Biomedical signal and image processing

W artykule powstałym jako zapis tak zwanego "wykładu mistrzowskiego", który autor miał obowiązek wygłosić podczas przyznawania mu godności honorowego profesora AGH, naszkicowano przebieg pierwszych działań, które w rezultacie doprowadziły do powstania kierunku studiów Inżynieria Biomedyczna na AGH. Wspomniano o tym, jak w latach 70. XX wieku zarówno inżynierowie, jak i medycy doszli do wniosku, że rozwój diagnostyki i terapii medycznej może zostać znacząco przyspieszony, gdy obok sposobu widzenia organizmu pacjenta jako swoistego reaktora chemicznego, weźmie się pod uwagę fakt, że w organizmie tym toczą się też liczne procesy fizyczne. Owe procesy można śledzić przy użyciu specjalnej technicznej aparatury diagnostycznej, a także można wspomagać za pomocą urządzeń technicznych pełniących rolę terapeutyczną. Droga od owego uświadomienia do realnego wykorzystania była jednak trudna i skomplikowana. Drogę tę w skrócie i uproszczeniu opisano w tekście artykułu.

Waste management has become a significant issue in recent years due to rapid urbanization and population growth. Conventional trash disposal techniques are hazardous to the environment and ineffective. This study suggests a smart waste management system using the Internet of Things (IoT) to manage garbage effectively. The system uses sensors to sort waste, cloud computing to store data, and machine learning to improve processing. By collecting real-time data from waste bins, it enhances recycling and streamlines collection, promoting sustainable urban living. Using realtime data collection and advanced analytics, the proposed approach improves waste collection efficiency, reduces operational costs, and promotes environmental sustainability. The system makes use of IoT-enabled smart bins with two ultrasonic sensors to monitor waste levels and classify different waste categories. Which Node MCU is responsible for .One ultrasonic sensor detects the quantity of waste in the bin, while another sensor detects a person approaching to dispose of waste. This detection allows the lid to open and close automatically. A servo motor that is mounted to the lid makes it easier to open and close. The collected data is processed in a cloud-based system using machine learning algorithms to enhance trash segregation and optimize collection schedules. Additionally, to ensure efficient waste collection and reduce fuel consumption and travel distances, an AI-powered route optimization system is employed. The research evaluates numerous IoT-based frameworks and proposes the ideal solution for smart cities.In comparison to traditional waste management techniques, experimental results show that the system is effective in optimizing waste segregation and collection, resulting in a 35% increase in efficiency. This research supports on-going efforts in developing smart cities, opening the door to a cleaner and more sustainable future.

Study objectives: To assess cardiovascular control during sleep in children with sleep-disordered breathing (SDB) and the effect of adenotonsillectomy in comparison to healthy nonsnoring children. Methods: Cardiorespiratory signals obtained from overnight polysomnographic recordings of 28 children with SDB and 34 healthy nonsnoring children were analyzed. We employed an autoregressive closed-loop model with heart period (RR) and pulse transit time (PTT) as outputs and respiration as an external input to obtain estimates of respiratory gain and baroreflex gain. Results: Mean and variability of PTT were increased in children with SDB across all stages of sleep. Low frequency power of RR and PTT were attenuated during non-rapid eye movement (REM) sleep. Baroreflex sensitivity was reduced in children with SDB in stage 2 sleep, while respiratory gain was increased in slow wave sleep. After adenotonsillectomy, these indices normalized in the SDB group attaining values comparable to those of healthy children. In children with mild-to-moderate SDB, vasomotor activity is increased and baroreflex sensitivity decreased during quiet, event-free non-REM sleep. Adenotonsillectomy appears to reverse this effect.

This work aims to develop a subject-independent Emotion Detection System (EDS) based on EEG signals and the 3D Valence-Arousal-Dominance (VAD) model. The DEAP database physiological signals are considered for the system design. A multi-domain feature extraction is performed using the Wavelet-based Atomic Function time-frequency domain technique; and various time and frequency domain feature extraction techniques. Further, principal component analysis reduces the data dimensionality and redundancy in the obtained feature set. The minimal feature set is analysed using machine learning classifiers, i.e., gradient boosting, decision tree, and random forest. Additionally, the hyperparameters of machine learning algorithms are tuned using Optuna to improve the performance of the proposed model. Three EDS are designed in this work; EDS1 considers the 3D VAD model for three class classifications. 2D VA model is used in EDS2 to determine 9 discrete emotions, and EDS3 detects 12 discrete emotions using the 3D VAD model. Results reveal the highest classification accuracy of about 99% with EDS1, whereas an average accuracy of 99.82% and 98.44% is obtained with EDS2 and EDS3, respectively. The results reveal that both EDS1 and EDS2 show improvement as compared to the existing literature. Also, the proposed EDS3 provides the classification of the highest number of discrete emotions, which may facilitate the design of an efficient human-computer interface system.

