Thanks to the rapidly evolving integration of LLMs into decision-support tools, a significant transformation is happening across large-scale systems. Like other medical fields, the use of LLMs such as GPT-4 is gaining increasing interest in radiation oncology as well. An attempt to assess GPT-4's performance in radiation oncology was made via a dedicated 100-question examination on the highly specialized topic of radiation oncology physics, revealing GPT-4's superiority over other LLMs. GPT-4's performance on a broader field of clinical radiation oncology is further benchmarked by the ACR Radiation Oncology In-Training (TXIT) exam where GPT-4 achieved a high accuracy of 74.57%. Its performance on re-labelling structure names in accordance with the AAPM TG-263 report has also been benchmarked, achieving above 96% accuracies. Such studies shed light on the potential of LLMs in radiation oncology. As interest in the potential and constraints of LLMs in general healthcare applications continues to rise5, the capabilities and limitations of LLMs in radiation oncology decision support have not yet been fully explored.
Elucidating the statistical properties of extreme meteo-climatic events and capturing the physical processes responsible for their occurrence are key steps for improving our understanding of climate variability and climate change and for better evaluating the associated hazards. It has recently become apparent that large deviation theory is very useful for investigating persistent extreme events, and specifically, for flexibly estimating long return periods and for introducing a notion of dynamical typicality. Using a methodological framework based on large deviation theory and taking advantage of long simulations by a state-of-the-art Earth System Model, we investigate the 2021 North America Heatwave. Indeed, our analysis shows that the 2021 event can be seen as an unlikely but possible manifestation of climate variability, whilst its probability of occurrence is greatly amplified by the ongoing climate change. We also clarify the properties of spatial coherence of the 2021 heatwave and elucidate the role played by the Rocky Mountains in modulating hot, dry, and persistent extreme events in the Western Pacific region of North America.
We present observations of near-infrared 2.12 micro-meter molecular hydrogen outflows emerging from 1.1 mm dust continuum clumps in the North America and Pelican Nebula (NAP) complex selected from the Bolocam Galactic Plane Survey (BGPS). Hundreds of individual shocks powered by over 50 outflows from young stars are identified, indicating that the dusty molecular clumps surrounding the NGC 7000 / IC 5070 / W80 HII region are among the most active sites of on-going star formation in the Solar vicinity. A spectacular X-shaped outflow, MHO 3400, emerges from a young star system embedded in a dense clump more than a parsec from the ionization front associated with the Pelican Nebula (IC 5070). Suspected to be a binary, the source drives a pair of outflows with orientations differing by 80 degrees. Each flow exhibits S-shaped symmetry and multiple shocks indicating a pulsed and precessing jet. The `Gulf of Mexico' located south of the North America Nebula (NGC 7000), contains a dense cluster of molecular hydrogen objects (MHOs), Herbig-Haro (HH) objects, and over 300 YSOs, indicating a recent burst of star formation. The largest outflow detected thus far in the North America and Pelican
In the area covering the complex of the North America and Pelican nebulae we identified 13 faint stars with J-H and H-Ks color indices which simulate heavily reddened O-type stars. One of these stars is CP05-4 classified as O5 V by Comeron and Pasquali (2005). Combining magnitudes of these stars in the passbands I, J, H, Ks and [8.3] we were able to suspect that two of them are carbon stars and five are late M-type AGB stars. Interstellar extinction in the direction of these stars was estimated from the background red clump giants in the J-H vs. H-Ks diagram and from star counts in the Ks passband. Four or five stars are found to have a considerable probability of being O-type stars, contributing to the ionization of North America and Pelican. If they really are O-type stars, their interstellar extinction A(V) should be from 16 to 35 mag. Two of them seem to be responsible for bright E and J radio rims discovered by Matthews and Goss (1980).
We propose a Bayesian, noisy-input, spatial-temporal generalised additive model to examine regional relative sea-level (RSL) changes over time. The model provides probabilistic estimates of component drivers of regional RSL change via the combination of a univariate spline capturing a common regional signal over time, random slopes and intercepts capturing site-specific (local), long-term linear trends and a spatial-temporal spline capturing residual, non-linear, local variations. Proxy and instrumental records of RSL and corresponding measurement errors inform the model and a noisy-input method accounts for proxy temporal uncertainties. Results focus on the decomposition of RSL over the past 3000 years along the Atlantic coast of North America.
