RESUMO
A simplified correlation index is proposed to be used in real-time pulse shape recognition systems. This index is similar to the classic Pearson's correlation coefficient, but it can be efficiently implemented in FPGA devices with far fewer logic resources and excellent performance. Numerical simulations with synthetic data and comparisons with the Pearson's correlation show the suitability of the proposed index in applications such as the discrimination and counting of pulses with a predefined shape. Superior performance is evident in signal-to-noise ratio scenarios close to unity. FPGA implementation of Person's method and the proposed correlation index have been successfully tested and the main results are summarized.
Assuntos
Algoritmos , Processamento de Sinais Assistido por Computador , Humanos , Razão Sinal-Ruído , Sistemas Computacionais , Reconhecimento PsicológicoRESUMO
In this study, we present a procedure to optimize a set of finite impulse response filter (FIR) coefficients for digital pulse-amplitude measurement. Such an optimized filter is designed using an adapted digital penalized least mean square (DPLMS) method. The effectiveness of the procedure is demonstrated using a dataset from a case study on high-resolution X-ray spectroscopy based on single-photon detection and energy measurements. The energy resolutions of the Kα and Kß lines of the Manganese energy spectrum have been improved by approximately 20%, compared to the reference values obtained by fitting individual photon pulses with the corresponding mathematical model.
Assuntos
Análise de Dados , Processamento de Sinais Assistido por Computador , Modelos Teóricos , Análise Espectral , Raios XRESUMO
In hard X-ray applications that require high detection efficiency and short response times, such as synchrotron radiation-based Mössbauer absorption spectroscopy and time-resolved fluorescence or photon beam position monitoring, III-V-compound semiconductors, and dedicated alloys offer some advantages over the Si-based technologies traditionally used in solid-state photodetectors. Amongst them, gallium arsenide (GaAs) is one of the most valuable materials thanks to its unique characteristics. At the same time, implementing charge-multiplication mechanisms within the sensor may become of critical importance in cases where the photogenerated signal needs an intrinsic amplification before being acquired by the front-end electronics, such as in the case of a very weak photon flux or when single-photon detection is required. Some GaAs-based avalanche photodiodes (APDs) were grown by a molecular beam epitaxy to fulfill these needs; by means of band gap engineering, we realised devices with separate absorption and multiplication region(s) (SAM), the latter featuring a so-called staircase structure to reduce the multiplication noise. This work reports on the experimental characterisations of gain, noise, and charge collection efficiencies of three series of GaAs APDs featuring different thicknesses of the absorption regions. These devices have been developed to investigate the role of such thicknesses and the presence of traps or defects at the metal-semiconductor interfaces responsible for charge loss, in order to lay the groundwork for the future development of very thick GaAs devices (thicker than 100 µm) for hard X-rays. Several measurements were carried out on such devices with both lasers and synchrotron light sources, inducing photon absorption with X-ray microbeams at variable and controlled depths. In this way, we verified both the role of the thickness of the absorption region in the collection efficiency and the possibility of using the APDs without reaching the punch-through voltage, thus preventing the noise induced by charge multiplication in the absorption region. These devices, with thicknesses suitable for soft X-ray detection, have also shown good characteristics in terms of internal amplification and reduction of multiplication noise, in line with numerical simulations.
RESUMO
Face recognition functions are today exploited through biometric sensors in many applications, from extended security systems to inclusion devices; deep neural network methods are reaching in this field stunning performances. The main limitation of the deep learning approach is an inconvenient relation between the accuracy of the results and the needed computing power. When a personal device is employed, in particular, many algorithms require a cloud computing approach to achieve the expected performances; other algorithms adopt models that are simple by design. A third viable option consists of model (oracle) distillation. This is the most intriguing among the compression techniques since it permits to devise of the minimal structure that will enforce the same I/O relation as the original model. In this paper, a distillation technique is applied to a complex model, enabling the introduction of fast state-of-the-art recognition capabilities on a low-end hardware face recognition sensor module. Two distilled models are presented in this contribution: the former can be directly used in place of the original oracle, while the latter incarnates better the end-to-end approach, removing the need for a separate alignment procedure. The presented biometric systems are examined on the two problems of face verification and face recognition in an open set by using well-agreed training/testing methodologies and datasets.
Assuntos
Face/fisiologia , Reconhecimento Facial/fisiologia , Algoritmos , Identificação Biométrica/métodos , Biometria/métodos , Confidencialidade , Bases de Dados Factuais , Humanos , Aprendizado de Máquina , Redes Neurais de ComputaçãoRESUMO
The dynamic response of gas sensors based on poly(3-hexylthiophene) (P3HT) nanofibers(NFs) to gaseous acetone was assessed using a setup based on flow-injection analysis, aimed atemulating actual breath exhalation. The setup was validated by using a commercially available sensor.The P3HT NFs sensors tested in dynamic flow conditions showed satisfactory reproducibility down toabout 3.5 ppm acetone concentration, a linear response over a clinically relevant concentration range(3.5-35 ppm), excellent baseline recovery and reversibility upon repeated exposures to the analyte,short pulse rise and fall times (less than 1 s and about 2 s, respectively) and low power consumption(few nW), with no relevant response to water. Comparable responses' decay times under eithernitrogen or dry air suggest that the mechanisms at work is mainly attributable to specific analytesemiconductingpolymer interactions. These results open the way to the use of P3HT NFs-basedsensing elements for the realization of portable, real-time electronic noses for on-the-fly exhaledbreath analysis.
