First Full-Event Reconstruction from Imaging Atmospheric Cherenkov Telescope Real Data with Deep Learning

The Cherenkov Telescope Array (CTA) will consist of two arrays of Imaging Atmospheric Cherenkov Telescopes (IACTs) at the northern and southern hemispheres. CTA will feature IACTs with mirrors of three different sizes optimized to cover different energy ranges. The pro- posed sub-arrays of four Large Size Telescopes (LST) at CTA-North and CTA-South target the lowest energy range between around 20 GeV and 100 GeV. Thanks to their low weight of around 110 tons the LSTs can move by 180 deg in azimuth in 20 seconds for Gamma Ray Burst (GRB) follow-up. An LST has a tessellated parabolic mirror of 23 m diameter equipped with a system of actuators to correct for gravity-induced deformations during data taking. Its low-weight 2 ton camera at the prime focus has a 4.5 deg diameter, 1855 high QE PMTs and an embedded readout with 1 GSps sampling speed designed for data acquisition rates exceeding 10 kHz. A fully equipped LST has been installed at the CTA-North site in 2018 and is expected to be finished commissioning during 2019. The remaining three LSTs in the north will be installed by 2022. We will review the status of the LSTs, describe the installation of the first LST and report on the first results of the commissioning tests.

Riccardo Rando∗ab, Daniele Cortia, Francesco Dazzic, Alessandro De Angelisd, Antonios Dettlaffc, Daniela Dornere, David Finkc, Nadia Fouque f , Felix Grundnerc, Werner Habererc, Alexander Hahnc, Richard Hermel f , Samo Korparg, Gasper Kukec Mezekh, Ronald Maierc, Christian Maneaa, Mose Mariottiab, Daniel Mazinci, Fatima Mehrez f , Razmik Mirzoyanc, Sergey Podkladkinc, Ignasi Reichardta, Wolfgang Rhode j, Sylvie Rosier f , Cornelia Schultzab, Carlo Stellak, Masahiro Teshimaci, Holger Wetteskindc and Marko Zavrtanikgh for the CTA Consortium† E-mail: riccardo.rando@pd.infn.it a INFN Section of Padova, Italy; b University of Padova, Italy; c Max Planck Institute for Physics, Munich, Germany; d LIP/IST Lisboa, Portugal; e University of Wurzburg, Germany; f LAPP, CNRS-IN2P3, Annecy, France; g Jožef Stefan Institute, Ljubljana, Slovenia; h University of Nova Gorica, Slovenia; i University of Tokyo and ICRR, Tokyo, Japan; j Technische Universitat, Dortmund, Germany; k University of Udine and INFN, Italy. The Cherenkov Telescope Array (CTA) is the the next generation facility of imaging atmospheric Cherenkov telescopes; two sites will cover both hemispheres. CTA will reach unprecedented sensitivity, energy and angular resolution in very-high-energy gamma-ray astronomy. Each CTA array will include four Large Size Telescopes (LSTs), designed to cover the low-energy range of the CTA sensitivity (∼20 GeV to 200 GeV). In the baseline LST design, the focal-plane camera will be instrumented with 265 photodetector clusters; each will include seven photomultiplier tubes (PMTs), with an entrance window of 1.5 inches in diameter. The PMT design is based on mature and reliable technology. Recently, silicon photomultipliers (SiPMs) are emerging as a competitor. Currently, SiPMs have advantages (e.g. lower operating voltage and tolerance to high illumination levels) and disadvantages (e.g. higher capacitance and cross talk rates), but this technology is still young and rapidly evolving. SiPM technology has a strong potential to become superior to the PMT one in terms of photon detection efficiency and price per square mm of detector area. While the advantage of SiPMs has been proven for high-density, small size cameras, it is yet to be demonstrated for large area cameras such as the one of the LST. We are working to develop a SiPM-based module for the LST camera, in view of a possible camera upgrade. We will describe the solutions we are exploring in order to balance a competitive performance with a minimal impact on the overall LST camera design. The 34th International Cosmic Ray Conference, 30 July6 August, 2015 The Hague, The Netherlands