We present and discuss broad band CCD $UBV(I)_C$ photometry and low resolution spectroscopy for stars in the region of the open cluster NGC 6996, located in the North America Nebula. The new data allow us to tightly constrain the basic properties of this object. We revise the cluster size, which in the past has been significantly underestimated. The width of the Main Sequence is mainly interpreted in terms of differential reddening, and indeed the stars' color excess $E_{B-V}$ ranges from 0.43 to 0.65, implying the presence of a significant and evenly distributed dust component. We cross-correlate our optical photometry with near infrared from 2MASS, and by means of spectral classification we are able to build up extinction curves for an handful of bright members. We find that the reddening slope and the total to selective absorption ratio $R_V$ toward NGC 6996 are anomalous. Moreover the reddening corrected colors and magnitudes allow us to derive estimates for the cluster distance and age, which turn out to be $760 \pm 70 pc$ ($V_{0}-M_{V} = 9.4 \pm 0.2$) and $\sim 350$ Myr, respectively. Basing on our results, we suggest that NGC 6996 is located in front of the North America Neb
We present spectroscopic observations of the double-lined early type eclipsing binary V1898\,Cyg. The radial velocities were obtained by means of the cross-correlation technique. Analyses of the BV light curves and RVs led to determination of the fundamental stellar parameters of the V1898\,Cyg's components. We derived new ephemerides for the eclipsing pair using the observed times of mid-eclipses. The residuals between the observed and computed times of mid-eclipses were analysed and a rate of the period change $\dot{P}/P= 6.68 \times 10^{-7}\,yr^{-1}$ was obtained. The orbital period is increased by about 0.38 s in the last 24 years due to the mass transfer from less massive secondary to the more massive primary star with an amount of $1.88\times10^{-7}\,$ \Msun in a year, assuming conservative case. Results of the light and radial velocity curves' analyses were combined and the physical parameters of the components were revealed. The absolute parameters for the stars are derived as: M$_1$=6.054$\pm$0.037 M$_{\odot}$, M$_2$=1.162$\pm$0.011 M$_{\odot}$, R$_1$=3.526$\pm$0.009 R$_{\odot}$, R$_2$=2.640$\pm$0.010 R$_{\odot}$, T$_{eff_1}$=18\,000$\pm$600 K, and T$_{eff_2}$=6\,200$\pm$2
We evaluate the performance of various configurations of the Canadian Regional Climate Model (CRCM6-GEM5) in simulating 10-meter wind speeds using data from 27 AmeriFlux stations across North America. The assessment employs a hierarchy of error metrics, ranging from simple mean bias to advanced metrics that account for the dependence of wind speeds on variables such as friction velocity and stability. The results reveal that (i) the value of roughness length (z0) has a large effect on the simulation of wind speeds, (ii) using a lower limit for the Obhukov length instead of a lower limit for the lowest level wind speed seems to deteriorate the simulation of wind speeds under very stable conditions, (iii) the choice of stability function has a small but noticeable impact on the wind speeds, (iv) the turbulent orographic form drag scheme shows improvement over effective roughness length approach.
