Search

Abstract Objectives: To audit services for prenatal diagnosis for haemoglobin disorders in the United Kingdom. Design: Comparison of the annual number of cases recorded in a United Kingdom register of [...]

Coronary bypass surgery is perhaps the most investigated surgical procedure (1) with clear end points for outcome such as operative mortality. Consequently, it has long been regarded as a sentinel [...]

This study was to investigate the distribution and clinical characteristics of teeth diagnosed with MIH at surface and defect type level in a cohort of German children. The study cohort included 242 children diagnosed with MIH which had been recorded during the compulsory dental school examinations of 20 German primary schools. The subjects had been enrolled by cluster sampling. All children attended the second to fourth grade (age 7–10 years, mean 8.1 ± 0.8). The children were examined by five calibrated examiners (kappa = 0.9) after tooth brushing. The recording comprised teeth, surfaces, type and severity of MIH defects and was conducted using a portable light, mirrors and cotton rolls. MIH was registered according to the EAPD criteria. Defects <1 mm were not recorded. Statistical analysis included descriptive statistics and Spearman’s correlation. Most affected teeth were first permanent molars (71.4 %) followed by the maxillary central incisors (15.6 %). The most common defects were demarcated opacities (82.2 %), while the remaining 17.8 % of the affected teeth exhibited severe enamel defects. The most frequently affected surface in molars was the occlusal surface (72.4 %); in incisors, it was the buccal surface (73.5 %). There were no atypical restorations in the affected incisors. Different types of MIH defects at various surfaces of the same tooth were common. The number of affected tooth surfaces was positively correlated with the severity of MIH at child (p< 0.001). The study demonstrates severe enamel defects involving in almost one-fifth of all MIH teeth. The knowledge of the intra-oral distribution and severity of MIH findings at the enamel surface level is important for assessing the treatment needs.

Invasive fungal diseases are a major cause of death in renal allograft recipients. We previously reported that adjunctive recombinant human interferon‐γ therapy has clinical utility for invasive fungal diseases after renal transplantation. We have now developed a rapid peripheral blood‐based quantitative real‐time PCR assay that enables accurate profiling of cytokine imbalances. Our preliminary studies in renal transplant patients with invasive fungal diseases suggest that they fail to mount an adequate interferon‐γ response to the fungal infection. In addition, they have reduced IL‐10 and increased TNF‐α when compared to stable renal transplant patients. These preliminary cytokine profiling‐based observations provide a possible explanation for the therapeutic benefit of adjunctive human interferon‐γ therapy in renal allograft recipients with invasive fungal diseases.

To determine if there are in vivo differences in -aminobutyric acid (GABA) in the motor cortex and subcortical white matter of patients with amyotrophic lateral sclerosis (ALS) compared with healthy controls using proton magnetic resonance spectroscopy (1H-MRS).

We propose a robust approach to discriminant kernel-based feature extraction for face recognition and verification. We show, for the first time, how to perform the eigen analysis of the within-class scatter matrix directly in the feature space. This eigen analysis provides the eigenspectrum of its range space and the corresponding eigenvectors as well as the eigenvectors spanning its null space. Based on our analysis, we propose a kernel discriminant analysis (KDA) which combines eigenspectrum regularization with a feature-level scheme (ER-KDA). Finally, we combine the proposed ER-KDA with a nonlinear robust kernel particularly suitable for face recognition/verification applications which require robustness against outliers caused by occlusions and illumination changes. We applied the proposed framework to several popular databases (Yale, AR, XM2VTS) and achieved state-of-the-art performance for most of our experiments.

This paper presents an algorithm to detect the eyes and mouth of the faces under various imaging conditions such as with different poses, dimensions, illuminations, resolutions, wearing glasses or not and having different expressions is proposed. Initially, the algorithm converted the face region detected by the Viola-Jones algorithm into gray-scale and reduced the resolution by decimation to achieve low resolution region. The pixels that are darker than their surroundings are later located by 8 individual filters. Finally, the center of eyes and mouth are searched using a triangle model. Experiment show that, the algorithm produced 98.7% accuracy in detecting the eyes and mouth.

The incidence of invasive fungal infections (IFIs) in nonneutropenic solid organ transplant patients is increasing. We report our clinical experience with the use of interferon-? (IFN-?) immunotherapy in seven renal transplant patients who developed life threatening, disseminated IFIs refractory to conventional antifungal drug therapy. The infections were all microbiologically and histologically proven. The rapid cure of these disseminated infections with exogenous IFN-? injections was not associated with impaired kidney allograft function despite the use of liposomal amphotericin B in all cases. No clinical toxicity from the IFN-? immunotherapy was seen and no IFI relapsed during long-term follow-up. Our experience is both uncontrolled and in patients with unpredictable fungal infection-related outcomes. However, compared to standard approaches, the accelerated cure of life threatening, disseminated IFIs with 6 weeks of combination antifungal drug therapy and IFN-? immunotherapy saved lives, retained allograft function and led to substantial cost savings in this small patient group.

