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.

Abstract: In a general framework, valid for any H -> H' l(-)(v) over bar (l) semileptonic decay, we analyze the d(2)Gamma/(d omega d cos theta(l)) and d(2)Gamma/(d omega dE(l)) distributions, with omega being the product of the hadron four-velocities, theta(l) the angle made by the three-momenta of the charged lepton and the final hadron in the W- center of mass frame and E-l the charged lepton energy in the decaying hadron rest frame. Within the Standard Model (SM), d(2)Gamma/(d omega dE(l)) proportional to (c(0) (omega) c(1) (omega)E-l/M + c(2) (omega)E-l(2)/M-2), with M the initial hadron mass. We find that c(2) (omega) is independent of the lepton flavor and thus it is an ideal candidate to look for lepton flavor universality (LFU) violations. We also find a correlation between the a(2) (omega) structure function, which governs the (cos theta(l))(2) dependence of d(2)Gamma/(d omega d cos theta(l)), and c(2) (omega). Apart from trivial kinematical and mass factors, the ratio of a(2) (omega)/c(2) (omega) is a universal function that can be measured in any semileptonic decay, involving not only b -> c transitions. These two SM predictions can be used as new tests in the present search for signatures of LFU violations. We also generalize the formalism to account for some new physics (NP) terms, and show that neither c(2) nor a(2) are modified by left and right scalar NP terms, being however sensitive to left and right vector corrections. We also find that the a(2)/c(2) ratio is not modified by these latter NP contributions. Finally, and in order to illustrate our findings, we apply our general framework to the Lambda(b) -> Lambda(c)l (v) over bar (l) decay. We show that a measurement of c(2) (or a(2)) for tau decay would not only be a direct measurement of the possible existence of NP, but it would also allow to distinguish from NP fits to b -> c tau(v) over bar (tau) anomalies in the meson sector, which otherwise give the same total and differential d Gamma/d omega widths. We show that the same occurs for the other two terms, c(0) and c(1), that appear in d(2)Gamma/(d omega dE(l)), and for the cos theta(l) linear term of the angular distribution.

Abstract: In a special run of the LHC with beta star=2.5 km, proton-proton elastic-scattering events were recorded at root s=13 TeV with an integrated luminosity of 340 μb(-1) using the ALFA subdetector of ATLAS in 2016. The elastic cross section was measured differentially in the Mandelstam t variable in the range from -t=2.5 center dot 10(-4) GeV2 to -t=0.46 GeV2 using 6.9 million elastic-scattering candidates. This paper presents measurements of the total cross section sigma(tot), parameters of the nuclear slope, and the rho-parameter defined as the ratio of the real part to the imaginary part of the elastic-scattering amplitude in the limit t -> 0. These parameters are determined from a fit to the differential elastic cross section using the optical theorem and different parameterizations of the t-dependence. The results for sigma(tot) and rho are sigma(tot) (PP -> X ) =104.7 +/- 1.1 mob , rho=0.098 +/- 0.011. The uncertainty in sigma(tot) is dominated by the luminosity measurement, and in rho by imperfect knowledge of the detector alignment and by modelling of the nuclear amplitude.

Abstract: In addition to vector (V) type new particles extensively discussed previously, both CP-even (S) and CP-odd (P) spin-0 dark matter (DM) mediators can couple to muons and be produced in the bremsstrahlung reaction mu- + N -mu- + N + S(P). Their possible subsequent invisible decay into a pair of Dirac DM particles, S(P) -chi chi over bar , can be detected in fixed target experiments through missing energy signature. In this paper, we focus on the case of experiments using high-energy muon beams. For this reason, we derive the differential cross sections involved using the phase space Weiszacker-Williams approximation and compare them to the exact-tree-level calculations. The formalism derived can be applied in various experiments that could observe muon-spin-0 DM interactions. This can happen in present and future proton beam-dump experiments such as NA62, SHIP, HIKE, and SHADOWS; in muon fixed target experiments as NA64 mu, MUonE and M3; in neutrino experiments using powerful proton beams such as DUNE. In particular, we focus on the NA64 μexperiment case, which uses a 160 GeV muon beam at the CERN Super Proton Synchrotron accelerator. We compute the derived cross sections, the resulting signal yields and we discuss the experiment projected sensitivity to probe the relic DM parameter space and the (g – 2)mu anomaly favored region considering 1011 and 1013 muons on target.

Abstract: In cancer patients undergoing proton therapy, a very intense secondary radiation is produced during the treatment, which lasts around one minute. About one billion prompt gamma-rays are emitted per second, and their detection with fast scintillation detectors is useful for monitoring a correct beam delivery. To cope with the expected count rate and pile-up, as well as the scarce statistics due to the short treatment duration, we developed an eidetic data acquisition system capable of continuously digitizing the detector signal with a high sampling rate and without any dead time. By streaming the fully unprocessed waveforms to the computer, complex pile-up decomposition algorithms can be applied and optimized offline. We describe the data acquisition architecture and the multiple experimental tests designed to verify the sustained data throughput speed and the absence of dead time. While the system is tailored for the proton therapy environment, the methodology can be deployed in any other field requiring the recording of raw waveforms at high sampling rates with zero dead time.