Weak Lensing for Precision Cosmology Mandelbaum, Rachel
Annual review of astronomy and astrophysics,
09/2018, Letnik:
56, Številka:
1
Journal Article
Recenzirano
Odprti dostop
Weak gravitational lensing, the deflection of light by mass, is one of the best tools to constrain the growth of cosmic structure with time and reveal the nature of dark energy. I discuss the sources ...of systematic uncertainty in weak lensing measurements and their theoretical interpretation, including our current understanding and other options for future improvement. These include long-standing concerns such as the estimation of coherent shears from galaxy images or redshift distributions of galaxies selected on the basis of photometric redshifts, along with systematic uncertainties that have received less attention to date because they are subdominant contributors to the error budget in current surveys. I also discuss methods for automated systematics detection using survey data of the 2020s. The goal of this review is to describe the current state of the field and what must be done so that if weak lensing measurements lead toward surprising conclusions about key questions such as the nature of dark energy, those conclusions will be credible.
We develop a simple yet comprehensive method to distinguish the underlying drivers of galaxy quenching, using the clustering and galaxy–galaxy lensing of red and blue galaxies in Sloan Digital Sky ...Survey. Building on the iHOD framework developed by Zu & Mandelbaum, we consider two quenching scenarios: (1) a ‘halo’ quenching model in which halo mass is the sole driver for turning off star formation in both centrals and satellites; and (2) a ‘hybrid’ quenching model in which the quenched fraction of galaxies depends on their stellar mass, while the satellite quenching has an extra dependence on halo mass. The two best-fitting models describe the red galaxy clustering and lensing equally well, but halo quenching provides significantly better fits to the blue galaxies above 1011 h
−2 M⊙. The halo quenching model also correctly predicts the average halo mass of the red and blue centrals, showing excellent agreement with the direct weak lensing measurements of locally brightest galaxies. Models in which quenching is not tied to halo mass, including an age-matching model in which galaxy colour depends on halo age at fixed M
*, fail to reproduce the observed halo mass for massive blue centrals. We find similar critical halo masses responsible for the quenching of centrals and satellites (∼1.5 × 1012 h
−1 M⊙), hinting at a uniform quenching mechanism for both, e.g. the virial shock heating of infalling gas. The success of the iHOD halo quenching model provides strong evidence that the physical mechanism that quenches star formation in galaxies is tied principally to the masses of their dark matter haloes rather than the properties of their stellar components.
To validate the revised 2018 International Federation of Gynecology and Obstetrics (FIGO) staging system for cervical cancer, with a particular focus on stage IB and stage III disease.
Two ...retrospective cohort studies were conducted using The Surveillance, Epidemiology, and End Results Program between 1988 and 2014. The stage IB cohort consisted of node-negative FIGO stage IB1 (tumor size <2 cm), IB2 (2–3.9 cm), and IB3 (≥4 cm) cervical cancer. The stage III cohort consisted of FIGO stage IIIA, IIIB, and stage IIIC1 (any pelvic nodal metastasis) cervical cancer. Multivariable analysis was performed for cause-specific survival based on cancer stage.
In the stage IB cohort (n = 8909), stage IB1 tumors were more likely to be adenocarcinoma and low-grade compared to other the groups (P < 0.001). On multivariable analysis, stage IB2 disease was independently associated with a nearly two-fold increased risk of cervical cancer mortality compared to stage IB1 disease (adjusted-hazard ratio HR 1.98, 95% confidence interval CI 1.62–2.41, P < 0.001). In the stage III cohort (n = 11,733), stage IIIC1 was independently associated with improved cause-specific survival compared to stage IIIB disease (adjusted-HR 0.79, 95%CI 0.74–0.85, P < 0.001). Survival of stage IIIC1 disease significantly differed based on T = stage, (5-year rates: 74.8% for T1, 58.7% for T2, and 39.3% for T3) with a 35.3% difference in absolute survival (P < 0.001).