Limestones in Cyprus are mainly quarried for the production of coarse and fine aggregates to be used in concrete. The objective of this paper is to examine the properties of crushed limestone aggregates. The petrographical and physico-mechanical properties of these aggregates are described and their suitability for concrete production is examined. The coarse crushed limestone aggregates from Cyprus have water absorption values exceeding 3.3%, which is considered high for concrete applications. Their abrasion resistance (Los Angeles) values are consistently above 23%, while their weathering coefficients generally range between 10 and 30%. The fine crushed limestone aggregates show significantly lower water absorption values (less than 2.2%) and higher weathering coefficients (above 35%) than the coarse aggregates. The weathering coefficient of crushed limestone aggregates increases with a decrease in the fraction size up to 5 mm, after which it remains fairly constant. The physico-mechanical properties of crushed limestone aggregates are distinctly variable irrespective of the fact that they belong to the same geological formation and show relatively similar petrography.

Heart failure is the usual cause of death in patients with amyloid cardiomyopathy. The commonest form of hereditary cardiac amyloidosis is associated with the Val122Ile variant of transthyretin (TTR), which is carried by 3-4 of the African American population. Here, we report the outcome of the first cardiac transplantation in a patient with TTR V122I. A 59-year-old Caribbean man presented with biventricular failure. Other than previous bilateral carpel tunnel syndrome, he had been well and had no evidence of extracardiac amyloidosis. An endomyocardial biopsy demonstrated amyloid of TTR type. Sequencing of TTR gene indicated homozygosity for V122I. He underwent cardiac transplantation and 3 years later, remains well with no evidence of allograft or systemic amyloid deposition.

We report on measuring RH changes in concrete using FBGs inscribed in multimode gradient-index polymer CYTOP fibre. The CYTOP Bragg gratings were inscribed using the plane-by-plane femtosecond laser inscription method in order to minimize the coupling between the propagation modes supported by the multimode fibre. The wavelength shift of the FBG was monitored every two-hours for seven-days, and the humidity desorption of the concrete mixture was noted. [ABSTRACT FROM AUTHOR]

Nonaccidental head injury can be a very serious result of child abuse with the clinical outcome being poor and including mental retardation, long-term disability and even death. Skull radiographs, head CT and increasingly MRI are very useful tools in making this very serious diagnosis. Radiologists must be familiar with the radiologic patterns of child abuse and must maintain a high level of suspicion. We review and present several of the common radiologic findings in pediatric nonaccidental head injury.

Abstract For many remote sensing applications it is beneficial to know how the amount of shadows on a surface depends on illumination. Many natural surfaces (planetary surfaces being an example) may be successfully described by a fractal model. While the fractal shadowing function can be easily calculated experimentally, to date no rigorous analytical model of self-shadowing on a fractal surface exists. In this paper we offer an integral form of the shadowing function for fractal surfaces with different fractal and roughness parameters. The analysis is based on working out the covariance matrix for an arbitrarily long sequence of heights in a fractal profile.

In this paper we introduce novel face representations, the masked Trace transform (MTT), the shape Trace transform (STT) and the weighted Trace transform (WTT), for recognising faces in a face authentication system. We first transform the image space to the Trace transform space to produce the MTT. We then identify the points of the MTT which take similar values irrespective of intraclass variations and this way we create the WTT. Next we threshold the MTT and extract the edges of the thresholded regions to produce some shapes that characterise the person. This is the STT. Therefore, each person in the database is represented by their WTT and STT. We estimate the dissimilarity between two shapes by a new measure we propose, the Hausdorff context. Reinforcement learning is used to search for the optimal parameter values of the algorithm. Shape and features from the MTT are then integrated at the decision level, by a classifier combination algorithm. Our system is evaluated with experiments on the XM2VTS database using 2360 face images. We achieve a Total Error Rate (TER) of 0.18%, which is the lowest error among all other reported methods which used the same data and the same evaluation protocol in a recently published study.

