduce the risk of CLDI in intensive care units (ITUs) exists.We present the development of image reconstruction algorithms to monitor neonate lung function in ITU’s, and a method base on wearable technology to integrate measures of the boundary diameter from the boundary form. This approach provides a reduction of image artefacts in the reconstructed image associated with incorrect boundary form assumptions. In terms of image reconstruction, we utilise the concept of subspace invariance to design a block adaptive preconditioning scheme, which yields a smaller error norm and can provide improvements in the condition number of the coefficients matrix, as compared to incomplete Cholesky factorization, followed by the application of conjugate gradient.

A fisher information matrix interpretation of the NOSER algorithm in electrical impedance tomography

In this paper, we employ the concept of the Fisher information matrix (FIM) to reformulate and improve on the “Newtons One-Step Error Reconstructor” (NOSER) algorithm. FIM is a systematic approach for incorporating statistical properties of noise, modeling errors and multi-frequency data. The method is discussed in a maximum likelihood estimator (MLE) setting. The ill-posedness of the inverse problem is mitigated by means of a nonlinear regularization strategy. It is shown that the overall approach reduces to the maximum a posteriori estimator (MAP) with the prior (conductivity vector) described by a multivariate normal distribution. The covariance matrix of the prior is a diagonal matrix and is computed directly from the Fisher information matrix. An eigenvalue analysis is presented, revealing the advantages of using this prior to a Gaussian smoothness prior (Laplace). Reconstructions are shown using measured data obtained from a shallow breathing of an adult human subject. The reconstructions show that the FIM approach clearly improves on the original NOSER algorithm.

Wavelet preconditioning for EIT

In this paper the forward problem of Electrical Impedance Tomography (EIT) is considered. To achieve a flexible treatment of boundaries, the original two-dimensional domain is extended to a simple square one and essential boundary conditions are imposed by means of Lagrange multipliers, resulting in a saddle point system. For discretisation, we employ a biorthogonal B-Spline wavelet basis for which it can be shown that in the suggested forward EIT formulation the condition number of the system matrix is asymptotically uniformly bounded, and therefore iterative solvers converge with a speed independent of the discretisation level.

Real-time colour segmentation and autofocus in retinal images

This research is applied to the processing of retinal images. These are often corrupted with low contrast, noise, poor illumination etc. Retinal images are typically high resolution, colour data, hence there is a need to identify the important areas of the retina, through the process of segmentation. In this work, we propose a real-time colour segmentation algorithm for the extraction of the useful part of the retina. In order to quantify the quality of the segmented image, we carry out a comparison of image focus techniques. The results verify the robust-performance of the algorithm

On sparse forward solutions in non-stationary domains for the EIT imaging problem

In the forward EIT-problem numerical solutions of an elliptic partial differential equation are required. Given the arbitrary geometries encountered, the Finite Element Method (FEM) is, naturally, the method of choice. Nowadays, in EIT applications, there is an increasing demand for finer Finite Element mesh models. This in turn results to a soaring number of degrees of freedom and an excessive number of unknowns. As such, only piece-wise linear basis functions can practically be employed to maintain inexpensive computations. In addition, domain reduction and/or compression schemes are often sought to further counteract for the growing number of unknowns. In this paper, we replace the piece-wise linear with wavelet basis functions (coupled with the domain embedding method) to enable sparse approximations of the forward computations. Given that the forward solutions are repeatedly, if not extensively, utilised during the image reconstruction process, considerable computational savings can be recorded whilst maintaining O(N) forward problem complexity. We verify with numerical results that, in practice, less than 5% of the involved coefficients are actually required for computations and, hence, needs to be stored. We finalise this work by addressing the impact to the inverse problem. It is worth underlining that the proposed scheme is independent of the actual family of wavelet basis functions of compact support.

Towards non-invasive EIT imaging of domains with deformable boundaries

We investigate on the use of the Domain Embedding Method (DEM) for the forward modelling in EIT. This approach is suitably configured to overcome the model meshing bottleneck since it does not require that the mesh on the domain is adapted to the boundary surface. This is of crucial importance for, e.g., clinical applications of EIT, as it avoids tedious and time-consuming (re-)meshing procedures. The suggested DEM approach can accommodate arbitrary yet Lipschitz smooth boundary surfaces and is not limited to polygonal domains. For the discretisation purposes, we employ B-splines as they allow for arbitrary accuracy by raising the polynomial degree and are easy to implement due to their inherent piecewise polynomial structure. Numerical experiments confirm that a B-spline discretization yields, similarly to conventional Finite Difference discretizations, increasing condition numbers of the system matrix with respect to the discretisation levels. Fortunately, multiresolution ideas based on B-splines allow for optimal wavelet preconditioning.