The development of a kilometer-scale E3SM Land Model (km-scale ELM) is an integral part of the E3SM project, which seeks to advance energy-related Earth system science research with state-of-the-art modeling and simulation capabilities on exascale computing systems. Through the utilization of high-fidelity data products, such as atmospheric forcing and soil properties, the km-scale ELM plays a critical role in accurately modeling geographical characteristics and extreme weather occurrences. The model is vital for enhancing our comprehension and prediction of climate patterns, as well as their effects on ecosystems and human activities. This study showcases the first set of full-capability, km-scale ELM simulations over various computational domains, including simulations encompassing 21.6 million land gridcells, reflecting approximately 21.5 million square kilometers of North America at a 1 km x 1 km resolution. We present the largest km-scale ELM simulation using up to 100,800 CPU cores across 2,400 nodes. This continental-scale simulation is 300 times larger than any previous studies, and the computational resources used are about 400 times larger than those used in prior efforts
We present a spectroscopic survey of over 3400 potential members in the North America and Pelican nebulae (NAP) using several low-resolution ($R\approx$ 1300-2000) spectrographs: Palomar/Norris, WIYN/HYDRA, Keck/DEIMOS, and MMT/Hectospec. We identify 580 young stars as likely members of the NAP region based on criteria involving infrared excess, Li I 6708 absorption, X-ray emission, parallax, and proper motions. The spectral types of individual spectra are derived by fitting them with templates that are either empirical spectra of pre-main sequence stars, or model atmospheres. The templates are artificially veiled, and a best-fit combination of spectral type and veiling parameter is derived for each star. We use the spectral types with archival photometry to derive $V$-band extinction and stellar luminosity. From the H-R diagram, the median age of the young stars is about 1 Myr, with a luminosity dispersion of $\sim$0.3--0.4 dex. We investigate the photometric variability of the spectroscopic member sample using ZTF data, and conclude that photometric variability, while present, does not significantly contribute to the luminosity dispersion. While larger than the formal errors, the
Far red spectra for 34 stars with V magnitudes between 15 and 18 in the direction of the North America and Pelican nebulae (NAP) star-forming region are obtained. Some of these stars were known earlier as emission-line objects, others were suspected as pre-main-sequence stars from photometry in the J, H, Ks and Vilnius systems. We confirm the presence of the H alpha line emission in the spectra of 19 stars, some of them exhibit also emission in the O I and Ca II lines. In some of the stars the H alpha absorption line is filled with emission. To estimate their evolutionary status, the spectral energy distributions, based on Vilnius, 2MASS, MSX and Spitzer photometry, are applied. Only eight emission-line stars are found to be located at a distance of the NAP complex. Others are either chromospherically active stars in front of the complex or distant luminous stars with H alpha absorption and emission components. For five stars with faint emission the data are not sufficient to estimate their distance. One star is found to be a heavily reddened K-supergiant located in the Outer arm. The stars, for which we failed to confirm the emission in H alpha, are mostly red dwarfs located in fr
Scientists collaborate through intricate networks, which impact the quality and scope of their research. At the same time, funding and institutional arrangements, as well as scientific and political cultures, affect the structure of collaboration networks. Since such arrangements and cultures differ across regions in the world in systematic ways, we surmise that collaboration networks and impact should also differ systematically across regions. To test this, we compare the structure of collaboration networks among prominent researchers in North America and Europe. We find that prominent researchers in Europe establish denser collaboration networks, whereas those in North-America establish more decentralized networks. We also find that the impact of the publications of prominent researchers in North America is significantly higher than for those in Europe, both when they collaborate with other prominent researchers and when they do not. Although Europeans collaborate with other prominent researchers more often, which increases their impact, we also find that repeated collaboration among prominent researchers decreases the synergistic effect of collaborating.
Despite high performance on clinical benchmarks, large language models may reach correct conclusions through faulty reasoning, a failure mode with safety implications for oncology decision support that is not captured by accuracy-based evaluation. In this two-cohort retrospective study, we developed a hierarchical taxonomy of reasoning errors from GPT-4 chain-of-thought responses to real oncology notes and tested its clinical relevance. Using breast and pancreatic cancer notes from the CORAL dataset, we annotated 600 reasoning traces to define a three-tier taxonomy mapping computational failures to cognitive bias frameworks. We validated the taxonomy on 822 responses from prostate cancer consult notes spanning localized through metastatic disease, simulating extraction, analysis, and clinical recommendation tasks. Reasoning errors occurred in 23 percent of interpretations and dominated overall errors, with confirmation bias and anchoring bias most common. Reasoning failures were associated with guideline-discordant and potentially harmful recommendations, particularly in advanced disease management. Automated evaluators using state-of-the-art language models detected error presence
Magnitudes and color indices in the Vilnius seven-color system are measured for 690 stars down to ~13.2 mag in the area of the North America and Pelican nebulae. Spectral types, absolute magnitudes, color excesses, interstellar extinctions and distances of the stars are determined. The plots of interstellar extinction Av versus distance for the North America Nebula and for the dark cloud L935 show that both areas are covered by the same absorbing cloud, situated at a distance of 600 pc. The maximal extinction in the area of the nebula is ~3 mag, while in the dark cloud L935 it is much greater.