Abstract: In this paper we present the Dempster-Shafer theory as a framework within which the results of a Bayesian network classifier and a fuzzy logic-based classifier are combined to produce a better final classification. We deal with the case when the two original classifiers use different classes for the outcome. The problem of different classes is solved by using a superset of finer classes which can be combined to produce classes according to either of the two classifiers. Within the Dempster-Shafer formalism not only can the problem of different number of classes be solved, but the relative reliability of the classifiers can also be considered.

We have reviewed the accuracy of prenatal diagnosis for the thalassaemias and sickle cell disorders performed for UK residents since the service began in 1974. Prenatal diagnosis has been performed in 3254 pregnancies: 517 by fetal blood analysis, 681 by Southern blotting and 2056 by polymerase chain reaction (PCR) methods, the majority using the amplification refractory mutation system (ARMS). The number of homozygotes diagnosed was 808 (24.8%). Twenty-five diagnostic errors have been recorded, ten arising from non-laboratory errors (0.31%) and 15 due to technical problems associated with the diagnostic techniques. The latter group consisted of eight misdiagnoses by globin chain synthesis (1.55%), five by Southern blot analysis (0.73%) and two by PCR methods (0.10%). The data show that the accuracy of prenatal diagnosis has improved with each development of diagnostic technique, and confirms that prenatal diagnosis of β-thalassaemia and sickle cell disorders by ARMS-PCR is very accurate and reliable. The overall error rate for prenatal diagnosis by PCR methods in the UK is now 0.41%. Copyright © 2000 John Wiley & Sons, Ltd.

The susceptibility of 12 isolates of Cryptococcus neoformans to amphotericin B, 5-fluorocytosine, fluconazole, itraconazole and ketoconazole was tested using the NCCLS and Etest methods with yeast nitrogen base (YNB) pH 5.6 and pH 7.0, RPMI MOPS pH 7.0 with and without added glucose (2%) and RPMI buffered with phosphate buffer to pH 7.0. Some isolates yielded poor growth in RPMI MOPS after 72 h. Tests indicated that YNB pH 5.6 was the best medium for 5-fluorocytosine but was unsuitable for ketoconazole. In conclusion, YNB pH 7.0 or RPMI MOPS with 2% glucose can be used with either method.

Abstract.: We present a method of colour shade grading for industrial inspection of random textures, the differences of which are at the threshold of human perception. This method uses image restoration techniques to recover an unblurred version of the image, and then blurs it the same way as the human visual system does, to emulate the process of the image being captured by the human sensor. Subsequently, the colour image is transformed into a perceptually uniform colour space, where colour gra ding takes place.

Over the last 15 years we have managed patients with malignant pleural effusion, using a single procedure with thoracoscopy and talc pleurodesis or shunt as appropriate. Talc pleurodesis remains our primary treatment option but in those patient shown to have the 'trapped lung syndrome', in whom pleurodesis would fail, we have been using a pleuroperitoneal shunt.

Abstract.: In this paper, we formulate the two-stage stock-cutting problem, according to which a set of rectangular pieces of prespecified dimensions are to be cut from an arbitrarily shaped object with arbitrarily shaped holes or defective regions. We show how mathematical morphological operators can be used in order to determine the optimal shifting for a given cutting pattern. It is then proved that the problem of obtaining the optimal cutting pattern is -hard and a solution to the unconstrained problem using mathematical programming is proposed. However, for the general problem, good sub-optimal solutions can be obtained using the technique of simulated annealing. Experimental results are also included.

AbstractThis paper studies the constrained two-dimensional bin-packing or rectangle packing (RP) problem, according to which a set of arbitrarily dimensioned rectangular boxes are to be packed into a single orthogonal bin without overlapping. An efficient representation that can handle various constraints is proposed. The solution to this combinatorial optimisation problem is obtained by simulated annealing by the use of suitable search operators that constrain the search in the valid subspace of the problem. Experimental results demonstrate the effectiveness of the proposed algorithm.

In this paper we introduce the two-point correlation function as a measure of interclass separability. We present a theoretical study of this statistic in a general M-dimensional feature space and propose a fast algorithm for the efficient computation of it. We test the algorithm and illustrate the properties of the statistic using test data in 1D and 2D feature spaces and discuss the boundary effects of the feature space. We also present a discussion of the limitations of the proposed statistic and apply it to the assessment of inter-class separability in a texture segmentation context.

We generalize here the use of the 1D Boolean model for the analysis of grey level textures. Each grey image is first split into eight binary images using different criteria. Each of these binary images is separately analysed with the help of the 1D Boolean model and features are extracted from it. The final grey texture recognition is performed on the basis of these features using several classification criteria. Experiments have been carried out using an image database of 30 grey level textures, all of them with 512×512 pixels in size, obtaining correct classification rates between 95% and 100%, according to the classification criterion used.

