Intelligent systems are transforming the world, as well as our healthcare system. We propose a deep learning-based cough sound classification model that can distinguish between children with healthy ...versus pathological coughs such as asthma, upper respiratory tract infection (URTI), and lower respiratory tract infection (LRTI). To train a deep neural network model, we collected a new dataset of cough sounds, labelled with a clinician's diagnosis. The chosen model is a bidirectional long-short-term memory network (BiLSTM) based on Mel-Frequency Cepstral Coefficients (MFCCs) features. The resulting trained model when trained for classifying two classes of coughs-healthy or pathology (in general or belonging to a specific respiratory pathology)-reaches accuracy exceeding 84% when classifying the cough to the label provided by the physicians' diagnosis. To classify the subject's respiratory pathology condition, results of multiple cough epochs per subject were combined. The resulting prediction accuracy exceeds 91% for all three respiratory pathologies. However, when the model is trained to classify and discriminate among four classes of coughs, overall accuracy dropped: one class of pathological coughs is often misclassified as the other. However, if one considers the healthy cough classified as healthy and pathological cough classified to have some kind of pathology, then the overall accuracy of the four-class model is above 84%. A longitudinal study of MFCC feature space when comparing pathological and recovered coughs collected from the same subjects revealed the fact that pathological coughs, irrespective of the underlying conditions, occupy the same feature space making it harder to differentiate only using MFCC features.
In this study, we evaluated and compared the pulse shape discrimination (PSD) performance of a typical photomultiplier tube (PMT) with that of multipixel photon counters (MPPCs), also known as ...silicon photomultipliers (SiPMs). We used CsI(Tl) scintillators for both setups and mainly evaluated the PSD performance in the <inline-formula> <tex-math notation="LaTeX">\alpha </tex-math></inline-formula>-<inline-formula> <tex-math notation="LaTeX">\beta </tex-math></inline-formula> separation. We used the charge comparison method, that is, identifying the particle type based on the ratio of the charges collected in two time gates: Short gate (SG) for the delayed part and long gate (LG) for the entire digitized waveform. With the same LG = 2250 ns, the PMT setup showed a better PSD performance with the figure of merit (FOM) being 3.33 (in <inline-formula> <tex-math notation="LaTeX">\alpha </tex-math></inline-formula>-<inline-formula> <tex-math notation="LaTeX">\beta </tex-math></inline-formula> separation), whereas the FOM of the MPPC setup was 2.26. This inferior performance of the MPPC setup was due to its slow response, mainly the slow decay time in the waveform of one photoelectron (p.e.) of the MPPCs. This study aimed to determine the relationships of the PSD performance with the number of p.e. collected and the time characteristics of the scintillator-photosensor coupling. From these relationships, we estimated the SG to obtain <inline-formula> <tex-math notation="LaTeX">\text {FOM}_{\text {max}} </tex-math></inline-formula> and afterward confirmed a method to further increase the PSD performance. The estimated SG gave a comparable FOM without an extensive search for the optimized SG for obtaining the actual <inline-formula> <tex-math notation="LaTeX">\text {FOM}_{\text {max}} </tex-math></inline-formula> in analysis. To further increase the PSD performance of the MPPC setup, we used a new data acquisition (DAQ) system and extended the LG to 8000 ns to collect more p.e. and mitigate the slow response of the MPPCs. Using the new DAQ system, we obtained an FOM of 3.91 for the MPPC setup, that is, an increase by a factor of 1.7.
We have studied the neutrino-less double beta decay(0νββ) of 48Ca with the CANDLES III system, which consists of CaF2(pure) scintillators. Rejection analyses for background events from radioactive ...contaminations in the CaF2(pure) scintillators were effective to reduce backgrounds in Qββ-value region. As the results, no events were observed in the region for the data of 131 days × 86 kg. It gave a lower limit 6.2 × 1022 year (90 % C.L.) for the half-life of 0νββ of 48Ca. For higher sensitive measurement of 48Ca 0νββ, we have developed new techniques for 48Ca enrichment and CaF2 scintillating bolometer. In this paper, we will also show current status of these techniques.
(1) Background: Cough is a major presentation in childhood asthma. Here, we aim to develop a machine-learning based cough sound classifier for asthmatic and healthy children. (2) Methods: Children ...less than 16 years old were randomly recruited in a Children’s Hospital, from February 2017 to April 2018, and were divided into 2 cohorts—healthy children and children with acute asthma presenting with cough. Children with other concurrent respiratory conditions were excluded in the asthmatic cohort. Demographic data, duration of cough, and history of respiratory status were obtained. Children were instructed to produce voluntary cough sounds. These clinically labeled cough sounds were randomly divided into training and testing sets. Audio features such as Mel-Frequency Cepstral Coefficients and Constant-Q Cepstral Coefficients were extracted. Using a training set, a classification model was developed with Gaussian Mixture Model–Universal Background Model (GMM-UBM). Its predictive performance was tested using the test set against the physicians’ labels. (3) Results: Asthmatic cough sounds from 89 children (totaling 1192 cough sounds) and healthy coughs from 89 children (totaling 1140 cough sounds) were analyzed. The sensitivity and specificity of the audio-based classification model was 82.81% and 84.76%, respectively, when differentiating coughs from asthmatic children versus coughs from ‘healthy’ children. (4) Conclusion: Audio-based classification using machine learning is a potentially useful technique in assisting the differentiation of asthmatic cough sounds from healthy voluntary cough sounds in children.
