Brian J. Thelen

Limited short-term prognostic utility of cerebral NIRS during neonatal therapeutic hypothermia

Objective: We evaluated the utility of amplitude-integrated EEG (aEEG) and regional oxygen saturation (rSO2) measured using near-infrared spectroscopy (NIRS) for short-term outcome prediction in neonates with hypoxic ischemic encephalopathy (HIE) treated with therapeutic hypothermia. Methods: Neonates with HIE were monitored with dual-channel aEEG, bilateral cerebral NIRS, and systemic NIRS throughout cooling and rewarming. The short-term outcome measure was a composite of neurologic examination and brain MRI scores at 7 to 10 days. Multiple regression models were developed to assess NIRS and aEEG recorded during the 6 hours before rewarming and the 6-hour rewarming period as predictors of short-term outcome. Results: Twenty-one infants, mean gestational age 38.8 ± 1.6 weeks, median 10-minute Apgar score 4 (range 0–8), and mean initial pH 6.92 ± 0.19, were enrolled. Before rewarming, the most parsimonious model included 4 parameters (adjusted R2 = 0.59; p = 0.006): lower values of systemic rSO2 variability (p = 0.004), aEEG bandwidth variability (p = 0.019), and mean aEEG upper margin (p = 0.006), combined with higher mean aEEG bandwidth (worse discontinuity; p = 0.013), predicted worse short-term outcome. During rewarming, lower systemic rSO2 variability (p = 0.007) and depressed aEEG lower margin (p = 0.034) were associated with worse outcome (model-adjusted R2 = 0.49; p = 0.005). Cerebral NIRS data did not contribute to either model. Conclusions: During day 3 of cooling and during rewarming, loss of physiologic variability (by systemic NIRS) and invariant, discontinuous aEEG patterns predict poor short-term outcome in neonates with HIE. These parameters, but not cerebral NIRS, may be useful to identify infants suitable for studies of adjuvant neuroprotective therapies or modification of the duration of cooling and/or rewarming.

Modeling acute respiratory illness during the 2007 San Diego wildland fires using a coupled emissions-transport system and generalized additive modeling

BackgroundA study of the impacts on respiratory health of the 2007 wildland fires in and around San Diego County, California is presented. This study helps to address the impact of fire emissions on human health by modeling the exposure potential of proximate populations to atmospheric particulate matter (PM) from vegetation fires. Currently, there is no standard methodology to model and forecast the potential respiratory health effects of PM plumes from wildland fires, and in part this is due to a lack of methodology for rigorously relating the two. The contribution in this research specifically targets that absence by modeling explicitly the emission, transmission, and distribution of PM following a wildland fire in both space and time.MethodsCoupled empirical and deterministic models describing particulate matter (PM) emissions and atmospheric dispersion were linked to spatially explicit syndromic surveillance health data records collected through the San Diego Aberration Detection and Incident Characterization (SDADIC) system using a Generalized Additive Modeling (GAM) statistical approach. Two levels of geographic aggregation were modeled, a county-wide regional level and division of the county into six sub regions. Selected health syndromes within SDADIC from 16 emergency departments within San Diego County relevant for respiratory health were identified for inclusion in the model.ResultsThe model captured the variability in emergency department visits due to several factors by including nine ancillary variables in addition to wildfire PM concentration. The model coefficients and nonlinear function plots indicate that at peak fire PM concentrations the odds of a person seeking emergency care is increased by approximately 50% compared to non-fire conditions (40% for the regional case, 70% for a geographically specific case). The sub-regional analyses show that demographic variables also influence respiratory health outcomes from smoke.ConclusionsThe model developed in this study allows a quantitative assessment and prediction of respiratory health outcomes as it relates to the location and timing of wildland fire emissions relevant for application to future wildfire scenarios. An important aspect of the resulting model is its generality thus allowing its ready use for geospatial assessments of respiratory health impacts under possible future wildfire conditions in the San Diego region. The coupled statistical and process-based modeling demonstrates an end-to-end methodology for generating reasonable estimates of wildland fire PM concentrations and health effects at resolutions compatible with syndromic surveillance data.