The 2018 FIGO staging system for cervical cancer is useful to distinguish survival groups; stage IB1 and stage IB2 disease have distinct characteristics and survival outcomes, while survival in stage IIIC1 varies depending on local tumor factors.
FIGO revised cervical cancer staging in 2018•Revised staging was validated in a population-based tumor registry.•Stage IB1 and IB2 disease have distinct tumor characteristics and survival.•Stage IIIC1 disease has superior survival compared to stage IIIA-B disease.•Survival of stage IIIC1 disease depends on local tumor factors.
Intrinsic alignments (IA) of galaxies, i.e. correlations of galaxy shapes with each other (II) or with the density field (gI), are potentially a major astrophysical source of contamination for weak ...lensing surveys. We present the results of IA measurements of galaxies on 0.1–200 h
−1 Mpc scales using the SDSS-III BOSS low-redshift (LOWZ) sample, in the redshift range 0.16 < z < 0.36. We extend the existing IA measurements for spectroscopic luminous red galaxies (LRGs) to lower luminosities, and show that the luminosity dependence of large-scale IA can be well described by a power law. Within the limited redshift and colour range of our sample, we observe no significant redshift or colour dependence of IA. We measure the halo mass of galaxies using galaxy–galaxy lensing, and show that the mass dependence of large-scale IA is also well described by a power law. We detect variations in the scale dependence of IA with mass and luminosity, which underscores the need to use flexible templates in order to remove the IA signal. We also study the environment dependence of IA by splitting the sample into field and group galaxies, which are further split into satellite and central galaxies. We show that group central galaxies are aligned with their haloes at small scales and also are aligned with the tidal fields out to large scales. We also detect the radial alignments of satellite galaxies within groups. These results can be used to construct better IA models for removal of this contaminant to the weak lensing signal.
Measurements of intrinsic alignments of galaxy shapes with the large-scale density field, and the inferred intrinsic alignments model parameters, are sensitive to the shape measurement methods used. ...In this paper, we measure the intrinsic alignments of the Sloan Digital Sky Survey-III (SDSS-III) Baryon Oscillation Spectroscopic Survey (BOSS) low redshift (LOWZ) galaxies using three different shape measurement methods (re-Gaussianization, isophotal, and de Vaucouleurs), identifying a variation in the inferred intrinsic alignments amplitude at the 40 per cent level between these methods, independent of the galaxy luminosity or other properties. We also carry out a suite of systematics tests on the shapes and their two-point correlation functions, identifying a pronounced contribution from additive point spread function systematics in the de Vaucouleurs shapes. Since different methods measure galaxy shapes at different effective radii, the trends we identify in the intrinsic alignments amplitude are consistent with the interpretation that the outer regions of galaxy shapes are more responsive to tidal fields, resulting in isophote twisting and stronger alignments for isophotal shapes. We observe environment dependence of ellipticity, with brightest galaxies in groups being rounder on average compared to satellite and field galaxies. We also study the anisotropy in intrinsic alignments measurements introduced by projected shapes, finding effects consistent with predictions of the non-linear alignment model and hydrodynamic simulations. The large variations seen using the different shape measurement methods have important implications for intrinsic alignments forecasting and mitigation with future surveys.
Although an infrequent occurrence, the placenta can adhere abnormally to the gravid uterus leading to significantly high maternal morbidity and mortality during cesarean delivery. Contemporary ...national statistics related to a morbidly adherent placenta, referred to as placenta accreta spectrum, are needed.
This study aimed to examine national trends, characteristics, and perioperative outcomes of women who underwent cesarean delivery for placenta accreta spectrum in the United States.
This is a population-based retrospective, observational study querying the National Inpatient Sample. The study cohort included women who underwent cesarean delivery from October 2015 to December 2017 and had a diagnosis of placenta accreta spectrum. The main outcome measures were patient characteristics and surgical outcomes related to placenta accreta spectrum assessed by the generalized estimating equation on multivariable analysis. The temporal trend of placenta accreta spectrum was also assessed by linear segmented regression with log transformation.