When very-high-energy gamma rays interact high in the Earth’s atmosphere, they produce cascades of particles that induce flashes of Cherenkov light. Imaging Atmospheric Cherenkov Telescopes (IACTs) detect these flashes and convert them into shower images that can be analyzed to extract the properties of the primary gamma ray. The dominant background for IACTs is comprised of air shower images produced by cosmic hadrons, with typical noise-to-signal ratios of several orders of magnitude. The standard technique adopted to differentiate between images initiated by gamma rays and those initiated by hadrons is based on classical machine learning algorithms, such as Random Forests, that operate on a set of handcrafted parameters extracted from the images. Likewise, the inference of the energy and the arrival direction of the primary gamma ray is performed using those parameters. State-of-the-art deep learning techniques based on convolutional neural networks (CNNs) have the potential to enhance the event reconstruction performance, since they are able to autonomously extract features from raw images, exploiting the pixel-wise information washed out during the parametrization process. Here we present the results obtained by applying deep learning techniques to the reconstruction of Monte Carlo simulated events from a single, next-generation IACT, the Large-Sized Telescope (LST) of the Cherenkov Telescope Array (CTA). We use CNNs to separate the gamma-ray-induced events from hadronic events and to reconstruct the properties of the former, comparing their performance to the standard reconstruction technique. Three independent implementations of CNN-based event reconstruction models have been utilized in this work, producing consistent results.

CTA (Cherenkov Telescope Array) is a project for the next generation of the ground-based gamma-ray observatory. CTA will cover a wide energy range, 20 GeV and 300 TeV, by 3 types of the telescopes whose diameters are different. The Large-Sized Telescopes (LSTs) of the CTA are designed to cover the lowest energy range. We are working on mirrors for 4 LSTs which will be built at La Palma, Spain. The parabolic primary mirror of CTA-LST is 23 m diameter and its focal length is 28 m. The primary mirror consists of 198 segmented mirrors. Total effective reflective area is about 370 m2. Each mirror has a hexagonal shape of 1.51 m side-by-side size. It has a sandwich structure which consists of aluminum honeycomb (60 mm thickness) and two glass surfaces (2.7 mm) and the total weight is about 47 kg. We used a sputtering deposition technique to coat in the surface with 5 layer. The coated layers protect the surface for long operation such as 10 years at outside with a reflectance degradation of less than 1 % a year. The reflectance of the mirror reaches about 92 % at 400 nm and its resolution is 0.5 mrad in diameter that contains the 80 % light reflected by the mirror. The production technique based on cold slumping gave us the stable its production. We report on the design of the segment mirror and the production of about 950 hexagonal 2 m2 mirrors to achieve the parabolic shape of the optical system for 4 LSTs.

The Cherenkov Telescope Array (CTA) is the next-generation gamma-ray observatory that is expected to reach one order of magnitude better sensitivity than that of current telescope arrays. The Large-Sized Telescopes (LSTs) have an essential role in extending the energy range down to 20 GeV. The prototype LST (LST-1) proposed for CTA was built in La Palma, the northern site of CTA, in 2018. LST-1 is currently in its commissioning phase and moving towards scientific observations. The LST-1 camera consists of 1855 photomultiplier tubes (PMTs) which are sensitive to Cherenkov light. PMT signals are recorded as waveforms sampled at 1 GHz rate with Domino Ring Sampler version 4 (DRS4) chips. Fast sampling is essential to achieve a low energy threshold by minimizing the integration of background light from the night sky. Absolute charge calibration can be performed by the so-called F-factor method, which allows calibration constants to be monitored even during observations. A calibration pipeline of the camera readout has been developed as part of the LST analysis chain. The pipeline performs DRS4 pedestal and timing corrections, as well as the extraction and calibration of charge and time of pulses for subsequent higher-level analysis. The performance of each calibration step is examined, and especially charge and time resolution of the camera readout are evaluated and compared to CTA requirements. We report on the current status of the calibration pipeline, including the performance of each step through to signal reconstruction, and the consistency with Monte Carlo simulations.

Deep learning (also known as deep structured learning, hierarchical learning, or deep machine learning) is a branch of machine learning based on a set of algorithms that attempt to model high-level abstractions in data by using a deep graph with multiple processing layers, composed of multiple linear and nonlinear transformations. Deep learning has been characterized as a class of machine learning algorithms with the following characteristics [257]: • They use a cascade of many layers of nonlinear processing units for feature extraction and transformation. Each successive layer uses the output from the previous layer as input. The algorithms may be supervised or unsupervised and applications include pattern analysis (unsupervised) and classification (supervised). • They are based on the (unsupervised) learning of multiple levels of features or representations of the data. Higher-level features are derived from lower-level features to form a hierarchical representation. • They are part of the broader machine learning field of learning representations of data. • They learn multiple levels of representations that correspond to different levels of abstraction; the levels form a hierarchy of concepts. These definitions have in common: multiple layers of nonlinear processing units and the supervised or unsupervised learning of feature representations in each layer, with the layers forming a hierarchy from low-level to high-level features. Various deep learning architectures such as deep neural networks, convolutional deep neural networks, deep belief networks (DBN), and recurrent neural networks have been applied to fields like computer vision, automatic speech recognition, natural language processing, audio recognition, and bioinformatics where they have been shown to produce state-of-the-art results on various tasks. In this chapter we start in Section 7.1 with an introduction, giving a brief history of this field, the relevant literature, and its applications. Then we study some basic concepts of deep learning such as convolutional neural networks, recurrent neural networks, backpropagation algorithm, restricted Boltzmann machines, and deep learning networks in Section 7.2. Then we illustrate three examples for Apache Spark implementation for mobile big data (MBD), user moving pattern extraction, and combination with nonparametric Bayesian learning, respectively, in Sections 7.3 through 7.5. Finally, we have summary in Section 7.6.