This study describes the design, creation, and testing of an automatic smart trash bin (STB) aimed at improving hygiene and independence for people with disabilities. Regular trash bins require hand operation, which can be tough for those with limited mobility. Our STB uses an Arduino UNO microcontroller, two ultrasonic sensors, a servo motor, and a 1602A LCD with an I2C module to allow touch-free waste disposal. One sensor detects when someone is near and opens the lid, while the other checks the trash level inside for better waste management. The LCD shows real-time updates to help users. Tests show the system accurately detects objects, uses little power, and gets good feedback from users, proving it can greatly help in homes, public spaces, and healthcare settings. By adding new ideas like the Internet of Things (IoT) and user-friendly design, this work pushes forward assistive technology (AT) to make waste management more accessible.

The insertion of the needle is difficult because the deformation and displacement of the organs are the key elements in the surgical act. Liver and tumor modeling are essential in the development of the needle insertion model. The role of the needle is to deliver into the tumor an active chemotherapeutic agent. We describe in this chapter the deformation of the needle during its insertion into the human liver in the context of surgery simulation of the high- robotic-assisted intraoperative treatment of liver tumors based on the integrated imaging-molecular diagnosis. The needle is a bee barbed type modeled as a flexible thread within the framework of the Cosserat (micropolar) elasticity theory.

A torsional nano-pendulum based on a single molecule suspended on a carbon nanotube with a single wall is considering in this paper. The molecule is rotated, resulting in large elastic torsional deformations of the nanotube. The motion equation of this system leads to the sine-Gordon equation. The closed solutions of the sine-Gordon equation, which are expressed in terms of Jacobi elliptic functions (cnoidal solutions) are analyzed by applying the variable separation method.

Pola sidik bibir manusia dapat dijadikan identifikasi pada individu. Pada saat ini belum ada pengidentifikasian individu dengan menggunakan metode Content Based Image Retrieval Based On Gabor Wavelet. Tujuan penelitian ini adalah untuk mengidentifikasi pola sidik bibir individu dengan menggunakan metode Content Based Image Retrieval Based Metode penelitian yang digunakan adalah metode Content Based Image Retrieval Based On Gabor Wavelet untuk mengekstraksi ciri sesudah dilakukannya pre-processing dan tahap selanjutnya adalah mengklasifikasikannya dengan menggunakan metode K-Nearest Neighbor. Sampel citra bibir diperoleh dari Laboratorium Forensik Odontologi Fakultas Kedokteran Gigi, Universitas Padjajaran. Hasil penelitian ini didapatkan bahwa akurasi dengan pixel 512x256 saat K=1 40.63%, saat K=3 40.63%, dan saat K=5 43.75%, dengan pixel 256x128 saat K=1 43.75%, saat K=3 40.63%, dan saat K=5 40.63%, dan dengan pixel 128x64 saat K=1 40.63%, saat K=3 34.38%, dan saat K=5 50%. Simpulan penelitian ini menunjukkan bahwa metode Content Based Image Retrieval Based On Gabor Wavelet dan klasifikasi K-Nearest Neighbor tidak optimal untuk dapat mengidentifikasi pola sidik bibir.

Removing artifacts from electroencephalography (EEG) signals is a common technique. Although numerous algorithms have been proposed, most rely solely on EEG data. In this study, we introduce a novel approach utilizing a hybrid convolutional neural network-long short-term memory (CNN-LSTM) architecture alongside simultaneous recording of facial and neck EMG signals. This setup enables the precise elimination of artifacts from the EEG signal. To validate the method, we collected a dataset from 24 participants who were presented with a light-emitting diode (LED) stimulus that elicited steady-state visual evoked potentials (SSVEPs) while they performed strong jaw clenching, an action known to induce significant artifacts. We then assessed the algorithm's ability to remove artifacts while preserving SSVEP responses. The results were compared against other commonly used algorithms, such as independent component analysis and linear regression. The findings demonstrate that the proposed method exhibits excellent performance, effectively removing artifacts while retaining the EEG signal's useful components.

