Cardiovascular disease (CVD) is the most common class of chronic and life-threatening diseases and, therefore, considered to be one main causes mortality. The proposed new neural architecture based on recent popularity convolutional networks (CNN) was a solution for development automatic heart diagnosis systems using electrocardiogram (ECG) signals. More specifically, ECG signals were passed directly properly trained CNN network. database consisted more than 4000 signal instances extracted...

Rainfall estimation based on the impact of rain electromagnetic waves is a novel methodology that has had notable advancements during last few years. Many studies conducted this topic in past considered only with frequencies greater than 10 GHz since rainfall wave attenuation reduced at lower frequencies. Over years, some authors have demonstrated there can be non-negligible even signals received global system for mobile communications terminal presence rain. In paper, we propose new...

The integration of artificial intelligence (AI) with Digital Twins (DTs) has emerged as a promising approach to revolutionize healthcare, particularly in terms diagnosis and management thoracic disorders. This study proposes comprehensive framework, named Lung-DT, which leverages IoT sensors AI algorithms establish the digital representation patient’s respiratory health. Using YOLOv8 neural network, Lung-DT system accurately classifies chest X-rays into five distinct categories lung...

Discontinuous transmission based on speech/pause detection represents a valid solution to improve the spectral efficiency of new generation wireless communication systems. In this context, robust voice activity (VAD) algorithms are required, as traditional solutions present high misclassification rate in presence background noise typical mobile environments. This paper presents algorithm which is noisy environments, thanks methodology adopted for matching process. More specifically, VAD...

The performance of traditional biometric identification systems is, as yet, unsatisfactory in certain applications. For this reason, other physiological or behavioral characteristics have recently been considered, using new electrical physical signals linked to a person's vital signs. This paper examines the phonocardiogram (PCG) from cardiac auscultation. idea is that PCG specific individual can be taken into consideration sign used system. More specifically, proposes preliminary study...

Today’s healthcare facilities require new digital tools to cope with the rapidly increasing demand for technology that can support operators. The advancement of is leading pervasive use IoT devices in daily life, capable acquiring biomedical and biometric parameters, providing an opportunity activate medical community. Digital twins (DTs) are a form gaining more prominence these scenarios. Many scientific research papers literature combining artificial intelligence (AI) DTs. In this work, we...

The paper proposes a performance evaluation and comparison of G.729, AMR, fuzzy voice activity detection (FVAD) algorithms. was made using objective, psychoacoustic, subjective parameters. A highly varied speech database also set up to evaluate the extent which VADs depend on language, signal-to-noise ratio (SNR), or power level.

An accurate estimate of rainfall levels is fundamental in numerous application scenarios: weather forecasting, climate models, design hydraulic structures, precision agriculture, etc. becomes essential to be able warn the imminent occurrence a calamitous event and reduce risk human beings. Unfortunately, date, traditional techniques for estimating present critical issues. The algorithm applies Convolution Neural Network (CNN) directly audio signal, using 3 s sliding windows with an offset...

The integration of Artificial Intelligence (AI) with Digital Twins (DTs) has emerged as a promising approach to revolutionize healthcare, particularly in the diagnosis and management thoracic disorders. This study proposes comprehensive framework, named Lung-DT, which leverages IoT sensors AI algorithms establish digital representation patient’s respiratory health. Using YOLOv8 neural network, Lung-DT system accurately classifies chest X-Rays into five distinct categories lung diseases,...

In recent years, automatic diagnosis of the state health heart by employing phonocardiogram (PCG) has achieved remarkable success. This letter proposes a low-complexity automated solution based on direct application multiclass convolutional neural network to PCG signals for purpose recognizing and classifying disease. are fed in network, bypassing transformations from time domain that frequencies (for example, mel-frequency cepstral coefficients (MFCC), Wavelet, etc.). Applying recurrence...

Recently, many systems and approaches that employ heart sounds as physiological traits for biometric recognition have been investigated. However, those are often tested on small, diverse closed datasets, making it difficult to compare their performance. In this paper, we present HSCT-11, an open dataset containing data collected from 206 people can be used performance evaluation of systems, use benchmark two such systems. The most performing one shows Equal Error Rate 13.66 % database, a...

The recent climatic changes imply an increasing manifestation of calamitous phenomena related to important hydrogeological disruptions in many parts the earth. For this reason, accurate estimate rainfall levels becomes essential be able warn imminent occurrence a event and reduce risk human beings. This paper proposes approach based on Convolutional Neural Networks (CNN) classification audio signal coming from new system.

This paper presents new results in human identity verification via frequency analysis of cardiac sounds. More specifically, the proposes a pattern recognition approach based on feature set 13 Mel Frequency Cepstral Coefficients (MFCCs) extracted from first (S1) and second (S2) heart sounds metric power ratio S1 to S2. The algorithm yields significantly better performances with respect previous method Chirp z-transform, guaranteeing an equal error rate (EER) below 9%.

Communication systems based on man-machine interaction play an increasingly important role in everyday life. Whereas these make life easier for many people, the hearing impaired they are indispensable. Automatic speech recognition particular, thanks to simultaneous display of text, reduce gap separating deaf, or general, from world hearing. The paper presents automatic aid system while behind wheel a car. It proposes algorithm, typical techniques, identification and signaling emergency...

The paper proposes a study of background noise classifier based on pattern recognition approach using neural network. signals submitted to the network are characterised by means set 12 MFCC (Mel frequency cepstral coefficient) parameters typically present in front end mobile terminal. performance classifier, evaluated terms percent misclassification, indicate an accuracy ranging between 73% and 95% depending duration decision window.

The paper proposes an intelligent data sensing and geo-localization algorithm, based on innovative mobile computing system that measures the power level of RF sources through a 2G/5G femtocell-UAV system. In natural disasters (mainly earthquakes floods) can identify any missing persons under rubble within range precision between 1 to 2 meters. this paper, more specifically, algorithm allows classifying terminal even in presence obstacles cause anisotropic propagation radio signals, series...

In this paper we propose a cardiac biometric system for human identity verification based on an automatic selection algorithm of the best subsequence DHS (digital heart sound) signal. The quality score is cepstral distance between homogeneous sounds. Performance proposed, expressed in terms equal error rate, similar to manual segmentation-based system, but offers advantages fully application.

One of the most recent innovations in field biometric recognition is usage heart sounds as physiological traits for identity verification. In this paper, we propose a statistical approach, supported by Gaussian Mixture Models, to problem verification based on sounds. The system validated database acquired from 147 people, and shows significant performance improvement with respect other approaches, yielding an Equal Error Rate 15.53%.

The recent development of the IoT (Internet Things), which has enabled new types sensors that can be easily interconnected to Internet, will also have a significant impact in near future on management natural disasters (mainly earthquakes and floods) with aim improving effectiveness research, identification, recovery missing persons, therefore increasing possibility saving lives. In this paper, more specifically, an innovative technique is proposed for search identification persons disaster...

The paper proposes a performance evaluation and comparison of recent ITU-T ETSI voice activity detection algorithms. was made using both objective psychoacoustic parameters, so as to have reliable judgements that were close subjective ones. A highly varied speech database also set up evaluate the extent which VAD depend on language, signal noise ratio, or power level.