High-breakdown linear discriminant analysis

The classification rules of linear discriminant analysis are defined by the true mean vectors and the common covariance matrix of the populations from which the data come. Because these true parameters are generally unknown, they are commonly estimated by the sample mean vector and covariance matrix of the data in a training sample randomly drawn from each population. However, these sample statistics are notoriously susceptible to contamination by outliers, a problem compounded by the fact that the outliers may be invisible to conventional diagnostics. High-breakdown estimation is a procedure designed to remove this cause for concern by producing estimates that are immune to serious distortion by a minority of outliers, regardless of their severity. In this article we motivate and develop a high-breakdown criterion for linear discriminant analysis and give an algorithm for its implementation. The procedure is intended to supplement rather than replace the usual sample-moment methodology of discriminant analysis either by providing indications that the dataset is not seriously affected by outliers (supporting the usual analysis) or by identifying apparently aberrant points and giving resistant estimators that are not affected by them.

Applying linear time-varying constraints to econometric models: With an application to demand systems

When linear equality constraints are invariant through time they can be incorporated into estimation by restricted least squares. If, however, the constraints are time-varying, this standard methodology cannot be applied. In this paper we show how to incorporate linear time-varying constraints into the estimation of econometric models. The method involves the augmentation of the observation equation of a state-space model prior to estimation by the Kalman filter. Numerical optimisation routines are used for the estimation. A simple example drawn from demand analysis is used to illustrate the method and its application.

Linear IgA dermatosis induced by pregnancy

Linear IgA disease is a rare autoimmune subepidermal bullous disorder characterized by linear IgA deposits at the epidermal basement membrane zone. According to the literature, in patients who have linear IgA disease and become pregnant, the disease tends to improve. We report a case of linear IgA disease induced by pregnancy, successfully treated with dapsone and prednisone with no adverse effects observed in the patient and her newborns.

Leptospirosis pulmonary haemorrhage syndrome is associated with linear deposition of immunoglobulin and complement on the alveolar surface

Leptospirosis is a zoonotic infection associated with severe diseases such as leptospirosis pulmonary haemorrhage syndrome (LPHS). The cause of pulmonary haemorrhage is unclear. Understanding which mechanisms and processes are involved in LPHS will be important in treatment regimens under development for this life-threatening syndrome. In the present study, we evaluated 30 lung specimens from LPHS patients and seven controls using histology and immunohistochemistry (detection of IgM, IgG, IgA and C3) in order to describe the pathological features associated with this syndrome. Immunoglobulin deposits were detected on the alveolar surface in 18/30 LPHS patients. Three staining patterns were observed for the immunoglobulins and C3 in the lung tissues of LPHS patients: AS, delicate linear staining adjacent to the alveolar surface, which was indicative of a membrane covering the luminal surface of type I and II pneumocyte cells; S, heterogeneous staining which was sporadically distributed along the alveolar septum; and IA, weak, focal intra-alveolar granular staining. Human LPHS is associated with individual and unique histological patterns that differ from those of other causes of pulmonary haemorrhage. In the present study, it was found that the linear deposition of immunoglobulins (IgA, IgG and IgM) and complement on the alveolar surface may play a role in the pathogenesis of pulmonary haemorrhage in human leptospirosis.

Obesity due to Melanocortin 4 Receptor (MC4R) Deficiency Is Associated with Increased Linear Growth and Final Height, Fasting Hyperinsulinemia, and Incompletely Suppressed Growth Hormone Secretion

Context: Melanocortin receptor 4 (MC4R) deficiency is characterized by increased linear growth greater than expected for the degree of obesity. Objective: The objective of the investigation was to study the somatotroph axis in obese MC4R-deficient patients and equally obese controls. Patients and Methods: We obtained anthropometric measurements and insulin concentrations in 153 MC4R-deficient subjects and 1392 controls matched for age and severity of obesity. We measured fasting IGF-I, IGF-II, IGF binding protein (IGFBP)-1, IGFBP-3, and acid-labile subunit levels in a subset of 33 MC4R-deficient patients and 36 control subjects. We examined pulsatile GH secretion in six adult MC4R-deficient subjects and six obese controls. Results: Height so score was significantly greater in MC4R-deficient children under 5 yr of age compared with controls (mean +/- SEM: 2.3 +/- 0.06 vs. 1.8 +/- 0.04, P < 0.001), an effect that persisted throughout childhood. Final height (cm) was greater in MC4R-deficient men (mean +/- SEM 173 +/- 2.5 vs. 168 +/- 2.1, P < 0.001) and women (mean 165 +/- 2.1 vs. 158 +/- 1.9, P < 0.001). Fasting IGF-I, IGF-II, acid-labile subunit, and IGFBP-3 concentrations were similar in the two groups. GH levels were markedly suppressed in obese controls, but pulsatile GH secretion was retained in MC4R deficiency. The mean maximal GH secretion rate per burst (P < 0.05) and mass per burst (P < 0.05) were increased in MC4R deficiency, consistent with increased pulsatile and total GH secretion. Fasting insulin levels were markedly elevated in MC4R-deficient children. Conclusions: In MC4R deficiency, increased linear growth in childhood leads to increased adult final height, greater than predicted by obesity alone. GH pulsatility is maintained in MC4R deficiency, a finding consistent with animal studies, suggesting a role for MC4R in controlling hypothalamic somatostatinergic tone. Fasting insulin levels are significantly higher in children carrying MC4R mutations. Both of these factors may contribute to the accelerated growth phenotype characteristic of MC4R deficiency. (J Clin Endocrinol Metab 96: E181-E188, 2011)