Phase-diverse adaptive optics for future telescopes

Phase Diversity (PD) is a wavefront-sensing technology that offers certain advantages in an Adaptive-Optics (AO) system. Historically, PD has not been considered for use in AO applications because computations have been prohibitive. However, algorithmic and computational-hardware advances have recently allowed use of PD in AO applications. PD is an attractive candidate for AO applications for a variety of reasons. The optical hardware required is simple to implement and eliminates non-common path errors. In addition, PD has also been shown to work well with extended scenes that are encountered, for example, when imaging low-contrast solar granulation. PD can estimate high-order continuous aberrations as well as wavefront discontinuities characteristic of segmented-aperture or sparse-aperture telescope designs. Furthermore, the fundamental information content in a PD data set is shown to be greater than that of the correlation Shack-Hartmann wavefront sensor for the limiting case of unresolved objects. These advantages coupled with recent laboratory results (extended-scene closed-loop AO with PD sampling at 100 Hz) highlight the maturation of not only the PD concept and algorithm but the technology as an emerging and viable wavefront sensor for use in AO applications.

Development of Methods for Detection and Monitoring of Fire Disturbance in the Alaskan Tundra Using a Two-Decade Long Record of Synthetic Aperture Radar Satellite Images

Using the extensive archive of historical ERS-1 and -2 synthetic aperture radar (SAR) images, this analysis demonstrates that fire disturbance can be effectively detected and monitored in high northern latitudes using radar technology. A total of 392 SAR images from May to August spanning 1992-2010 were analyzed from three study fires in the Alaskan tundra. The investigated fires included the 2007 Anaktuvuk River Fire and the 1993 DCKN178 Fire on the North Slope of Alaska and the 1999 Uvgoon Creek Fire in the Noatak National Preserve. A 3 dB difference was found between burned and unburned tundra, with the best time for burned area detection being as late in the growing season as possible before frozen ground conditions develop. This corresponds to mid-August for the study fires. In contrast to electro-optical studies from the same region, measures of landscape recovery as detected by the SAR were on the order of four to five years instead

Compressive sensing and 3-D radar imaging

Nonstandard image formation methods can enable fully three-dimensional fine resolution radar images of some objects of interest to be constructed from certain types of sparsely sampled three-dimensional apertures, which contain too little collected data to support traditional imaging methods. Such three-dimensional image products provide more target information than traditional SAR and ISAR imagery, and eliminate most of the difficulties associated with interpretation, mensuration, and recognition that result from overlay effects, self shadowing, scaling and viewing angle uncertainties.

Operation of Phase-Diverse Adaptive-Optics with Extended Scenes

Previously, we demonstrated useful and novel features of the General Dynamics QuickStar adaptive-optics testbed utilizing Phase Diversity (PD) as the wavefront sensor operating on a point object. Point objects are relatively easy to produce in the laboratory and simplify the calibration procedure. However, for some applications, natural or artificial beacons may not be readily available and a wavefront sensor that operates on extended scenes is required. Accordingly, the QuickStar testbed has been augmented to allow PD to operate on natural three-dimensional solar-illuminated scenes external to the QuickStar laboratory. In addition, a computationally efficient chip-selection strategy has been developed that allows PD to operate on chips with favorable scene content. Finally, a covariance matrix has been developed that provides an accuracy estimate for PD wavefront-parameter estimates. The covariance can be used by the controls algorithm to properly weight the correction applied according to the accuracy of the estimates. These advances suggest that PD is a sufficiently mature technology for use in adaptive optics systems that require operation with extended scenes.

Theoretical performance assessment and empirical analysis of super-resolution under unknown affine sensor motion.

