|
|
Conde, D., Castillo, F. L., Escobar, C., García, C., Garcia Navarro, J. E., Sanz, V., et al. (2023). Forecasting Geomagnetic Storm Disturbances and Their Uncertainties Using Deep Learning. Space Weather, 21(11), e2023SW003474–27pp.
Abstract: Severe space weather produced by disturbed conditions on the Sun results in harmful effects both for humans in space and in high-latitude flights, and for technological systems such as spacecraft or communications. Also, geomagnetically induced currents (GICs) flowing on long ground-based conductors, such as power networks, potentially threaten critical infrastructures on Earth. The first step in developing an alarm system against GICs is to forecast them. This is a challenging task given the highly non-linear dependencies of the response of the magnetosphere to these perturbations. In the last few years, modern machine-learning models have shown to be very good at predicting magnetic activity indices. However, such complex models are on the one hand difficult to tune, and on the other hand they are known to bring along potentially large prediction uncertainties which are generally difficult to estimate. In this work we aim at predicting the SYM-H index characterizing geomagnetic storms multiple-hour ahead, using public interplanetary magnetic field (IMF) data from the Sun-Earth L1 Lagrange point and SYM-H data. We implement a type of machine-learning model called long short-term memory (LSTM) network. Our scope is to estimate the prediction uncertainties coming from a deep-learning model in the context of forecasting the SYM-H index. These uncertainties will be essential to set reliable alarm thresholds. The resulting uncertainties turn out to be sizable at the critical stages of the geomagnetic storms. Our methodology includes as well an efficient optimization of important hyper-parameters of the LSTM network and robustness tests.
|
|
|
|
Schaffter, T. et al, Albiol, F., & Caballero, L. (2020). Evaluation of Combined Artificial Intelligence and Radiologist Assessment to Interpret Screening Mammograms. JAMA Netw. Open, 3(3), e200265–15pp.
Abstract: Importance Mammography screening currently relies on subjective human interpretation. Artificial intelligence (AI) advances could be used to increase mammography screening accuracy by reducing missed cancers and false positives. Objective To evaluate whether AI can overcome human mammography interpretation limitations with a rigorous, unbiased evaluation of machine learning algorithms. Design, Setting, and Participants In this diagnostic accuracy study conducted between September 2016 and November 2017, an international, crowdsourced challenge was hosted to foster AI algorithm development focused on interpreting screening mammography. More than 1100 participants comprising 126 teams from 44 countries participated. Analysis began November 18, 2016. Main Outcomes and Measurements Algorithms used images alone (challenge 1) or combined images, previous examinations (if available), and clinical and demographic risk factor data (challenge 2) and output a score that translated to cancer yes/no within 12 months. Algorithm accuracy for breast cancer detection was evaluated using area under the curve and algorithm specificity compared with radiologists' specificity with radiologists' sensitivity set at 85.9% (United States) and 83.9% (Sweden). An ensemble method aggregating top-performing AI algorithms and radiologists' recall assessment was developed and evaluated. Results Overall, 144231 screening mammograms from 85580 US women (952 cancer positive <= 12 months from screening) were used for algorithm training and validation. A second independent validation cohort included 166578 examinations from 68008 Swedish women (780 cancer positive). The top-performing algorithm achieved an area under the curve of 0.858 (United States) and 0.903 (Sweden) and 66.2% (United States) and 81.2% (Sweden) specificity at the radiologists' sensitivity, lower than community-practice radiologists' specificity of 90.5% (United States) and 98.5% (Sweden). Combining top-performing algorithms and US radiologist assessments resulted in a higher area under the curve of 0.942 and achieved a significantly improved specificity (92.0%) at the same sensitivity. Conclusions and Relevance While no single AI algorithm outperformed radiologists, an ensemble of AI algorithms combined with radiologist assessment in a single-reader screening environment improved overall accuracy. This study underscores the potential of using machine learning methods for enhancing mammography screening interpretation. Question How do deep learning algorithms perform compared with radiologists in screening mammography interpretation? Findings In this diagnostic accuracy study using 144231 screening mammograms from 85580 women from the United States and 166578 screening mammograms from 68008 women from Sweden, no single artificial intelligence algorithm outperformed US community radiologist benchmarks; including clinical data and prior mammograms did not improve artificial intelligence performance. However, combining best-performing artificial intelligence algorithms with single-radiologist assessment demonstrated increased specificity. Meaning Integrating artificial intelligence to mammography interpretation in single-radiologist settings could yield significant performance improvements, with the potential to reduce health care system expenditures and address resource scarcity experienced in population-based screening programs. This diagnostic accuracy study evaluates whether artificial intelligence can overcome human mammography interpretation limits with a rigorous, unbiased evaluation of machine learning algorithms.
|
|
|
|
ATLAS Collaboration(Aaboud, M. et al), Alvarez Piqueras, D., Aparisi Pozo, J. A., Bailey, A. J., Barranco Navarro, L., Cabrera Urban, S., et al. (2018). Search for pair and single production of vectorlike quarks in final states with at least one Z boson decaying into a pair of electrons or muons in pp collision data collected with the ATLAS detector at root s=13 TeV. Phys. Rev. D, 98(11), 112010–53pp.