A programme of prospective heterozygote detection and counselling, fetal diagnostic testing, and abortion offetuses affected by thalassaemia major introduced in Britain in 1977 has proved highly acceptable to at-risk couples of Cypriot and East African Asian origin, but less so to couples of Pakistani origin. However, many at-risk couples are still not detected prospectively, the proportion of thalassaemia-major births prevented was only 32% by the end of 1981, and there is little evidence of a further fall since then. The thalassaemia-major birth-rate had fallen by 60% in Cypriots and by 20% in East African Asians, but it had not fallen at all in Pakistanis. Improved approaches to fetal diagnosis of thalassaemia major are becoming available, so a concerted effort is needed to inform all the at-risk ethnic groups of the existence of the problem and the possibility of detection of affected fetuses.

Prenatal diagnoses were performed on six selected pairs of parents known to be carriers of Hb mutations by testing transcervical cells (TCCs) retrieved, prior to chorionic villus sampling (CVS), by aspiration of the cervical mucus from the pregnant mothers at 10–12 weeks of gestation. A concordance between the results of testing chorionic villus cells and isolated clumps of trophoblastic cellular elements was observed in four of the six cases. © 1997 John Wiley & Sons, Ltd.

In this paper we propose the use of the Randomized Hough Transform algorithm for the determination of the proportions of pure classes present in sets of mixed pixels, for large datasets (for which the deterministic Hough is prohibitively slow) and in the presence of outliers (i.e. in cases that the classical Least Square Error method cannot cope). We show that the Randomized Hough is of constant CPU time, irrespective of the size of the data sets and can be made even more accurate than the deterministic Hough for datasets larger than about 50 sample points. We demonstrate our results both with simulated and laboratory data.

Changes in contraction and relaxation parameters during chronic electrical stimulation can exert profound effects on diastolic augmentation during skeletal muscle assistance (SMA) of the circulation, in both short and long term. The physiological properties of latissimus dorsi muscle (LD) performance in a system that mimics the clinical setting has not been adequately studied. Objective: To quantify changes in the biomechanical performance of trained and untrained skeletal muscle in relation to circulatory assistance using an ex-vivo Windkessel mock circulation. Methods: Twelve Welsh Mountain sheep were divided into 2 groups: Group A (n = 6) underwent implantation of intramuscular electrodes into the left LD connected to a myostimulator (Telectronics Pacing Systems, Inc., Colorado) and progressively trained by burst stimulation over a 12-week period using standard stimulation parameters (2.5-5 V, 35 Hz, 6 pulses per burst, 240 μs per pulse); Group B (n = 6) were the untrained controls. At the end of 12 weeks, the LD was mobilised on its neurovascular pedicle and wrapped around a latex rubber aorta connected to two Windkessel chambers pressurised to 70 mmHg and stimulated to contract 40 times per minute continuously for 60 min. Pressure change per contraction (augmentation, ΔP), volume displacement, contraction (Ct) and relaxation to 90% (Rt90) times, and the standardised rate of change of pressure generation (+dP/dt:ΔP) and decay (−dP/dt:ΔP) were determined and assessed for potential clinical efficacy. Results: In Group A, the LD was fatigue-resistant in all 6 animals with a mean pressure augmentation of 13.7 (s.e.m. 1.3) mmHg and mean stroke volume of 12.5 (s.e.m. 1.0) ml. These muscles were slow with a mean Ct and +dP/dt:Δmax of 243.2 (s.e.m. 6.1) ms and +6.5 s−1, respectively, and Rt90 and −dP/dt:Δmax of 261.0 (s.e.m. 4.8) ms and −7.8 s−1, respectively. In contrast, the LD in Group B was fatiguable with a mean pressure augmentation and stroke volume of 24.6 (s.e.m. 0.9) mmHg and 21.1 (s.e.m. 0.7) ml at 1 min and only 5.4 (s.e.m. 0.3) mmHg and 5.2 (s.e.m. 0.3) ml, respectively, at 30 min (P < 0.001). These muscles were faster at all time points compared to group A (P < 0.02). Acute diminution of power output per contraction in Group B coincided with a prolongation in the Rt90 by 101% compared to the Ct which decreased by less than 5% (P < 0.001). The Ct/Rt90 ratio did not significantly change during performance testing in Group A (fatigue-resistant animals) (P > 0.1). Conclusion: Using a mock circulation system, we have identified significant differences in biomechanical properties of trained and untrained skeletal muscle. Optimisation of these parameters during and after electrical training may alter the clinical efficacy of SMA.