This paper presents a FIR adaptive filter to cancelling interference due to 50 Hz power line in human electroencephalographic (EEG) inspection. An overview of EEG techniques to study brain activity will be outlined. Then, the problem of spurious signal rejection, in this particular handset, shows the meaningfull application of adaptive filter structure. Poles of filter are driven to find an optimum bandwidth to eliminate noise. Least-Mean-Square (LMS) algorithm is used for adaptive process. Effect of finite filter length and non-zero response time, and their effect on time response will be also showed. I.Introduction Normal brain activity consist of a superposing of four types of “characteristic waves” denoted from greek alphabetical letter α, β, γ, δ. During an EEG exam, that shows these waves superposing, a very low amplitude periodic signal due to power line interference could be observed. So, an extraction signal method is necessary to preserve interesting biopotential. Figure 1...

Health monitoring of pregnant women is essential throughout the entire course of pregnancy to ensure proper growth of fetus. In gynecology centers and hospitals some special instruments are used to monitor the health conditions of mother and foetus. In practice, hospital based monitoring system is employed where the pregnant ladies move to the concerned centre to monitor their health. It becomes complex for them to travel and wait for long hours. To overcome this issue an instrument which is used to monitor the important parameters such as ECG (Electro cardio gram), respiration rate, heart rate and rate of dehydration is designed. All the parameters can be monitored on the computer screen which is IoT(Internet of Things) enabled so that the doctor can monitor these parameters as and when the patient uses the instrument kept at her home by fixing an appointment with the doctor. This technology has a desired impact on industrial and hospital automation. The hardware of this project is built on a printed circuit board, constituting PIC micro controller 18F4550. It is a versatile micro controller with in-built features such as 10 bit multi-channel ADC, USART, synchronous serial port, programmable low Voltage detection circuit etc., which is to be interfaced to PC system through RS232C. The software can be easily modified for any alarm setting or record intervals since it is written in Visual Basic.

this paper presents a new approach based on the continuous wavelet transform CWT and JADE algorithm for the blind source separation. The JADE algorithm has been widely used to separate the fECG and mECG signals from 8 recordings or channels, while not all of this number is needed. The present work will show that it isn’t the number of channels that counts but rather the quality of the channel used in terms of the energy or the information that it carries. That’s why before introducing the 8 channels to the JADE algorithm we will make a selection by the mathematical microscope Continuous Wavelet Transform CWT and we will show that the number of channels will be reduced to 5 or 3 or even to 2 channels. This algorithm has been validated on several real data. Keywords—blind source separation BSS. continuous wavelet transform. Independent Component Analysis ICA. JADE.

This paper discusses a linear time invariant (LTI) systems approach to signal enhancement via projective subspace techniques. It provides closed form expressions for the frequency response of data adaptive finite impulse response eigenfilters. An illustrative example ...

Monitoring systems in intensive care units (ICU) generate a high rate of false alarms 80% . This has the effect of desensitizing the medical staff and extending the response time to these alarms. In this work we present a new method to reduce false alarms in intensive care units that is based on packet wavelet decomposition and alarms classification by SVM.

Suicide, considered as one of the most leading causes of death, has not given enough and appropriate attention in order to reduce its rate such that the humans in all over the world deserve it. The problem addressed in this paper is analysis of the relation between an extra stimulus and physiological data's responses. In order to record the physiological data set from multiple subjects over many weeks, we used an acoustic startle during a Paced Auditory Serial Addition Task (PASAT) test that spontaneously leads subjects to real emotional reactivity, without any deliberate laboratory setting. Crucially, we show that, by inducing anxiety during the test, changes appear in Electrodermal activity, Electrocardiogram, Heart Rate and Respiration Rate. A wide range of physiological features from various analysis domains, including modeling, time/frequency analysis, an algorithm and etc., is proposed in order to find the best emotional reactivity feature to correlate them with emotional states which can be considered as a suicide factor. More specifically, this paper is focused on the EDA data analysis. Experimental results highlight that all cited techniques perform well and we achieved a high resolution of tonic and phasic components which allow us to measure the latency, onsets and amplitudes of EDA responses to a stimulus. This paper follows the association of recommendations for advancement of health care instruments.

This paper presents a smart garbage monitoring and clearance alert system using the Internet of Things (IoT) to address inefficiencies in traditional waste management. The system integrates load and ultrasonic sensors to monitor both the weight and fill level of garbage bins. An ESP32 microcontroller processes the sensor data and activates a servo motor to close the bin lid when full. Real-time alerts are sent to authorities for timely waste clearance. This approach ensures cleaner environments, reduces manual inspections, and promotes efficient resource usage.

This paper presents a smart garbage monitoring and clearance alert system using the Internet of Things (IoT) to address inefficiencies in traditional waste management. The system integrates load and ultrasonic sensors to monitor both the weight and fill level of garbage bins. An ESP32 microcontroller processes the sensor data and activates a servo motor to close the bin lid when full. Real-time alerts are sent to authorities for timely waste clearance. This approach ensures cleaner environments, reduces manual inspections, and promotes efficient resource usage.