This paper deals with super-resolution (SR) processing and associated theoretical performance assessment for under-sampled video data collected from a moving imaging platform with unknown motion and assuming a relatively flat scene. This general scenario requires joint estimation of the high-resolution image and the parameters that determine a projective transform that relates the collected frames to one another. A quantitative assessment of the variance in the random error as achieved through a joint-estimation approach (e.g., SR image reconstruction and motion estimation) is carried out via the general framework of M-estimators and asymptotic statistics. This approach provides a performance measure on estimating the fine-resolution scene when there is a lack of perspective information and represents a significant advancement over previous work that considered only the more specific scenario of mis-registration. A succinct overview of the theoretical framework is presented along with some specific results on the approximate random error for the case of unknown translation and affine motions. A comparison is given between the approximated random error and that actually achieved by an M-estimator approach to the joint-estimation problem. These results provide insight on the reduction in SR reconstruction accuracy when jointly estimating unknown inter-frame affine motion.

Incorporating prior knowledge of urban scene spatial structure in aperture code designs for surveillance systems

Two major missions of Surveillance systems are imaging and ground moving target indication (GMTI). Recent advances in coded aperture electro optical systems have enabled persistent surveillance systems with extremely large fields of regard. The areas of interest for these surveillance systems are typically urban, with spatial topologies having a very definite structure. We incorporate aspects of a priori information on this structure in our aperture code designs to enable optimized dealiasing operations for undersampled focal plane arrays. Our framework enables us to design aperture codes to minimize mean square error for image reconstruction or to maximize signal to clutter ratio for GMTI detection. In this paper we present a technical overview of our code design methodology and show the results of our designed codes on simulated DIRSIG mega-scene data.

Decomposition approaches to separate clutter/background from buried object signatures

Ground penetrating radar measurements are dominated by the strong return from the ground interface and volume scattering from distributed subsurface in-homogeneities. Buried object detection performance can be improved if these clutter sources can be reduced relative to the scattering from the buried objects of interest. This paper applies two recently developed methods of separating a signal into a low-rank component (representing the background) and a sparse component (the buried object), robust principal component analysis (RPCA) and dynamic mode decomposition (DMD), to the problem of separating subsurface scattering anomalies from a slowly varying background. The algorithms are described and an example application to field-collected impulse GPR data is shown. The target-to-clutter ratio is significantly improved in the sparse component compare to that in the original data suggesting that these techniques are viable methods of suppressing surface clutter and distributed volumetric clutter.

Divergences and estimating tight bounds on Bayes error with applications to multivariate Gaussian copula and latent Gaussian copula

In Bayesian decision theory, there has been a great amount of research into theoretical frameworks and information– theoretic quantities that can be used to provide lower and upper bounds for the Bayes error. These include well-known bounds such as Chernoff, Battacharrya, and J-divergence. Part of the challenge of utilizing these various metrics in practice is (i) whether they are ”loose” or ”tight” bounds, (ii) how they might be estimated via either parametric or non-parametric methods, and (iii) how accurate the estimates are for limited amounts of data. In general what is desired is a methodology for generating relatively tight lower and upper bounds, and then an approach to estimate these bounds efficiently from data. In this paper, we explore the so-called triangle divergence which has been around for a while, but was recently made more prominent in some recent research on non-parametric estimation of information metrics. Part of this work is motivated by applications for quantifying fundamental information content in SAR/LIDAR data, and to help in this, we have developed a flexible multivariate modeling framework based on multivariate Gaussian copula models which can be combined with the triangle divergence framework to quantify this information, and provide approximate bounds on Bayes error. In this paper we present an overview of the bounds, including those based on triangle divergence and verify that under a number of multivariate models, the upper and lower bounds derived from triangle divergence are significantly tighter than the other common bounds, and often times, dramatically so. We also propose some simple but effective means for computing the triangle divergence using Monte Carlo methods, and then discuss estimation of the triangle divergence from empirical data based on Gaussian Copula models.