University of Texas at Austin

About: University of Texas at Austin is a education organization based out in Austin, Texas, United States. It is known for research contribution in the topics: Population & Poison control. The organization has 94352 authors who have published 206297 publications receiving 9070052 citations. The organization is also known as: UT-Austin & UT Austin.

Structural topic models for open ended survey responses

Abstract: Collection and especially analysis of open-ended survey responses are relatively rare in the discipline and when conducted are almost exclusively done through human coding. We present an alternative, semiautomated approach, the structural topic model (STM) (Roberts, Stewart, and Airoldi 2013; Roberts et al. 2013), that draws on recent developments in machine learning based analysis of textual data. A crucial contribution of the method is that it incorporates information about the document, such as the author's gender, political affiliation, and treatment assignment (if an experimental study). This article focuses on how the STM is helpful for survey researchers and experimentalists. The STM makes analyzing open-ended responses easier, more revealing, and capable of being used to estimate treatment effects. We illustrate these innovations with analysis of text from surveys and experiments.

A Randomized Trial Comparing Perioperative Chemotherapy and Surgery With Surgery Alone in Resectable Stage IIIA Non-Small-Cell Lung Cancer

Abstract: BACKGROUND Patients with resectable stage IIIA non-small-cell lung cancer have a low survival rate following standard surgical treatment. Nonrandomized trials in which induction chemotherapy or a combination of chemotherapy and radiation prior to surgery were used to treat patients with regionally advanced primary cancers have suggested that survival is improved when compared with treatment by surgery alone. PURPOSE We performed a prospective, randomized study of patients with previously untreated, potentially resectable clinical stage IIIA non-small-cell lung cancer to compare the results of perioperative chemotherapy and surgery with those of surgery alone. METHODS This trial was designed to test the null hypothesis that the proportion of patients surviving 3 years is 12% for either treatment group against the alternate hypothesis that the 3-year survival rate would be 12% in the surgery alone group and 32% in the perioperative chemotherapy group. The estimated required sample size was 65 patients in each group. The trial was terminated at an early time according to the method of O'Brien and Fleming following a single unplanned interim analysis. The decision to terminate the trial was based on ethical considerations, the magnitude of the treatment effect, and the high degree of statistical significance attained. In total, 60 patients were randomly assigned between 1987 and 1993 to receive either six cycles of perioperative chemotherapy (cyclophosphamide, etoposide, and cisplatin) and surgery (28 patients) or surgery alone (32 patients). For patients in the former group, tumor measurements were made before each course of chemotherapy and the clinical tumor response was evaluated after three cycles of chemotherapy; they then underwent surgical resection. Patients who had documented tumor regression after preoperative chemotherapy received three additional cycles of chemotherapy after surgery. RESULTS After three cycles of preoperative chemotherapy, the rate of clinical major response was 35%. Patients treated with perioperative chemotherapy and surgery had an estimated median survival of 64 months compared with 11 months for patients who had surgery alone (P < .008 by log-rank test; P < .018 by Wilcoxon test). The estimated 2- and 3-year survival rates were 60% and 56% for the perioperative chemotherapy patients and 25% and 15% for those who had surgery alone, respectively. CONCLUSIONS In this trial, the treatment strategy using perioperative chemotherapy and surgery was more effective than surgery alone. IMPLICATIONS This clinical trial strengthens the validity of using perioperative chemotherapy in the management of patients with resectable stage IIIA non-small-cell lung cancer. Further investigation of the perioperative chemotherapy strategy in earlier stage lung cancer is warranted.