This paper relates to the separation of single channel source signals from a single mixed signal by means of independent component analysis (ICA). The proposed idea lies in a time-frequency representation of the mixed signal and the use of ICA on spectral rows corresponding to different time intervals. In our approach, in order to reconstruct true sources, we proposed a novelty idea of grouping statistically independent time-frequency domain (TFD) components of the mixed signal obtained by ICA. The TFD components are grouped by hierarchical clustering and k-mean partitional clustering. The distance between TFD components is measured with the classical Euclidean distance and the β distance of Gaussian distribution introduced by as. In addition, the TFD components are grouped by minimizing the negentropy of reconstructed constituent signals. The proposed method was used to separate source signals from single audio mixes of two- and three-component signals. The separation was performed...

Fire hazards continue to be a major concern across various environments, including residential, commercial, and industrial spaces. Conventional fire detection systems, such as smoke or heat sensors, often suffer from limitations like delayed response times and false alarms. This paper presents a vision-based fire detection system utilizing OpenCV and deep learning techniques to address these challenges. The proposed system employs a Convolutional Neural Network (CNN) trained on a curated dataset comprising fire and non-fire images to accurately identify fire in real-time video streams. OpenCV is used for image preprocessing and frame extraction, enabling efficient analysis of live surveillance footage. The model demonstrates high accuracy and low latency in fire detection, making it suitable for deployment in critical environments requiring rapid response. The integration of deep learning with computer vision offers a scalable and cost-effective alternative to traditional sensor-based fire detection methods. Furthermore, the system's adaptability allows it to be integrated with existing surveillance infrastructure, reducing the need for specialized hardware installations. Its performance was evaluated under various lighting conditions and backgrounds, demonstrating robust detection capabilities even in complex visual environments. The modular design also opens possibilities for future enhancements, such as incorporating smoke detection, alert mechanisms, or deployment on edge devices for on-site processing. This research contributes to the advancement of intelligent safety systems by combining the strengths of deep learning and computer vision for proactive fire risk management.

The introduction of renewable energy sources, distributed energy systems, and power electronics equipment has led to the emergence of the Smart Grid. However, these developments have also caused the worsening of power quality. Selecting the correct sampling frequency and feature extraction techniques are essential for appropriately analyzing power quality disturbances. This work compares the performance of an algorithm based on a Support Vector Machine and Discrete Wavelet Transform for the classification of power quality disturbances using eight sampling rates and five different mother wavelets. The algorithm was tested in noisy and noiseless scenarios to show the methodology. The results indicate that a success rate of 99.9% is obtained for the noiseless signals using a sampling rate of 9.6 kHz and 95.2% for signals with a signal-to-noise ratio of 30 dB with a sampling rate of 30 kHz.

The model suggested for detection of financial fraud detection a Modified Graph Convolutional Network with Temporal Features Network (MGCN-TFNet) for spatial as well as temporal data analysis. The model initiates with acquiring the dataset and normalizing it, followed by applying Graph Convolutional Networks to learn about transactional structural patterns and embeddings of features. Temporal attention layers emphasize corresponding time-based features and identify transactions according to relevance for various windows of time. An integration layer of temporal-structural embeddings enhances fraud detection. A fine-tune and final classification fully connected layer by a SoftMax classifier detect valid and invalid transactions. The performance of the model is tested based on a confusion matrix, with an accuracy of 0.991, precision of 0.992, sensitivity of 0.990, specificity of 0.992, and 0.991 in Fmeasure. The Negative Predictive Value (NPV) is 0.939, which is indicative of superior performance in identifying true fraudulent transactions. This methodology is very effective in integrating structure learning and temporal pattern discovery for detecting financial fraud.

Estados Unidos y los países en los que tiene influencia técnica utilizan redes de 60 Hz, mientras que Europa y los países en los que tuvo capacidad de influir cuando se comenzó la red de deistribución de electricidad, prefieren 50 Hz. Por número de países, son mayoritarias las redes de 50 Hz. ¿Qué tipo de frecuencia ofrece mayores ventajas? Una de mis alumnas, Nazaret Sinde Lorenzo, me remitió esta respuesta, que me parece interesante y técnicamente correcta: La utilización de diferentes frecuencias en las diferentes zonas geográficas desarrolladas es más bien una razón de tradición, que una razón técnica. Desde el punto de vista técnico y en comparativa de las frecuencias de 50 Hz y 60 Hz no se puede decir que una frecuencia sea muy ventajosa con respecto a la otra, y si así fuera no habría ambas frecuencias, se utilizaría la mejor. Pero también es cierto que si se observan ciertas diferencias entre ambas; por ejemplo en algunas lámparas que utilizan rectificadores, el parpadeo se aprecia menos en el caso de frecuencias de 60 Hz que 50 Hz, debido a que el ciclo se repite cada menos tiempo.