The Large-Sized Telescopes (LSTs) of the Cherenkov Telescope Array (CTA) are designed for gamma-ray studies focusing on low energy threshold, high flux sensitivity and rapid telescope repositioning. The LST has a tessellated parabolic mirror of 23 m diameter and a weight of about 100 tons, with the capability of pointing to any position in the sky in 20 seconds or less to catch transients. A 2 ton 3x3 m camera, placed in the focus of the mirror, is equipped with 1855 high QE PMTs corresponding to a FoV of about 4.5 degrees. LSTs will dominate the CTA performance between 20 GeV and approximately 200 GeV. The first LST (LST-1) was inaugurated in La Palma (Spain) in October 2018 and since then it is in the commissioning phase. The current status and performance of the LST-1 are discussed. The telescope behavior is well understood and the performance has reached the design level in terms of sensitivity, angular and energy resolutions. Still some work is needed to achieve the duty cycle goal. We briefly present here some of the first physics results. Finally the outlooks of the project are presented.

The Cherenkov Telescope Array (CTA) consortium aims to create the next generation Very High Energy gamma-ray observatory. It will be devoted to the observation of gamma rays over a wide band of energy, from 20 GeV to 300 TeV. Three different classes, Large, Medium and Small Size Telescopes, are foreseen to cover the low, intermediate and high energy regions, respectively. The energy range of the Small Size Telescopes (SSTs) extends from 1 TeV to 300 TeV. Among them, the Gamma-ray Cherenkov Telescope (GCT), a telescope based on a Schwarzschild-Couder dual-mirror optical design, is one of the prototypes under construction proposed for the SST sub-array of CTA. This contribution focuses on the mechanical structure of GCT. It reports on last progress on the mechanical design and discusses this in the context of CTA specifications. Recent advances in the assembly and installation of the opto-mechanical prototype of GCT on the French site of the Paris Observatory are also described.

The Cherenkov Telescope Array (CTA) is the future large observatory in the very high energy (VHE) domain. Operating from 20 GeV to 300 TeV, it will be composed of tens of Imaging Air Cherenkov Telescopes (IACTs) displaced in a large area of a few square kilometers in both the southern and northern hemispheres. The CTA/DATA On-Site Analysis (OSA) is the system devoted to the development of dedicated pipelines and algorithms to be used at the CTA site for the reconstruction, data quality monitoring, science monitoring and realtime science alerting during observations. The OSA integral sensitivity is computed here for the most studied source at Gamma-rays, the Crab Nebula, for a set of exposures ranging from 1000 seconds to 50 hours, using the full CTA Southern array. The reason for the Crab Nebula selection as the first example of OSA integral sensitivity is twofold: (i) this source is characterized by a broad spectrum covering the entire CTA energy range; (ii) it represents, at the time of writing, the standard candle in VHE and it is often used as unit for the IACTs sensitivity. The effect of different Crab Nebula emission models on the CTA integral sensitivity is evaluated, to emphasize the need for representative spectra of the CTA science targets in the evaluation of the OSA use cases. Using the most complete model as input to the OSA integral sensitivity, we obtain a significant detection of the Crab nebula (about 10% of flux) even for a 1000 second exposure, for an energy threshold less than 10 TeV.

