Search

We introduce a novel method to estimate the causal effects of an intervention over multiple treated units by combining the techniques of probabilistic forecasting with global forecasting methods using deep learning (DL) models. Considering the counterfactual and synthetic approach for policy evaluation, we recast the causal effect estimation problem as a counterfactual prediction outcome of the treated units in the absence of the treatment. Nevertheless, in contrast to estimating only the counterfactual time series outcome, our work differs from conventional methods by proposing to estimate the counterfactual time series probability distribution based on the past preintervention set of treated and untreated time series. We rely on time series properties and forecasting methods, with shared parameters, applied to stacked univariate time series for causal identification. This article presents DeepProbCP, a framework for producing accurate quantile probabilistic forecasts for the counterfactual outcome, based on training a global autoregressive recurrent neural network model with conditional quantile functions on a large set of related time series. The output of the proposed method is the counterfactual outcome as the spline-based representation of the counterfactual distribution. We demonstrate how this probabilistic methodology added to the global DL technique to forecast the counterfactual trend and distribution outcomes overcomes many challenges faced by the baseline approaches to the policy evaluation problem. Oftentimes, some target interventions affect only the tails or the variance of the treated units' distribution rather than the mean or median, which is usual for skewed or heavy-tailed distributions. Under this scenario, the classical causal effect models based on counterfactual predictions are not capable of accurately capturing or even seeing policy effects. By means of empirical evaluations of synthetic and real-world datasets, we show that our framework delivers more accurate forecasts than the state-of-the-art models, depicting, in which quantiles, the intervention most affected the treated units, unlike the conventional counterfactual inference methods based on nonprobabilistic approaches.

This report describes local administration of submicron particle paclitaxel (SPP) (NanoPac®: ~ 800-nm-sized particles with high relative surface area with each particle containing ~ 2 billion molecules of paclitaxel) in preclinical models and clinical trials evaluating treatment of carcinomas. Paclitaxel is active in the treatment of epithelial solid tumors including ovarian, peritoneal, pancreatic, breast, esophageal, prostate, and non-small cell lung cancer. SPP has been delivered directly to solid tumors, where the particles are retained and continuously release the drug, exposing primary tumors to high, therapeutic levels of paclitaxel for several weeks. As a result, tumor cell death shifts from primarily apoptosis to both apoptosis and necroptosis. Direct local tumoricidal effects of paclitaxel, as well as stimulation of innate and adaptive immune responses, contribute to antineoplastic effects. Local administration of SPP may facilitate tumor response to systemically administered chemotherapy, targeted therapy, or immunotherapy without contributing to systemic toxicity. Results of preclinical and clinical investigations described here suggest that local administration of SPP achieves clinical benefit with negligible toxicity and may complement standard treatments for metastatic disease. Graphical abstract

The foregrounding of economic interests to the exclusion of broadly termed ‘cultural’ interests in water governance has been challenged by the grant of legal personality to rivers, including the Ga...

Emergency Medical Services (EMS) demand load has become a considerable burden for many government authorities, and EMS demand is often an early indicator for stress in communities, a warning sign of emerging problems. In this paper, we introduce Deep Planning and Policy Making Net (DeepPPMNet), a Long Short-Term Memory network based, global forecasting and inference framework to forecast the EMS demand, analyse causal relationships, and perform `what-if' analyses for policy-making across multiple local government areas. Unless traditional univariate forecasting techniques, the proposed method follows the global forecasting methodology, where a model is trained across all the available EMS demand time series to exploit the potential cross-series information available. DeepPPMNet also uses seasonal decomposition techniques, incorporated in two different training paradigms into the framework, to suit various characteristics of the EMS related time series data. We then explore causal relationships using the notion of Granger Causality, where the global forecasting framework enables us to perform `what-if' analyses that could be used for the national policy-making process. We empirically evaluate our method, using a set of EMS datasets related to alcohol, drug use and self-harm in Australia. The proposed framework is able to outperform many state-of-the-art techniques and achieve competitive results in terms of forecasting accuracy. We finally illustrate its use for policy-making in an example regarding alcohol outlet licenses.

• Evaporites occur in an unsaturated silt layer, which is underlain by a sand and gravel aquifer. • These evaporites are rich in chloride across the site. • Uranium concentrations are higher in the evaporites that overlie the uranium contaminant plume. • Flooding can solubilize the evaporites in the silt layer and release chloride, sulfate (not shown), and uranium into the underlyingsand and gravel aquifer. • The uranium-rich evaporites can delay natural flushing, creating plume persistence near the Little Wind River.