In a neutrinoless double-beta decay (<inline-formula> <tex-math notation="LaTeX">0\nu \beta \beta </tex-math></inline-formula>) experiment, energy resolution is important to distinguish between ...<inline-formula> <tex-math notation="LaTeX">0\nu \beta \beta </tex-math></inline-formula> and background events. CAlcium fluoride for studies of Neutrino and Dark matters by Low Energy Spectrometer (CANDLES) discerns the <inline-formula> <tex-math notation="LaTeX">0\nu \beta \beta </tex-math></inline-formula> of 48 Ca using a CaF 2 scintillator as the detector and source. Photomultiplier tubes (PMTs) collect scintillation photons. At the <inline-formula> <tex-math notation="LaTeX">Q </tex-math></inline-formula> value of 48 Ca, the current energy resolution (2.6%) exceeds the ideal statistical fluctuation of the number of photoelectrons (1.6%). Because of CaF 2 's long decay constant of 1000 ns, a signal integration within 4000 ns is used to calculate the energy. The baseline fluctuation (<inline-formula> <tex-math notation="LaTeX">\sigma _{\mathrm{ baseline}} </tex-math></inline-formula>) is accumulated in the signal integration, thus degrading the energy resolution. This article studies <inline-formula> <tex-math notation="LaTeX">\sigma _{\mathrm{ baseline}} </tex-math></inline-formula> in the CANDLES detector, which severely degrades the resolution by 1% at the <inline-formula> <tex-math notation="LaTeX">Q </tex-math></inline-formula> value of 48 Ca. To avoid <inline-formula> <tex-math notation="LaTeX">\sigma _{\mathrm{ baseline}} </tex-math></inline-formula>, photon counting can be used to obtain the number of photoelectrons in each PMT; however, a significant photoelectron signal overlapping probability in each PMT causes missing photoelectrons in counting and reduces the energy resolution. "Partial photon counting" reduces <inline-formula> <tex-math notation="LaTeX">\sigma _{\mathrm{ baseline}} </tex-math></inline-formula> and minimizes photoelectron loss. We obtain improved energy resolutions of 4.5%-4.0% at 1460.8 keV (<inline-formula> <tex-math notation="LaTeX">\gamma </tex-math></inline-formula>-ray of 40 K) and 3.3%-2.9% at 2614.5 keV (<inline-formula> <tex-math notation="LaTeX">\gamma </tex-math></inline-formula>-ray of 208 Tl). The energy resolution at the <inline-formula> <tex-math notation="LaTeX">Q </tex-math></inline-formula> value is estimated to be improved from 2.6% to 2.2%, and the detector sensitivity for the <inline-formula> <tex-math notation="LaTeX">0\nu \beta \beta </tex-math></inline-formula> half-life of 48 Ca can be improved by 1.09 times.
Material properties of epoxidized natural rubber (ENR 50)-based adhesive were determined using DSC, TGA and FTIR analysis which was prepared using gum rosin and toluene as tackifier and solvent ...respectively. Quadratic regression was fitted to model the effect of peel strength and shear strength of the adhesives. Result shows that peel and shear strength increases up to an optimum molecular weight of 4.2 × 10
4
g/mol of ENR 50. Peel strength and shear strength also increases with increase in rate of testing, an observation which is associated to the viscoelastic response of the adhesive. Statistical analysis was carried out to further investigate the effect of molecular weight and testing rate on peel and shear strength.
The observation of neutrino-less double beta decay (0vßß) would be the most practical way to prove the Majorana nature of the neutrino and lepton number violation. CANDLES studies 48Ca double beta ...decay using CaF2 scintillator. The main advantage of 48Ca is that it has the highest Q-value (4.27 MeV) among all the isotope candidates for 0vßß. The CANDLES III detector is currently operating with 300kg CaF2 crystals in the Kamioka underground observatory, Japan. In 2014, a detector cooling system and a magnetic cancellation coil was installed with the aim to increase light emission of CaF2 scintillator and photo-electron collection efficiency of the photo-multipliers. After this upgrade, light yield was increased to 1000 p.e./MeV which is 1.6 times larger than before. According to data analysis and simulation, main background source in CANDLES is turned out to be high energy external gamma-ray originating neutron capture on the surrounding materials, so called (n,γ). Upgrading the detector by installing neutron and gamma-ray shield can reduce the remaining main backgrounds by two order magnitude. In this report, we discuss the detail of (n,γ) and background reduction by additional shielding.
We have studied the neutrino-less double beta decay(0νββ) of 48Ca by using CaF2(pure) scintillators. Analysis for rejection of backgrounds(212Bi→ 212Po events and 208Tl events) was effective to ...reduce backgrounds in Qββ-value region. No events are observed in the Qββ-value region for the data of 131 days × 86 kg. It gives a lower limit (90% confidence level) of (math) > 6.2 × 1022 year (preliminary) for the half-life of 0νββ of 48Ca.
The energy calibration system for CANDLES using (n,γ) reaction Iida, T.; Mizukoshi, K.; Ohata, T. ...
Nuclear instruments & methods in physics research. Section A, Accelerators, spectrometers, detectors and associated equipment,
01/2021, Letnik:
986
Journal Article
Recenzirano
Odprti dostop
Calcium fluoride for the study of Neutrinos and Dark matters by Low-energy Spectrometer (CANDLES) searches for neutrino-less double-beta decay of 48Ca using a CaF2 scintillator array. A high Q-value ...of 48Ca at 4268 keV enabled us to achieve low background, however, at the same time it causes difficulties in calibrating the detector’s Q-value region because of the absence of a standard high-energy γ-ray source. Therefore, we have developed a novel calibration system based on γ-ray emission by neutron capture on 28Si, 56Fe, and 58Ni nuclei. In this paper, we report the development of the new calibration system as well as the results of energy calibration in CANDLES up to 9 MeV.
PDF