Focal and linear endocardial and epicardial catheter-based cryoablation of normal and infarcted ventricular tissue

Background: This study of a chronic porcine postinfarction model examined whether linear epicardial cryoablation was capable of creating large, homogenous lesions in regions of the myocardium including scarred ventricle. Endocardial and epicardial focal cryolesions were also compared to determine if there were significant differences in lesion characteristics. Methods: Eighty focal endocardial and 28 focal epicardial cryoapplications were delivered to eight normal caprine and four normal porcine ventricular myocardium, and 21 linear cryolesions were applied along the border of infarcted epicardial tissue in a chronic porcine infarct model in six swines. Results: Focal endocardial cryolesions in normal animals measured 9.7 +/- 0.4 mm (length) by 7.3 +/- 1.4 mm (width) by 4.8 +/- 0.2 mm (depth), while epicardial lesions measured 10.2 +/- 1.4 mm (length) by 7.7 +/- 2 mm (width) by 4.6 +/- 0.9 mm (depth); P > 0.05. Linear epicardial cryolesions in the chronic porcine infarct model measured 36.5 +/- 7.8 mm (length) by 8.2 +/- 1.3 mm (width) by 6.0 +/- 1.2 mm (depth). The mean depth of linear cryolesions applied to the border of the infarct scar was 7 +/- 0.7 mm, as measured by magnetic resonance imaging. Conclusions:Cryoablation can create deep lesions when delivered to the ventricular epicardium. Endocardial and epicardial cryolesions created by a focal cryoablation catheter are similar in size and depth. The ability to rapidly create deep linear cryolesions may prove to be beneficial in substrate-based catheter ablation of ventricular arrhythmias.

Assessment of linear measurements of bone for implant sites in the presence of metallic artefacts using cone beam computed tomography and multislice computed tomography

The objective was to evaluate the influence of dental metallic artefacts on implant sites using multislice and cone-beam computed tomography techniques. Ten dried human mandibles were scanned twice by each technique, with and without dental metallic artefacts. Metallic restorations were placed at the top of the alveolar ridge adjacent to the mental foramen region for the second scanning. Linear measurements (thickness and height) for each cross-section were performed by a single examiner using computer software. All mandibles were analysed at both the right and the left mental foramen regions. For the multislice technique, dental metallic artefact produced an increase of 5% in bone thickness and a reduction of 6% in bone height; no significant differences (p > 0.05) were detected when comparing measurements performed with and without metallic artefacts. With respect to the cone-beam technique, dental metallic artefact produced an increase of 6% in bone thickness and a reduction of 0.68% in bone height. No significant differences (p > 0.05) were observed when comparing measurements performed with and without metallic artefacts. The presence of dental metallic artefacts did not alter the linear measurements obtained with both techniques, although its presence made the location of the alveolar bone crest more difficult.

Assessment of linear and angular measurements on three-dimensional cone-beam computed tomographic images

Objective. The purpose of this research was to provide further evidence to demonstrate the precision and accuracy of maxillofacial linear and angular measurements obtained by cone-beam computed tomography (CBCT) images. Study design. The study population consisted of 15 dry human skulls that were submitted to CBCT, and 3-dimensional (3D) images were generated. Linear and angular measurements based on conventional craniometric anatomical landmarks, and were identified in 3D-CBCT images by 2 radiologists twice each independently. Subsequently, physical measurements were made by a third examiner using a digital caliper and a digital goniometer. Results. The results demonstrated no statistically significant difference between inter-and intra-examiner analysis. Regarding accuracy test, no statistically significant differences were found of the comparison between the physical and CBCT-based linear and angular measurements for both examiners (P = .968 and .915, P = .844 and .700, respectively). Conclusions. 3D-CBCT images can be used to obtain dimensionally accurate linear and angular measurements from bony maxillofacial structures and landmarks. (Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2009; 108: 430-436)

Anisotropy-based performance analysis of linear discrete time invariant control systems

The anisotropic norm of a linear discrete-time-invariant system measures system output sensitivity to stationary Gaussian input disturbances of bounded mean anisotropy. Mean anisotropy characterizes the degree of predictability (or colouredness) and spatial non-roundness of the noise. The anisotropic norm falls between the H-2 and H-infinity norms and accommodates their loss of performance when the probability structure of input disturbances is not exactly known. This paper develops a method for numerical computation of the anisotropic norm which involves linked Riccati and Lyapunov equations and an associated special type equation.

A flux ratio theorem for multicomponent linear diffusion equations

Ussing [1] considered the steady flux of a single chemical component diffusing through a membrane under the influence of chemical potentials and derived from his linear model, an expression for the ratio of this flux and that of the complementary experiment in which the boundary conditions were interchanged. Here, an extension of Ussing's flux ratio theorem is obtained for n chemically interacting components governed by a linear system of diffusion-migration equations that may also incorporate linear temporary trapping reactions. The determinants of the output flux matrices for complementary experiments are shown to satisfy an Ussing flux ratio formula for steady state conditions of the same form as for the well-known one-component case. (C) 2000 Elsevier Science Ltd. All rights reserved.

Stability of linear difference and differential inclusions

Any given n X n matrix A is shown to be a restriction, to the A-invariant subspace, of a nonnegative N x N matrix B of spectral radius p(B) arbitrarily close to p(A). A difference inclusion x(k+1) is an element of Ax(k), where A is a compact set of matrices, is asymptotically stable if and only if A can be extended to a set B of nonnegative matrices B with \ \B \ \ (1) < 1 or \ \B \ \ (infinity) < 1. Similar results are derived for differential inclusions.