Automatic Speech Recognition (ASR) and spoken language systems have indeed made remarkable strides, fueled by advances in artificial intelligence, deep learning, and computational power. Modern ASR systems are now more accurate, robust, and capable of handling a variety of applications, from voice assistants to real-time transcription services. This review discusses the developments in isolated Arabic speech recognition using different AI methodologies. It emphasizes fundamental techniques, such as deep learning and machine learning algorithms, and then evaluates their effectiveness in enhancing recognition accuracy. This paper highlights the inherent obstacles of the Arabic language, including dialectal differences and phonetic complexities. It also examines the importance of feature extraction and model training in improving performance. The methodologies used in speech detection and processing, identifying gaps and correlations between known patterns, and presenting recent patterns are illustrated in this paper. A systematic review of the selected studies was conducted to identify and select relevant papers. The evaluation indicates that despite significant progress, additional research is needed to overcome current limitations and enhance practical application.

Abstract—Recent years have seen significant research in finding closed form expressions for the delay of the RLC interconnect which improves upon the Elmore delay model. However, several of these formulae assume a step excitation. But in practice, the input waveform does have a non zero time of flight. There are few works reported so far which do consider the ramp inputs but lacks in the explicit nature which could work for a wide range of possible input slews. Elmore delay has been widely used as an analytical estimate of ...

In this paper, we describe the use of deep learning-based model for classifying kidney stone images in order to overcome the lack of medical infrastructures within the hospitals in the development countries like Madagascar. Moreover, the use of mobile phones such as smartphone is very popularized in these countries. This situation leads us to perform a lightweight deep learning-based model inspired from the architecture of SSDLiteX. The Balancing technical was used to achieve higher performance. Our system provides 100% of accuracy, validation accuracy and F1-score. The result shows that this model surpass all lightweight models recorded to the state of the art about kidney stone classification. According to its nature, the proposed approach could run on the low resource environment like Smartphone. Thus, it is reliable and answer the need of the specialist within the hospitals in the development country.

The world suffers the impacts of the climate change caused by forest destruction and air pollution. Lung cancer is among of impacts where
the population suffers mainly in the development country. The aim of this paper is to demonstrate the performance, the reliability and the effectiveness of using a lightweight deep learning base model for classifying lung cancer CT scan images. Furthermore, in the development country the lack of medical infrastructures and the specialist within the hospitals result numerous misdiagnoses. Or, some previous works were more complex and heavies. These situations conduct us to perform a lightweight
deep learning base model inspired from the SSDLiteX approach for classifying lung cancer to aid specialist. The train data was balanced to obtain higher performance. Our model achieves 97% of accuracy, validation accuracy and F1-score. The result shows that the proposed approach is reliable for classifying lung cancer and could run on the low
resource environment.

Proses identifikasi pada individu dilakukan dengan dua metode yaitu identifikasi primer dan sekunder. Sidik bibir menjadi identifikasi sekunder jika pemeriksaan pada identifikasi primer tidak mendapatkan hasil yang akurat. Sidik bibir memiliki pola yang unik dan stabil dan tidak berubah seiring bertambahnya umur individu. Penelitian ini menggunakan metode ekstraksi ciri Gray Level Co-Occurrence Matrix (GLCM) dan metode Local Binary Pattern (LBP) dengan klasifikasi ciri Decision Tree. Citra latih yang digunakan berjumlah 50 dan 36 citra uji. Penelitian ini bertujuan untuk memudahkan proses identifikasi kepemilikan sidik bibir yang merupakan hasil kerjasama Fakultas Kedokteran Gigi Universitas Padjajaran dengan mahasiswa Universitas Telkom. Hasil dari penelitian ini didapatkan metode terbaik adalah GLCM dengan nilai akurasi sebesar 83,3333% menggunakan parameter orde dua energi, korelasi, homogenitas, kontras dan 8 level kuantisasi. Pada LBP dihasilkan akurasi sebesar 61,1111% dengan radius R=1 dengan masing-masing metode menggunakan klasifikasi Decision Tree.