RESUMO
Computational techniques allow breaking the limits of traditional imaging methods, such as time restrictions, resolution, and optics flaws. While simple computational methods can be enough for highly controlled microscope setups or just for previews, an increased level of complexity is instead required for advanced setups, acquisition modalities or where uncertainty is high; the need for complex computational methods clashes with rapid design and execution. In all these cases, Automatic Differentiation, one of the subtopics of Artificial Intelligence, may offer a functional solution, but only if a GPU implementation is available. In this paper, we show how a framework built to solve just one optimisation problem can be employed for many different X-ray imaging inverse problems.
RESUMO
Computational methods are driving high impact microscopy techniques such as ptychography. However, the design and implementation of new algorithms is often a laborious process, as many parts of the code are written in close-to-the-hardware programming constructs to speed up the reconstruction. In this article, we present SciComPty, a new ptychography software framework aiming at simulating ptychography datasets and testing state-of-the-art and new reconstruction algorithms. Despite its simplicity, the software leverages GPU accelerated processing through the PyTorch CUDA interface. This is essential for designing new methods that can readily be employed. As an example, we present an improved position refinement method based on Adam and a new version of the rPIE algorithm, adapted for partial coherence setups. Results are shown on both synthetic and real datasets. The software is released as open-source.
RESUMO
We report here an experimental setup to perform three-pulse pump-probe measurements over a wide wavelength and temperature range. By combining two pump pulses in the visible (650 nm-900 nm) and mid-IR (5 µm-20 µm) range, with a broadband supercontinuum white-light probe, our apparatus enables both the combined selective excitation of different material degrees of freedom and a full time-dependent reconstruction of the non-equilibrium dielectric function of the sample. We describe here the optical setup, the cryogenic sample environment, and the custom-made acquisition electronics capable of referenced single-pulse detection of broadband spectra at the maximum repetition rate of 50 kHz, achieving a sensitivity of the order of 10-4 over an integration time of 1 s. We demonstrate the performance of the setup by reporting data on a mid-IR pump, optical push, and broadband probe in a single crystal of Bi2Sr2Y0.08Ca0.92Cu2O8+δ across the superconducting and pseudogap phases.
RESUMO
Micro and nanomechanical resonators represent a promising platform for proteins label-free detection because of their extreme sensitivity, fast response and low cost. Micro-pillars are columnar resonators that can be easily arranged in dense arrays of several thousand sensors in a squared mm. To exploit such a large density, however, a method for tracking independently micropillars resonance frequency is required. Here we present a detection method based on CCD imaging and software image analysis, which can measure the resonance frequency of tens of pillars in parallel. Acquiring simultaneously the frequency shift of up to 40 sensors and applying a proper statistical analysis, we were able to overcome the variability of the single measures improving the device sensitivity at low analyte concentration range. As a proof of concept, this method has been tested for the detection of a tumor marker, the Prostate Specific Membrane Antigen (PSMA). Pillars have been functionalized with an antibody against PSMA. The tumor marker (PSMA) has been detected in a range of concentrations between 300 pM and 100 nM, in buffer and in diluted bovine serum. The sensitivity of our method was limited only by the affinity constant of the antigen-antibody recognition. Moreover, this detection technique demonstrated to be effective in the 1-6 nM range, which is the window of PSMA concentration of clinical interest.
Assuntos
Técnicas Biossensoriais/instrumentação , Sistemas Microeletromecânicos/instrumentação , Antígeno Prostático Específico/sangue , Anticorpos Imobilizados/química , Biomarcadores Tumorais/sangue , Desenho de Equipamento , Humanos , Limite de Detecção , Masculino , Próstata/patologia , Neoplasias da Próstata/sangue , Neoplasias da Próstata/diagnósticoRESUMO
Regular moderate-intensity exercise is strongly recommended for its beneficial effects in all people. In patients with type 1 diabetes, however, the exercise-associated glycemic imbalances remain an unresolved clinical challenge. Current guidelines require an in-depth understanding of the glycemic responses to exercise and each patient has to discover, by trial-and-error, his/her own strategy, several attempts being usually required to gain sufficient experience. Consequently, fear of hypoglycemia remains the strongest barrier to physical activity. This paper explores the potential strategies that may be employed to minimize the risk of exercise related glycemic imbalances. Moreover, a newly developed algorithm (ECRES, Exercise Carbohydrate Requirement Estimating Software) is described, which estimates on a patient- and situation-specific basis the glucose supplement required by the patient to maintain safe blood glucose levels. The algorithm was tested on 27 patients who performed three 1-hr constant intensity walks (each starting at a different time interval following insulin injection). Results showed that in 70.4% of the trials, independent of the time of day, the algorithm provided a satisfactory estimate of the carbohydrates needed by patients to complete the exercise with a glucose level within safe thresholds (i.e. 3.9 - 10 mmol·L-1). Despite the algorithm requires further experimental testing to be applied by the majority of patients, these results indicate its potential usefulness as a tool for preventing immediate exercise-induced glycemic imbalances (i.e. during exercise) in type 1 diabetic patients, in particular for spontaneous activities not planned in advance, thus allowing all insulin-dependent patients to safely enjoy the benefits of exercise.