A possibility of applying 2MASS J, H, Ks, IPHAS r, i and MegaCam u, g photometry of red giants for determining distances to dark clouds is investigated. Red clump giants with a small admixture of G5-K1 and M2-M3 stars of the giant branch can be isolated and used in determining distances to separate clouds or spiral arms. The method is applied to an area of the North America and Pelican nebulae complex. Interstellar extinctions of background red giants can be also used for mapping dust surface density in the cloud.
Surgical procedures are often not "standardised" (i.e., defined in a unique and unambiguous way), but rather exist as implicit knowledge in the minds of the surgeon and the surgical team. This reliance extends to pre-surgery planning and effective communication during the procedure. We introduce a novel approach for the formal and automated analysis of surgical procedures, which we model as security ceremonies, leveraging well-established techniques developed for the analysis of such ceremonies. Mutations of a procedure are used to model variants and mistakes that members of the surgical team might make. Our approach allows us to automatically identify violations of the intended properties of a surgical procedure.
Despite the availability of computer-aided simulators and recorded videos of surgical procedures, junior residents still heavily rely on experts to answer their queries. However, expert surgeons are often overloaded with clinical and academic workloads and limit their time in answering. For this purpose, we develop a surgical question-answering system to facilitate robot-assisted surgical scene and activity understanding from recorded videos. Most of the existing VQA methods require an object detector and regions based feature extractor to extract visual features and fuse them with the embedded text of the question for answer generation. However, (1) surgical object detection model is scarce due to smaller datasets and lack of bounding box annotation; (2) current fusion strategy of heterogeneous modalities like text and image is naive; (3) the localized answering is missing, which is crucial in complex surgical scenarios. In this paper, we propose Visual Question Localized-Answering in Robotic Surgery (Surgical-VQLA) to localize the specific surgical area during the answer prediction. To deal with the fusion of the heterogeneous modalities, we design gated vision-language embedding
Every year approximately 234 million major surgeries are performed, leading to plentiful, highly diverse data. This is accompanied by a matching number of novel algorithms for the surgical domain. To garner all benefits of surgical data science it is necessary to have an unambiguous, shared understanding of algorithms and data. This includes inputs and outputs of algorithms and thus their function, but also the semantic content, i.e. meaning of data such as patient parameters. We therefore propose the establishment of a new ontology for data and algorithms in surgical data science. Such an ontology can be used to provide common data sets for the community, encouraging sharing of knowledge and comparison of algorithms on common data. We hold that this is a necessary foundation towards new methods for applications such as semantic-based content retrieval and similarity measures and that it is overall vital for the future of surgical data science.
The prevailing conceptual model for the production of severe local storm (SLS) environments over North America asserts that upstream elevated terrain and the Gulf of Mexico are both essential to their formation. This work tests this hypothesis using two prescribed-ocean climate model experiments with North American topography removed or the Gulf of Mexico converted to land and analyzes how SLS environments and associated synoptic-scale drivers (southerly Great Plains low-level jets, drylines, elevated mixed layers, and extratropical cyclones) change relative to a control historical run. Overall, SLS environments depend strongly on upstream elevated terrain but weakly on the Gulf of Mexico. Removing elevated terrain substantially reduces SLS environments especially over the continental interior due to broad reductions in both thermodynamic and kinematic parameters, leaving a more zonally-uniform residual distribution that is maximized near the Gulf coast and decays toward the continental interior. This response is associated with a strong reduction in synoptic-scale drivers and a cooler and drier mean-state atmosphere. Replacing the Gulf of Mexico with land modestly reduces SLS envi
Surgical skill assessment is important for surgery training and quality control. Prior works on this task largely focus on basic surgical tasks such as suturing and knot tying performed in simulation settings. In contrast, surgical skill assessment is studied in this paper on a real clinical dataset, which consists of fifty-seven in-vivo laparoscopic surgeries and corresponding skill scores annotated by six surgeons. From analyses on this dataset, the clearness of operating field (COF) is identified as a good proxy for overall surgical skills, given its strong correlation with overall skills and high inter-annotator consistency. Then an objective and automated framework based on neural network is proposed to predict surgical skills through the proxy of COF. The neural network is jointly trained with a supervised regression loss and an unsupervised rank loss. In experiments, the proposed method achieves 0.55 Spearman's correlation with the ground truth of overall technical skill, which is even comparable with the human performance of junior surgeons.