In this study we propose an automatic method for solving convolutive mixtures separation. The independent components are extracted by frequency domain analysis, where the convolutive model can be solved by instantaneous mixing model approach. The signals are reconstructed back in the observation space resolving the ICA model ambiguities. Simulations are carried out to test the validity of the proposed method

In this paper we compare the performance of different algorithms employed in solving frequency domain blind source separation of convolutive mixtures. The convolutive model is an extension of the instantaneous one and it allows to relax the hypothesis of a linear mixing process in which all the sources are supposed to reach the electrodes at the same time. This test is carried out in the frequency domain, where the algorithms developed for independent component analysis can be employed with minor modifications. The decomposition performance of such algorithms is evaluated on simulated dataset of convultive mixtures of biomedical signals.

Due to the advancements in digital technologies and social networking, image collections are growing exponentially. The important aim in content-based image retrieval (CBIR) is to reduce the semantic gap issue that improves the performance of image retrieval. In this paper, the objective is achieved by introducing effective visual words fusion technique based on speeded-up robust features (SURF) and histograms of oriented gradients (HOG) feature descriptors. HOG is used to extract global features, whereas SURF is used for the extraction of local features. Global features are preferred for large-scale image retrieval, whereas local features perform better on those systems that support semantic queries with close visual appearance. Moreover, SURF is scale and rotation-invariant as compared to HOG descriptor and it works better for low illumination. On the contrary, HOG performs better for scene-recognition-or activity-recognition-based applications. In the proposed technique, visual words fusion of SURF and HOG feature descriptors is carried which performed better than features fusion of SURF and HOG feature descriptors as well as state-of-the-art CBIR techniques. The proposed technique based on visual words fusion gives classification accuracy of 98.40% using support vector machine while image retrieval accuracy of 80.61%. Qualitative and quantitative analyses performed on four standard image collections namely, Corel-1000, Corel-1500, Corel-5000, and Caltech-256 demonstrate the effectiveness of the proposed technique based on visual words fusion of SURF and HOG feature descriptors. Content-based image retrieval • Classification • Local and global features • Semantic gap • Visual words fusion B Zahid Mehmood

Parkinson's disease (PD) is responsible for a broad spectrum of signs and symptoms, including relevant motor impairments generally rated by clinical experts. In recent years, motor measurements gathered by technology-based systems have been used more and more to provide objective data. In particular, wearable devices have been adopted to evidence differences in the gait capabilities between PD patients and healthy people. Within this frame, despite the key role that the upper limbs' swing plays during walking, no studies have been focused on their harmonic content, to which this work is devoted. To this end, we measured, by means of IMU sensors, the walking capabilities of groups of PD patients (both de novo and under-chronic-dopaminergic-treatment patients when in an off-therapy state) and their healthy counterparts. The collected data were FFT transformed, and the frequency content was analyzed. According to the results obtained, PD determines upper limb rigidity objectively evidenced and correlated to lower harmonic contents.

As in every developing country, Madagascar is still fighting against the tuberculosis (TB), the second leading cause of death from a communicable infectious disease. Its incidence remains high, exacerbated by limited medical infrastructure and unequal distribution of health resources. Several researches were conducted to automatically diagnosis disease to greatly lowering the total expenses. Deep learning approaches demonstrated its high effectiveness in the domain using chest x-ray images (CXR) that are the most cost-effective and widely utilized imaging technology to classify TB. Some preprocessing was necessary for getting better features from the images. Using a more lightweight model inspired by SSD Lite X, this work achieves the highest accuracy (training, validation and testing), recall, F1-Score and specificity of 100% each, making this model the most effective one in the state-of-the-art. It could be deployed in a low-resource environments, enabling automated diagnosis in every region of Madagascar and other developing countries.

Antimicrobial agents are widely used for treatment of animal and human diseases. Heavy use of antimicrobial agents permits bacteria to develop resistance to these agents specifically when a dose of antibiotic is insufficient or course of treatment is incomplete. Antimicrobial resistance genes (ARGs) are usually associated with mobile genetic elements (MGEs) including Integron therefore; these genes can transmit among bacteria via horizontal transmission. The current study was conducted to assess the possible role of manure in dissemination of antimicrobial resistance. The presence, quantitate, and diversity of resistance genes associated with Integron class 1 have been assessed using conventional and quantitative polymerase chain reaction (PCR) combined with sequencing of gene cassette within Integron and analysis of sequenced data by blast tool. Thirty-eight samples were found a positive for Integron and concentration of Integron in positive sample ranged from from 10 6-10 10 copies/g of manure. High frequencies were detected to genes that encoded to sulphonamide and ammonium compound resistance. These genes were detected in 25% and 23% respectively of the total manure samples. In general, the detected genes in manure functionally belong to five protein families including Efflux pump, DNA repair, heavy metal resistance, membrane protein, and antibiotic resistance. Manure might act as a hotspot from which ARGs emerge and transfer to the environment and then to the animal and human environments.