Of 2,727,477 cases who underwent cesarean delivery during the study period, 8030 (0.29%) had the diagnosis of placenta accreta spectrum. Placenta accreta was the most common diagnosis (n=6205, 0.23%), followed by percreta (n=1060, 0.04%) and increta (n=765, 0.03%). The number of placenta accreta spectrum cases increased by 2.1% every quarter year from 0.27% to 0.32% (P=.004). On multivariable analysis, (1) patient demographics (older age, tobacco use, recent diagnosis, higher comorbidity, and use of assisted reproductive technology), (2) pregnancy characteristics (placenta previa, previous cesarean delivery, breech presentation, and grand multiparity), and (3) hospital factors (urban teaching center and large bed capacity hospital) represented the independent characteristics related to placenta accreta spectrum (all, P<.05). The median gestational age at cesarean delivery was 36 weeks for placenta accreta and 34 weeks for both placenta increta and percreta vs 39 weeks for non–placenta accreta spectrum cases (P<.001). On multivariable analysis, cesarean delivery complicated by placenta accreta spectrum was associated with increased risk of any surgical morbidities (78.3% vs 10.6%), Centers for Disease Control and Prevention–defined severe maternal morbidity (60.3% vs 3.1%), hemorrhage (54.1% vs 3.9%), coagulopathy (5.3% vs 0.3%), shock (5.0% vs 0.1%), urinary tract injury (8.3% vs 0.2%), and death (0.25% vs 0.01%) compared with cesarean delivery without placenta accreta spectrum. When further analyzed by subtype, cesarean delivery for placenta increta and percreta was associated with higher likelihood of hysterectomy (0.4% for non–placenta accreta spectrum, 45.8% for accreta, 82.4% for increta, 78.3% for percreta; P<.001) and urinary tract injury (0.2% for non–placenta accreta spectrum, 5.2% for accreta, 11.8% for increta, 24.5% for percreta; P<.001). Moreover, women in the placenta increta and percreta groups had markedly increased risks of surgical mortality compared with those without placenta accreta spectrum (increta, odds ratio, 19.9; and percreta, odds ratio, 32.1).
Patient characteristics and outcomes differ across the placenta accreta spectrum subtypes, and women with placenta increta and percreta have considerably high surgical morbidity and mortality risks. Notably, 1 in 313 women undergoing cesarean delivery had a diagnosis of placenta accreta spectrum by the end of 2017, and the incidence seems to be higher than reported in previous studies.
Historically, the Cox proportional hazard regression model has been the mainstay for survival analyses in oncologic research. The Cox proportional hazard regression model generally is used based on ...an assumption of linear association. However, it is likely that, in reality, there are many clinicopathologic features that exhibit a nonlinear association in biomedicine.
The purpose of this study was to compare the deep-learning neural network model and the Cox proportional hazard regression model in the prediction of survival in women with cervical cancer.
This was a retrospective pilot study of consecutive cases of newly diagnosed stage I–IV cervical cancer from 2000–2014. A total of 40 features that included patient demographics, vital signs, laboratory test results, tumor characteristics, and treatment types were assessed for analysis and grouped into 3 feature sets. The deep-learning neural network model was compared with the Cox proportional hazard regression model and 3 other survival analysis models for progression-free survival and overall survival. Mean absolute error and concordance index were used to assess the performance of these 5 models.
There were 768 women included in the analysis. The median age was 49 years, and the majority were Hispanic (71.7%). The majority of tumors were squamous (75.3%) and stage I (48.7%). The median follow-up time was 40.2 months; there were 241 events for recurrence and progression and 170 deaths during the follow-up period. The deep-learning model showed promising results in the prediction of progression-free survival when compared with the Cox proportional hazard regression model (mean absolute error, 29.3 vs 316.2). The deep-learning model also outperformed all the other models, including the Cox proportional hazard regression model, for overall survival (mean absolute error, Cox proportional hazard regression vs deep-learning, 43.6 vs 30.7). The performance of the deep-learning model further improved when more features were included (concordance index for progression-free survival: 0.695 for 20 features, 0.787 for 36 features, and 0.795 for 40 features). There were 10 features for progression-free survival and 3 features for overall survival that demonstrated significance only in the deep-learning model, but not in the Cox proportional hazard regression model. There were no features for progression-free survival and 3 features for overall survival that demonstrated significance only in the Cox proportional hazard regression model, but not in the deep-learning model.