PurposeThe aim of this study was to analyse and review deep learning convolutional neural networks for detecting and diagnosing early-stage dental caries on periapical radiographs.Materials and MethodsIn order to conduct this review, the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were followed. Studies published from 2015 to 2021 under the keywords (deep convolutional neural network) AND (caries), (deep learning caries) AND (convolutional neural network) AND (caries) were systematically reviewed.ResultsWhen dental caries is improperly diagnosed, the lesion may eventually invade the enamel, dentin, and pulp tissue, leading to loss of tooth function. Rapid and precise detection and diagnosis are vital for implementing appropriate prevention and treatment of dental caries. Radiography and intraoral images are considered to play a vital role in detecting dental caries; nevertheless, studies have shown that 20% of suspicious areas are mistakenly diagnosed as dental caries using this technique; hence, diagnosis via radiography alone without an objective assessment is inaccurate. Identifying caries with a deep convolutional neural network-based detector enables the operator to distinguish changes in the location and morphological features of dental caries lesions. Deep learning algorithms have broader and more profound layers and are continually being developed, remarkably enhancing their precision in detecting and segmenting objects.ConclusionClinical applications of deep learning convolutional neural networks in the dental field have shown significant accuracy in detecting and diagnosing dental caries, and these models hold promise in supporting dental practitioners to improve patient outcomes.

We provide an updated assessment of the power of the Cherenkov Telescope Array (CTA) to search for thermally produced dark matter at the TeV scale, via the associated gamma-ray signal from pair-annihilating dark matter particles in the region around the Galactic centre. We find that CTA will open a new window of discovery potential, significantly extending the range of robustly testable models given a standard cuspy profile of the dark matter density distribution. Importantly, even for a cored profile, the projected sensitivity of CTA will be sufficient to probe various well-motivated models of thermally produced dark matter at the TeV scale. This is due to CTA's unprecedented sensitivity, angular and energy resolutions, and the planned observational strategy. The survey of the inner Galaxy will cover a much larger region than corresponding previous observational campaigns with imaging atmospheric Cherenkov telescopes. CTA will map with unprecedented precision the large-scale diffuse emission in high-energy gamma rays, constituting a background for dark matter searches for which we adopt state-of-the-art models based on current data. Throughout our analysis, we use up-to-date event reconstruction Monte Carlo tools developed by the CTA consortium, and pay special attention to quantifying the level of instrumental systematic uncertainties, as well as background template systematic errors, required to probe thermally produced dark matter at these energies.

The Cherenkov Telescope Array (CTA) observatory will probe the non-thermal universe above 20 GeV up to several hundreds of TeV with a significant improvement in sensitivity and angular resolution compared to current experiments. Its outstanding capabilities will allow to increase the number of extragalactic cosmic accelerators detected at very high energy (VHE) and therefore to better constrain the population of VHE accelerators and the gamma-ray absorption processes in the intergalactic medium. For the first time in the history of imaging atmospheric Cherenkov telescopes (IACTs), CTA will be an open observatory and high-level data will be made available to the astronomical community. Gammapy is an open-source Python package developed by the Cherenkov telescope community that provides tools to simulate the gamma-ray sky and analyse IACT data. The versatile architecture of, and steady user contributions to Gammapy enable a large variety of high-level data analyses. Examples of Gaammapy applications are presented, particularly in the context of extragalactic science with CTA.

Estimation of the height of the first interaction in gamma-ray showers observed by Cherenkov telescopes

Stereoscopic arrays of Imaging Cherenkov Telescopes allow to reconstruct gamma‐ray‐induced showers in 3 dimensions. An analysis method based on a simple 3D‐model of electromagnetic showers and implemented in the framework of the H.E.S.S. experiment was recently improved by an additional quality criterion which reduces the background contamination by a factor of about 2 in the case of extended sources, while hardly affecting gamma‐ray selection efficiency. Moreover, the dramatic flares of PKS 2155‐304 in July 2006, which provided H.E.S.S. data with an almost pure gamma‐ray sample, offered the unique opportunity of a precision test of the 3D‐reconstruction method as well as of the H.E.S.S. simulations used in its calibration. An agreement at a few percent level is found between data and simulations for the distributions of all 3D shower parameters.

Night Sky Background (NSB) is a complex phenomenon, consisting of all light detected by Imaging Atmospheric Cherenkov Telescopes (IACTs) not attributable to Cherenkov light emission. Understanding the effect of NSB on cameras for the next-generation Cherenkov Telescope Array (CTA) is important, as it affects the systematic errors on observations, the energy threshold, the thermal control of the cameras and the ability of the telescopes to operate under partial moonlight conditions. This capacity to observe under partial moonlight conditions is crucial for the CTA transient science programme, as it substantially increases the potential observing time. Using tools initially developed for H.E.S.S. (in combination with the prototype CTA analysis package ctapipe) we will present predictions for the NSB present in images taken by the CTA Small Sized Telescope Camera (SSTCAM), showing that SSTCAM will likely be able to meet the associated CTA requirements. Additionally, we calculate the potential observing time gain by operating under high NSB conditions.