Structure and formation of elliptical and spheroidal galaxies

Abstract: New surface photometry of all known elliptical galaxies in the Virgo cluster is combined with published data to derive composite profiles of brightness, ellipticity, position angle, isophote shape, and color over large radius ranges These provide enough leverage to show that S?rsic log I r 1/n functions fit the brightness profiles I(r) of nearly all ellipticals remarkably well over large dynamic ranges Therefore, we can confidently identify departures from these profiles that are diagnostic of galaxy formation Two kinds of departures are seen at small radii All 10 of our ellipticals with total absolute magnitudes MVT ? ?2166 have cuspy cores?missing light?at small radii Cores are well known and naturally scoured by binary black holes (BHs) formed in dissipationless (dry) mergers All 17 ellipticals with ?2154 ? MVT ? ?1553 do not have cores We find a new distinct component in these galaxies: all coreless ellipticals in our sample have extra light at the center above the inward extrapolation of the outer S?rsic profile In large ellipticals, the excess light is spatially resolved and resembles the central components predicted in numerical simulations of mergers of galaxies that contain gas In the simulations, the gas dissipates, falls toward the center, undergoes a starburst, and builds a compact stellar component that, as in our observations, is distinct from the S?rsic-function main body of the elliptical But ellipticals with extra light also contain supermassive BHs We suggest that the starburst has swamped core scouring by binary BHs That is, we interpret extra light components as a signature of formation in dissipative (wet) mergers Besides extra light, we find three new aspects to the (E-E) dichotomy into two types of elliptical galaxies Core galaxies are known to be slowly rotating, to have relatively anisotropic velocity distributions, and to have boxy isophotes We show that they have S?rsic indices n > 4 uncorrelated with MVT They also are ?-element enhanced, implying short star-formation timescales And their stellar populations have a variety of ages but mostly are very old Extra light ellipticals generally rotate rapidly, are more isotropic than core Es, and have disky isophotes We show that they have n 3 ? 1 almost uncorrelated with MVT and younger and less ?-enhanced stellar populations These are new clues to galaxy formation We suggest that extra light ellipticals got their low S?rsic indices by forming in relatively few binary mergers, whereas giant ellipticals have n > 4 because they formed in larger numbers of mergers of more galaxies at once plus later heating during hierarchical clustering We confirm that core Es contain X-ray-emitting gas whereas extra light Es generally do not This leads us to suggest why the E-E dichotomy arose If energy feedback from active galactic nuclei (AGNs) requires a working surface of hot gas, then this is present in core galaxies but absent in extra light galaxies We suggest that AGN energy feedback is a strong function of galaxy mass: it is weak enough in small Es not to prevent merger starbursts but strong enough in giant Es and their progenitors to make dry mergers dry and to protect old stellar populations from late star formation Finally, we verify that there is a strong dichotomy between elliptical and spheroidal galaxies Their properties are consistent with our understanding of their different formation processes: mergers for ellipticals and conversion of late-type galaxies into spheroidals by environmental effects and by energy feedback from supernovae In an appendix, we develop machinery to get realistic error estimates for S?rsic parameters even when they are strongly coupled And we discuss photometric dynamic ranges necessary to get robust results from S?rsic fits

OVERVIEW: Neoproterozoic-Paleozoic geography and tectonics: Review, hypothesis, environmental speculation

Abstract: The ever-changing distribution of continents and ocean basins on Earth is fundamental to the environment of the planet. Recent ideas regarding pre-Pangea geography and tectonics offer fresh opportunities to examine possible causative relations between tectonics and environmental and biologic changes during the Neoproterozoic and Paleozoic eras. The starting point is an appreciation that Laurentia, the rift-bounded Precambrian core of North America, could have been juxtaposed with the cratonic cores of some present-day southern continents. This has led to reconstructions of Rodinia and Pannotia, supercontinents that may have existed in early and latest Neoproterozoic time, respectively, before and after the opening of the Pacific Ocean basin. Recognition that the Precordillera of northwest Argentina constitutes a terrane derived from Laurentia may provide critical longitudinal control on the relations of that craton to Gondwana during the Precambrian-Cambrian boundary transition, and in the early Paleozoic. The Precordillera was most likely derived from the general area of the Ouachita embayment, and may have been part of a hypothetical promontory of Laurentia, the “Texas plateau,” which was detached from the Cape of Good Hope embayment within Gondwana between the present-day Falkland-Malvinas Plateau and Transantarctic Mountains margins. Thus the American continents may represent geometric “twins” detached from the Pannotian and Pangean supercontinents in Early Cambrian and Early Cretaceous time, respectively—the new mid-ocean ridge crests of those times initiating the two environmental supercycles of Phanerozoic history 400 m.y. apart. In this scenario, the extremity of the Texas plateau was detached from Laurentia during the Caradocian Epoch, in a rift event ca. 455 Ma that followed Middle Ordovician collision with the proto-Andean margin of Gondwana as part of the complex Indonesian-style Taconic-Famatinian orogeny, which involved several island arc-continent collisions between the two major continental entities. Laurentia then continued its clockwise relative motion around the proto-Andean margin, colliding with other arc terranes, Avalonia, and Baltica en route to the Ouachita-Alleghanian-Hercynian-Uralian collision that completed the amalgamation of Pangea. The important change in single-celled organisms at the Mesoproterozoic-Neoproterozoic boundary (1000 Ma) accompanied assembly of Rodinia along Grenvillian sutures. Possible divergence of metazoan phyla, the appearance and disappearance of the Ediacaran fauna (ca. 650–545 Ma), and the Cambrian “explosion” of skeletalized metazoans (ca. 545–500 Ma) also appear to have taken place within the framework of tectonic change of truly global proportions. These are the opening of the Pacific Ocean basin; uplift and erosion of orogens within the newly assembled Gondwana portion of Pannotia, including a collisional mountain range extending ≈7500 km from Arabia to the Pacific margin of Antarctica; the development of a Pannotia-splitting oceanic spreading ridge system nearly 10 000 km long as Laurentia broke away from Gondwana, Baltica, and Siberia; and initiation of subduction zones along thousands of kilometres of the South American and Antarctic-Australian continental margins. The Middle Ordovician sea-level changes and biologic radiation broadly coincided with initiation of the Appalachian-Andean mountain system along >7000 km of the Taconic and Famatinian belts. These correlations, based on testable paleogeographic reconstructions, invite further speculation about possible causative relations between the internally driven long-term tectonic evolution of the planet, its surface environment, and life.