Our study suggests that the deep-learning neural network model may be a useful analytic tool for survival prediction in women with cervical cancer because it exhibited superior performance compared with the Cox proportional hazard regression model. This novel analytic approach may provide clinicians with meaningful survival information that potentially could be integrated into treatment decision-making and planning. Further validation studies are necessary to support this pilot study.
We use galaxy–galaxy lensing to study the dark matter haloes surrounding a sample of locally brightest galaxies (LBGs) selected from the Sloan Digital Sky Survey. We measure mean halo mass as a ...function of the stellar mass and colour of the central galaxy. Mock catalogues constructed from semi-analytic galaxy formation simulations demonstrate that most LBGs are the central objects of their haloes, greatly reducing interpretation uncertainties due to satellite contributions to the lensing signal. Over the full stellar mass range, 10.3 < log M
*/M⊙ < 11.6, we find that passive central galaxies have haloes that are at least twice as massive as those of star-forming objects of the same stellar mass. The significance of this effect exceeds 3σ for log M
*/M⊙ > 10.7. Tests using the mock catalogues and on the data themselves clarify the effects of LBG selection and show that it cannot artificially induce a systematic dependence of halo mass on LBG colour. The bimodality in halo mass at fixed stellar mass is reproduced by the astrophysical model underlying our mock catalogue, but the sign of the effect is inconsistent with recent, nearly parameter-free age-matching models. The sign and magnitude of the effect can, however, be reproduced by halo occupation distribution models with a simple (few-parameter) prescription for type dependence.
On the level of cluster assembly bias in SDSS Zu, Ying; Mandelbaum, Rachel; Simet, Melanie ...
Monthly notices of the Royal Astronomical Society,
09/2017, Letnik:
470, Številka:
1
Journal Article
Recenzirano
Abstract
Recently, several studies have discovered a strong discrepancy between the large-scale clustering biases of two subsamples of galaxy clusters at the same halo mass, split by their average ...projected membership distances 〈Rmem〉. The level of this discrepancy significantly exceeds the maximum halo assembly bias predicted by Λ cold dark matter (ΛCDM). We explore whether some of the large-scale bias differences could be caused by projection effects in 〈Rmem〉 due to other systems along the line of sight. We thoroughly investigate the assembly bias of the redMaPPer clusters in Sloan Digital Sky Survey (SDSS), by defining a new variant of the average membership distance estimator $\tilde{R}_{\mathrm{mem}}$ that is robust against projection effects in the cluster membership identification. Using the angular mark correlation functions, we show that the large-scale bias differences when splitting by 〈Rmem〉 can be mostly attributed to projection effects. After splitting by $\tilde{R}_{\mathrm{mem}}$, the anomalously large signal is reduced, giving a ratio of 1.02 ± 0.14 between the two clustering biases as measured from weak lensing. Using a realistic mock cluster catalogue, we predict that the bias ratio between two $\tilde{R}_{\mathrm{mem}}$-split subsamples should be ≃1.10, which is >60 per cent weaker than the maximum halo assembly bias (1.24) when split by halo concentration. Therefore, our results demonstrate that the level of halo assembly bias exhibited by clusters in SDSS is consistent with the ΛCDM prediction. With a 10-fold increase in cluster numbers, deeper ongoing surveys will enable a more robust detection of halo assembly bias. Our findings also have important implications for quantifying the impact of projection effects on cosmological constraints using photometrically selected clusters.