In this paper we present an optimization algorithm for locating peaks in the accumulator of the Hough algorithm with robust voting kernel. We present a detailed discussion of the accuracy that can be achieved by locating these peaks in the accumulator, and show that the error bounds on the estimates of line parameters are always within those based upon least squares. This arises from the robust nature of the voting kernel. We describe the optimization algorithm in some detail since the shape of the peaks in the standard parameter space for straight lines are sinusoidal ridges. Standard approaches therefore fail, but the method described is shown to be robust from the experimental results presented. Some discussion of post-processing is also made, in which the shortcomings of standard Hough techniques, splitting long lines across parameter bins, can be remedied. We also discuss the use of a confidence measure in the line parameters based upon the value of the accumulator, and show that this is related to the mean squared distance from the line of the edge pixels associated with it. Finally, we present results produced by this optimizing Hough technique on a disparate set of images, with various application areas in mind, to demonstrate the versatility of the method and the accuracy that can be achieved at little computational overhead.

In this paper we show how a Bayesian network of inference can be used with a GIS to combine information from different sources of data for classification. Data may include satellite sensor images, topographic maps, geological maps etc, each one with its own resolution and accuracy. We show how this uncertainty in the input data can be incorporated in the network and present various methods by which the conditional probability matrices used by the network can be constructed. We demonstrate our approach within the framework of the problem of assessing the risk of desertification of some burned forests in the Mediterranean region.

In this paper the problem of image feature extraction is considered with emphasis on developing methods which are resilient in the presence of data contamination. The issue of robustness of estimation procedures has received considerable attention in the statistics community [1–3] but its results are only recently being applied to specific image analysis tasks [4–7]. In this paper we show how the design of robust methods applies to image description tasks posed within a statistical hypothesis testing and parameter estimation framework. The methodology is illustrated by applying it to finding robust, optimal estimation kernels for line detection and edge detection. We then discuss the relationship of these optimal solutions to both the well established Hough Transform technique and the standard estimation kernels developed in the statistics literature. The application of standard robust kernels to image analysis tasks is illustrated by two examples which involve circular arc detection in gray-level imagery and planar surface segmentation in depth data. Robust methods are found to be effective general tools for generating 2D and 3D image descriptions.

The increasing volume of remotely-sensed data has led to a need for efficient automatic analysis procedures. A system is presented for automatic, knowledge-based segmentation of remotely-sensed images of the land. The system uses the information in time sequences of remotely-sensed data together with cartographic map data and domain expertise to build a model of the scene in terms of segments and their possible classes. The accuracies of segmentation and subsequent classification are shown to be superior to traditional automatic techniques. In addition, potential changes in the scene are isolated.

In this paper we consider a Hough transform line finding algorithm in which the voting kernel is a smooth function of differences in both line parameters. The shape of the voting kernel is decided in terms of a hypothesis testing approach, and the shape is adjusted to give optimal results. We show that this new kernel is robust to changes in the distribution of the underlying noise and the implementation is very fast, taking typically 2-3 s on a Sparc 2 workstation for a 256 × 256 image.Copyright 1993, 1999 Academic Press

In this paper we consider a Hough transform line finding algorithm in which the voting kernel is a smooth function of differences in both line parameters. The shape of the voting kernel is decided in terms of a hypothesis testing approach, and the shape is adjusted to give optimal results. We show that this new kernel is robust to changes in the distribution of the underlying noise and the implementation is very fast, taking typically 2-3 s on a Sparc 2 workstation for a 256 × 256 image. Copyright 1993, 1999 Academic Press

Abstract.: We present a method of colour shade grading for industrial inspection of surfaces, the differences of which are at the threshold of human perception. This method converts the input data from the electronic sensor to the corresponding data as they would have been viewed using the human vision system. Then their differences are computed using a perceptually uniform colour space, thus approximating the way the human experts would grade the product. The transformation from the electronic sensor to the human sensor makes use of synthetic metameric data to determine the transformation parameters. The method has been tested using real data.

Remotely-sensed data constitute a major potential source of input to geographical information systems (GIS)However, these data often have a relatively poor classification accuracy compared with that of the cartographic data from maps with which they may be combined in the course of GIS analysis. The possibility exists of using data sets (in the form of digital maps) resident within a GIS in order to improve this accuracy, before the classified image is incorporated into the GIS. Results are discussed from a British Alvey Information Technology project to develop a system for the knowledge-based segmentation and classification of remotely-sensed terrain images, in which the knowledge contained in digital map