Abstract: A search for vectorlike quarks is presented, which targets their decay into a Z boson and a third-generation Standard Model quark. In the case of a vectorlike quark T (B) with charge +2/3e (-1/3e), the decay searched for is T -> Zt (B -> Zb). Data for this analysis were taken during 2015 and 2016 with the ATLAS detector at the Large Hadron Collider and correspond to an integrated luminosity of 36.1 fb(-1) of pp collisions at root s = 13 TeV. The final state used is characterized by the presence of b-tagged jets, as well as a Z boson with high transverse momentum, which is reconstructed from a pair of opposite-sign same-flavor leptons. Pair and single production of vectorlike quarks are both taken into account and are each searched for using optimized dileptonic exclusive and trileptonic inclusive event selections. In these selections, the high scalar sum of jet transverse momenta, the presence of high-transverse-momentum large-radius jets, as well as-in the case of the single-production selections-the presence of forward jets are used. No significant excess over the background-only hypothesis is found and exclusion limits at 95% confidence level allow masses of vectorlike quarks of m(T) > 1030 GeV (m(T) > 1210 GeV) and m(B) > 1010 GeV (m(B) > 1140 GeV) in the singlet (doublet) model. In the case of 100% branching ratio for T -> Zt (B -> Zb), the limits are m(T) > 1340 GeV (m(B) > 1220 GeV). Limits at 95% confidence level are also set on the coupling to Standard Model quarks for given vectorlike quark masses.
|
|
|
|
Carames, T. F., Fontoura, C. E., Krein, G., Vijande, J., & Valcarce, A. (2018). Charmed baryons in nuclear matter. Phys. Rev. D, 98(11), 114019–9pp.
Abstract: We study the temperature and baryon density dependence of the masses of the lightest charmed baryons Lambda(c), Sigma(c) and Sigma(c)*. We also look at the effects of the temperature and baryon density on the binding energies of the Lambda N-c and Lambda(c)Lambda(c) systems. Baryon masses and baryon-baryon interactions are evaluated within a chiral constituent quark model. Medium effects are incorporated in those parameters of the model related to the dynamical breaking of chiral symmetry, which are the masses of the constituent quarks, the sigma and pi meson masses, and quark-meson couplings. We find that while the in-medium Lambda(c) mass decreases monotonically with temperature, those of Sigma(c) and Sigma(c)* have a nonmonotonic dependence. These features can be understood in terms of a simple group theory analysis regarding the one-gluon exchange interaction in those hadrons. The in-medium Lambda N-c and Lambda(c)Lambda(c) interactions are governed by a delicate balance involving a stronger attraction due to the decrease of the sigma meson mass, suppression of coupled-channel effects and lower thresholds, leading to shallow bound states with binding energies of a few MeV. The Lambda(c) baryon could possibly be bound to a large nucleus, in qualitative agreement with results based on relativistic mean field models or QCD sum rules. Ongoing experiments at RHIC or LHCb or the planned ones at FAIR and J-PARC may take advantage of the present results.
|
|
|
|
BABAR and Belle Collaborations(Adachi, I. et al), Martinez-Vidal, F., & Oyanguren, A. (2018). Measurement of cos 2 beta in B-0 -> D((*))h(0) with D -> K-S(0)pi(+) pi(-) decays by a combined time-dependent Dalitz plot analysis of BABAR and Belle data. Phys. Rev. D, 98(11), 112012–29pp.
Abstract: We report measurements of sin 2 beta and cos 2 beta using a time-dependent Dalitz plot analysis of B-0 -> D((*))h(0) with D -> K-S(0)pi(+)pi(-) decays, where the light unflavored and neutral hadron h(0) is a pi(0),eta, or omega meson. The analysis uses a combination of the final data sets of the BABAR and Belle experiments containing 471 x 10(6) and 772 x 10(6) B (B) over bar pairs collected at the gamma(4S) resonance at the asymmetric-energy B factories PEP-II at SLAC and KEKB at KEK, respectively. We measure sin 2 beta = 0.80 +/- 0.14(stat) +/- 0.06(syst) +/- 0.03(model) and cos 2 beta = 0.91 +/- 0.22(stat) +/- 0.09(syst) +/- 0.07(model). The result for the direct measurement of the angle is beta = (22.5 +/- 4.4(stat) +/- 1.2(syst) +/- 0.6(model))degrees. The last quoted uncertainties are due to the composition of the D-0 -> K-S(0)pi(+)pi(-) decay amplitude model, which is newly established by a Dalitz plot amplitude analysis of a high-statistics e(+) e(-) -> c (c) over bar data sample as part of this analysis. We find the first evidence for cos 2 beta > 0 at the level of 3.7 standard deviations. The measurement excludes the trigonometric multifold solution pi/2 – beta = (68.1 +/- 0.7)degrees at the level of 7.3 standard deviations and therefore resolves an ambiguity in the determination of the apex of the CKM Unitarity Triangle. The hypothesis of beta = 0 degrees is ruled out at the level of 5.1 standard deviations, and thus CP violation is observed in B-0 -> D-(*) h(0) decays. The measurement assumes no direct CP violation in B-0 -> D-(*) h(0) decays.
|
|