Objective. To review, from a legal perspective, the potential for using the Internet for inter-institutional transfer of patient medical records. Methods. Basic issues and recent legislation that relate to protection of both medical data, and those transferring that data over public network systems is reviewed. Results. Many laws already in existence can be applied to Internet transmission, but questions of jurisdiction remain. Providing signatures on requests for information, which are in essence contracts, is a problem. Signatures must both prove the identity of the participants and provide for non-repudiation of the agreement. Cryptographic digital signatures appear secure and effective, but their use is difficult to implement. Simpler methods are fraught with risks, yet are more easily accomplished. The patient's rights of privacy must be balanced against the need for access by government, physician, or healthcare institutions to confidential information. In general, information holders must put forth reasonable efforts to keep information confidential. The development of acknowledged standards will provide guidance. Multiple laws provide some deterrence and hence some reassurance to healthcare institutions, for example, by criminalizing acts of electronic interception of patient records in transit. Conclusion. Some believe the expense of secure transfer of medical records by electronic means is a major obstacle; this is false: such transfers are now technologically quite easy. The greatest obstacle to electronic transfer of medical records at this point is the development of workable standards for signing agreements and protecting transmissions, but the perceived advantages will likely drive the necessary developments.

We present the first integrated multimode photonic spectrograph, a device we call PIMMS #1. The device comprises a set of multimode fibres that convert to single-mode propagation using a matching set of photonic lanterns. These feed to a stack of cyclic array waveguides (AWGs) that illuminate a common detector. Such a device greatly reduces the size of an astronomical instrument at a fixed spectroscopic resolution. Remarkably, the PIMMS concept is largely independent of the telescope diameter, input focal ratio and entrance aperture – i.e. one size fits all! The instrument architecture can also exploit recent advances in astrophotonics (e.g . OH suppression fibres). We present a movie of the instrument’s operation and discuss the advantages and disadvantages of this approach. Keywords: multimode fibre; integrated photonic spectrograph; astronomical instrumentation – diffraction limited 1. INTRODUCTION In 2006, we explored the possibility of designing and building an integrated photonic spectrograph for the first time (Bland-Hawthorn & Horton 2006; hereafter BH06). The stated goals in that paper were to revolutionize our approach to instrument building in order to halt the spiraling trends in instrument development. In the following year, such a device was made to operate with a single-mode fibre input (Fig. 1; Bland-Hawthorn & Horton 2007). In 2008, the instrument was used successfully at Siding Spring Observatory, NSW which led to the first continuous spectrum ever obtained with a photonic device (Cvetojevic et al 2009). As anticipated, the coupling efficiency of diffuse incoherent skylight was very low but we still managed to observe the familiar OH atmospheric lines across the H band. For this technology to be viable on front-line telescopes, it is imperative that we extend it to accept light from a multimode fibre. This is a more challenging prospect that has not been solved until now. In solving this problem, we arrive at a device – PIMMS #1 – with revolutionary properties as we show. In principle, the PIMMS instrument will accept light from (almost) any input f/ratio, either natural seeing or AO-corrected, any input fibre diameter, and any telescope diameter. For spectroscopic resolutions in the range of 100s to a few u 10

We consider lamellar gratings made of dielectric or lossy materials used in classical diffraction mounts. We show how the modal diffraction formulation may be generalized to deal with slanted lamellar gratings and illustrate the accuracy and versatility of the new method through study of highly slanted gratings in a homog- enization limit. We also comment on the completeness of the eigenmode basis and present tests enabling this completeness to be verified numerically. © 2008 Optical Society of America.

Results of a nationwide study of managerialperceptions ofperformance hindrances in a Fortune 100 Corporation (Honeywell) is reported. This paper relies on skill relapse-prevention theories and self-regulation theories to interpret the perceptual responses of 3,580 managers concerning relevant performance hindrances. The managers were divided into "higher" and "lower" performance groups, and perceptual response patterns regarding performance obstacles were contrasted between the two groups. In general, it was found that, while no apparent difference occurred between the number of hindrances perceived, higher performing managers more often focused on external performance hindrances, e.g., lack of promotion opportunities and non-performance-related appraisal systems. etc., while lower performers focused more on personal skill deficiencies, e.g., lack of people-managing and communication skills, and lack of technical skills, etc.

Milling activities at a former uranium mill site near Riverton, Wyoming, USA, contaminated the shallow groundwater beneath and downgradient of the site. Although the mill operated for

In the “real world”, Photonics is somewhat invisible to those who rely upon it worldwide. We would like students to connect their everyday experiences of communications with the underlying ideas in Photonics. To do this, we have developed the Photonics Simulator to illustrate to high school students how text or information is coded into binary optical signals which are relayed through photonic communications networks from sender to receiver. Using our simulator, students construct a virtual network, and then test it by sending messages. The messages are coded using ASCII binary code as digital signals in data packets with address headers, which need to be switched, combined, amplified, or delayed to get to their designated address. The students must manage their power budget, correctly target each message address, and avoid collisions of data packets to send their messages uncorrupted and error-free. We tested an early version of the simulator with five Year 9 and 10 classes. The students provided many constructive comments and their feedback was used to improve the graphical interface of the simulator. We subsequently tested the simulator with 80 Year 9 students in short workshops. Overall we had a very positive response - it was more fun than a normal class, and interactivity helped students retain information. Students enjoy the visual aspects— they see how messages are delivered, and learn the function of each network component by experiment. Tests of the simulator at the Macquarie Siemens Science Experience were also encouraging, with one student even sneaking back to class to complete his challenges!