Resumen: En este trabajo se evalúan tres algoritmos de segmentación automática del intervalo RT. Estos algoritmos son: el algoritmo basado en un filtro derivador, el algoritmo basado en la descomposición de ondículas y el algoritmo basado en la modelización de la onda T. La evaluación se efectuó sobre señales electrocardiográficas sintéticas y reales. En la primera evaluación se contaminó un electrocardiograma (ECG) sintético con cuatro tipos de ruido en cuadro condiciones de Relación Señal a Ruido (SNR) (10, 15, 20 y 25 dB), en la segunda evaluación se compararon los resultados arrojados por los algoritmos con las anotaciones de dos expertos. Los resultados obtenidos muestran comportamientos satisfactorios de los tres enfoques en condiciones favorables de SNR. Sin embargo se observó un aumento de la variabilidad del intervalo RT inducida por el ruido.

Artificial intelligence techniques in computer vision have made a substantial contribution to the development of imaging analysis in medicine by improving the accuracy of predictions, which has resulted in more suitable treatment and diagnostics. By offering a second view, these techniques can help hematologists and other medical professionals make better diagnoses in the area of automated leukemidical imaging and cancer in the blood diagnosis. A thorough examination of the existing DM and DL image processing algorithms is provided in this study, with a particular emphasis on how to identify of leukocytes in smear blood imaging alongside various clinical imaging regions. The primary aim of the suggested investigation is to identify the best DM and DL techniques for clinical imaging, particularly for identifying the types of lymphocytes from smear blood data. This review article delves closely into the sophisticated algorithms for DL, especially the emerging models built around CNNs that operate in the clinical computational imaging area. According to a review of associated research, white blood cell identification using micro smear captures is a common application of standard AI technologies. They aid in the diagnosis of numerous illnesses, including blood cancer, and give medical professionals important information. As the researchers and professionals assigned to the processing of medical images, we construct recommendations for future investigations based on the extensive analysis of white blood cell associated research study that is laid out through the current research.

Chronic kidney disease (CKD) is considered a risk factor for the development of cardiovascular disease. QT interval is an electrocardiographic parameter that quantifies the duration of ventricular repolarization. An increase of its spatial variability measured from the selected leads of a standard electrocardiogram (ECG), named QT dispersion (QTd), is considered a risk factor for malign ventricular arrhythmias and sudden death in the CKD. An algorithm for automatic measurement of QTd in the ECG leads DI, aVF and V2 using the continuous wavelet transform with splines is presented. Validation of QRS complex detection has been done on records from MIT-BIH database, and the accuracy is 99.5%. Validation of detection of QRS wave onset and T wave end has been done on records from CSE and QT databases, and the measurements were within the tolerance limits for deviations with respect to the manual measurements defined by the experts. The algorithm was applied in two studies. In the first st...

There are two main processes to create a 3D animatable facial model from photographs. The first is to extract features such as eyes, nose, mouth, and chin curves on the photographs. The second is to create 3D individualized facial model using extracted feature information. The final facial model is expected to have an individualized shape, photograph-realistic skin color, and animatable structures. Here, we describe our novel approach to detect features automatically using a statistical analysis for facial information. We are not only interested in the location of the features but also the shape of local features. How to create 3D models from detected features is also explained and several resulting 3D facial models are illustrated and discussed.

Dans cet article nous allons présenter une nouvelle méthode temps-fréquence pour la reduction des terms d'interférence. On va considérer le cas des signaux multi-composants, pour lequel l'élimination des termes d'interférence est difficile à résoudre. Pour extraire les composants utiles, nous proposons une technique basée sur le concept de décomposition atomique. Nous définissons une structure algorithmique pyramidale, en envisageant deux buts: l'amélioration des performances du processus d'estimation et la réduction de la complexité. On met en évidence les avantages de cette méthode, en ce qui concerne l'estimation du taux de modulation linéaire et la complexité.

Over the years people have been monitoring some vital signs in human's health. Notably, in the physiological study of mankind, various methods and techniques are being used to track the heartbeat, body temperature, respiratory rate and arterial blood pressure. This quest was to ensure that humanity is in good condition at all time. Since health is wealth, the human's health becomes a crucial and vital to every medical professionals and scientists as well. Up till today, it is still in progress, and more ways of solving human's health problems keep on unfolding on daily bases. This work takes a critical survey of human health monitoring system, with focus on heart beat, body temperature and blood pressure monitoring devices and systems, how they evolved over the years, and what the human health monitoring system of today are capable of. It also noted some research gaps and sets some agenda for further research work. The ideas presented in this paper will open up more areas for further studies. It will enable researchers to further advance the methods and tools used in monitoring the identified vital signs in human health.

Smart agriculture uses the internet of things, usually for irrigation purposes. IoT is employed in this case to manage the water log on fields. The soil moisture sensor is used to keep track of the soil moisture in agricultural areas. The Cloud Service Brokerage issues a directive to the relay telling it to turn ON the suction motor when the moisture level reaches a specific level. The farmland's extra water is removed by the suction motor. With this proposal, we want to use IoT to analyze the availability of micronutrients and control the excess water log in farming. This semi-automated water management system can operate independently or be managed using a smartphone app.The weather at that particular place may be analyzed using the DHT sensor values, and it can be projected using a prediction algorithm for the following 10 days. IoT refers to the idea of linking all gadgets to the internet and enabling internet-based communication between them. The Internet of Things (IoT) is a vast network of interconnected gadgets that collectively gather and share information about their usage and the settings in which they function. IoT is an agricultural use of contemporary information and communication technologies. A system for agricultural monitoring is constructed using sensors

Very low bit-rate video coding has recently become one of the most important areas of image communication and a large variety of applications have already been identified. Since conventional approaches are reaching a saturation point, in terms of coding efficiency, a new generation of video coding techniques, aiming at a deeper "understanding" of the image, is being studied. In this context, image analysis, particularly the identification of objects or regions in images (segmentation), is a very important step. This paper describes a segmentation algorithm based on split and merge. Images are first simplified using mathematical morphology operators, which eliminate perceptually less relevant details. The simplified image is then split according to a quad tree structure and the resulting regions are finally merged in three steps: merge, elimination of small regions and control of the number of regions.

Monitoring weather plays a crucial role in human life, making the collection of data on climate changes imperative. This paper aims to develop a weather station to monitor climate parameters in off-grid regions. The project includes sensors for detecting temperature, humidity, rainfall, LPG levels, carbon monoxide, smoke, altitude, and barometric pressure. The Arduino gathers information from the sensors, displays it on an LCD screen, and additionally sends SMS alerts via a GSM module. Accurate weather updates are essential for successful planning in agriculture, fishing, and various other activities. Unfortunately, weather information often lacks accuracy for specific regions, relying instead on data from the nearest weather station. This project calculates and compares the error level between national climate forecasts and actual readings. It is designed to assist individuals in remote and off-grid areas who rely on weather information for their livelihoods.

As fossil fuels are decreasing day by day, renewable power turns out to be the main energy source. Anyway, the interest is still lower than expected. Among them, one of the newest energy-gathering techniques is changing the mechanical energy from falling raindrops into power. This task focuses on the small, creating power with the help of falling raindrops. The benefits of this paper are that it can help to add to minimizing the disadvantages of diminishing fossil-based fuels. Regardless, in the utilization of making power from the rain, the available head and stream rate is extremely low for the utilization of Pelton, Francis, Kaplan, or various kinds of water-driven turbomachines. This task revolves around the use of the developed Savonius rotor, which is mounted in an even circumstance to isolate energy from falling raindrops. In this examination, water is first stored in a tank before being conveyed like a stream of water onto a rotor. This rotor is unequivocally planned for this application and is an adjusted version of the Savonius water rotor. This instrument was used to generate a small amount of power. SOLIDWORKS was used to design the modified water turbine, which was then analyzed using Computational Fluid Dynamics (CFD) in ANSYS simulation software. The performance is analyzed with the test variables of rotational speed, torque, power, torque coefficient, tip speed ratio, and power coefficient. Then the experimented values are compared with the numerical values to know the accuracy of the values. The output of the experimental value was 15.452 watts, and in the numerical analysis, it was found to be 16.501 watts.

The Electrocardiogram (ECG) signal is used to record the electrical activity of heart. The subtle variations in ECG attributes are used by cardiologists for diagnosis of heart anomalies. But, for prognosis of cardiac ailments feature extraction from electrocardiographic signal becomes extremely difficult due to presence of noise. With the aim of noise reduction, a hybrid technique involving Extended Kalman filter along with Discrete Wavelet transform for effectively improving signal quality is focused as a powerful tool. The performance of denoising algorithm is evaluated in terms of signal to noise ratio and mean square error. On denoised signal, a quick, simple and effectual approach based on Principal Component Analysis is proposed for R-peak and QRS complex detection. The beat detector performance is validated with MIT-BIH arrhythmia database, yielding a sensitivity of 99.93%, positive predictivity of 99.98% and a 0.079% detection error rate, being a positive outcome in comparison with recent researches. Later, different types of arrhythmias are detected on the basis of heart rate and morphological characteristics of ECG waveform.