Biomedical signal and image processing

Speaker Recognition is a most prominent emerging technology in today's society Speaker recognition is a process of authentication of the claimed of a person from voice characteristics. The main goal of speaker recognition is to extract, classify and identify the information about speaker identity. In this paper shows the usage of Mel Frequency Cepstral Coefficients (MFCC) and Artificial Neural Network for Speaker Recognition System of Kannada Languages. The standard multilingual database is not available experiments are carried out on our own created database of ten speakers. The code developed in MATLAB Environment.

This paper provides a systematic implementation of th e techniques of artificial intelligence for agricultural crop man agement. Agriculture faces many challenges, such as disease a nd infestation of pests, unsuitable soil treatment, inadequate dr ainage and irrigation and many more. Such result in severe crop failure along with environmental hazards caused by excessive chemical use. Several researches were carried out to deal with these issues. With its robust learning capabilities, the fields of artificial intelligence have become a crucial technique for solving various agricultural related problems. Systems are being developed to assist the agricultural experts around the world in finding better solutions. The sector faces numerous challenges to optimize its production, including inadequate soil care, disease and infestation of pests, big data requirements, low output and knowledge gap between farmers and technology. In agriculture the main concept of AI is its versatility, high perf ormance, precision and cost. This paper presents a review of th e soil management, crop management, weed management and disease management applications.

In this paper, we describe an algorithm for estimating heart rate from an optically measured PPG signal when physical exercises are performed. In this case, the PPG signal is contaminated by motion artifacts caused by hand movements, making it difficult to find its fundamental frequency that corresponds to the heart rate. To overcome the noise, a soft decision approach is taken, by which several candidates for the fundamental frequency of the PPG signal are extracted and assigned grades. By appropriate grade weighting, the candidate having the maximal grade is selected. The presented algorithm is of low complexity and shown to provide good results. As such, it can be used in low-power portable devices for real time heart rate estimation.

— An automatic verification of person's identity from its voice is a part of modern telecommunication services. In order to execute a verification task, a speech signal has to be transmitted to a remote server. So, a performance of the verification system can be influenced by various distortions that can occur when transmitting a speech signal through a communication channel. This paper studies an effect of the state of art wideband (WB) speech codecs on a performance of automatic speaker verification in the context of a channel/codec mismatch between enrollment and test utterances. The speaker verification system is developed on GMM-UBM method. The results show that EVS codec provides the best performance over all the investigated scenarios in this study. Moreover, deploying G.729.1 codec in a training process of the verification system provides the best equal error rate in the fully-codec mismatched scenario. Anyhow, differences between the equal error rates reported for all of the codecs involved in this scenario are mostly nonsignificant.

Substantial progress has been achieved in voice-based biometrics in recent times but a variety of challenges still remain for speech research community. One such obstacle is reliable speaker authentication from speech signals degraded by lossy compression. Compression is commonplace in modern telecommunications, such as mobile telephony, VoIP services, teleconference, voice messaging or gaming. In this study, the authors investigate the effect of lossy speech compression on text-independent speaker verification. Voice biometrics performance is evaluated on clean speech signals distorted by the state-of-the-art narrowband (NB) as well as wideband (WB) speech codecs. The tests are performed in both channel-matched and channel-mismatched scenarios. The test results show that coded WB speech improves voice authentication precision by 1–3% of equal error rate over coded NB speech, even at the lowest investigated bitrates. It is also shown that the enhanced voice services codec does not p...

The number of wearable and smart devices which are connecting every day in the Internet of Things (IoT) is continuously growing. We have a great opportunity though to improve the quality of life (QoL) standards by adding medical value to these devices. Especially, by exploiting IoT technology, we have the potential to create useful tools which utilize the sensors to provide biometric data. This novel study aims to use a smartwatch, independent from other hardware, to predict the Blood Pressure (BP) drop caused by postural changes. In cases that the drop is due to orthostatic hypotension (OH) can cause dizziness or even faint factors, which increase the risk of fall in the elderly but, as well as, in younger groups of people. A mathematical prediction model is proposed here which can reduce the risk of fall due to OH by sensing heart rate variability (data and drops in systolic BP after standing in a healthy group of 10 subjects. The experimental results justify the efficiency of the model, as it can perform correct prediction in 86.7% of the cases, and are encouraging enough for extending the proposed approach to pathological cases, such as patients with Parkinson's disease, involving large scale experiments.

Image processing is a process, method, technology which is used to perform some operations on an image, in order to get an upgrade image or to extract some useful information from it. In urban region voting is the major issue, identification of person is most important. Sometime people can use fake voter identity for voting. For avoiding this problem we have to develop a system which can identify the face of person and check in present database if face is match in that point the system can return output as UID of the person. After verification of valid face and UID the voter can enter to vote. Data will be store at only one place, so winning party information can be announced within short time period..

In this paper, we describe an algorithm for estimating heart rate from an optically measured PPG signal when physical exercises are performed. In this case, the PPG signal is contaminated by motion artifacts caused by hand movements, making it difficult to find its fundamental frequency that corresponds to the heart rate. To overcome the noise, a soft decision approach is taken, by which several candidates for the fundamental frequency of the PPG signal are extracted and assigned grades. By appropriate grade weighting, the candidate having the maximal grade is selected. The presented algorithm is of low complexity and shown to provide good results. As such, it can be used in low-power portable devices for real time heart rate estimation.

Multimodal biometric systems that fuse information from several biometrics are most spreads lately because they can overcome problems in unimodal biometric systems. Most of the proposed multibiometric systems offer one level of fusion. In this paper, a comparison between two levels of fusion has been proposed; a proposed fusion system of three biometrics at the feature level based on Particle Swarm Optimization approach (PSO) is presented with a new multi objective fitness function for PSO has been used. Also the score level fusion rule is optimized using (PSO) Particle Swarm Optimization. Results shown that matching score fusion outperforms matching score fusion in one multimodal system (palmprint_Knuckle), while matching score fusion outperforms feature fusion in the other two systems.

Nocturnal frontal lobe epilepsy (NFLE) is distinguished by seizures with composite, often bizarre, fierce behavior arising only or mainly during sleep. These strange seizures and their happening during sleep are often followed by normal electroencephalogram tracings and neurological findings, making it hard to distinguish nocturnal frontal lobe epilepsy seizures from other non-epileptic nocturnal paroxysmal consequences, namely parasomnias. Nocturnal Frontal Lobe Epilepsy was depicted for the first time in 1981, but, as its epileptic origin was contentious, the condition was named nocturnal paroxysmal dystonia. Nocturnal Frontal Lobe Epilepsy is the most difficult syndrome in the sleep related disturbances .This paper restrains an overview of Nocturnal Frontal Lobe Epilepsy, which is a sleeping disorder and the prediction of Nocturnal Frontal Lobe Epilepsy through electroencephalogram in which electrical activity of brain is recorded. Nocturnal frontal lobe can be treated by Anti-seizure drugs.

The electrocardiogram (ECG) is a signal reflecting the electrical activity of the heart muscles. Computer processing of ECG signals has been a very active area of research for decades. All modern recording systems of ECG incorporate compression software, automatic analysis and a guide for interpretation of the signal. The ECG signal is usually modeled by descriptors to facilitate the automatic analysis. We have proposed in this work, the modeling of the characteristic waves of ECG signal by the low degree polynomials, prior to compression and automatic analysis. To this end, we have implemented a complex QRS detection algorithm, and the extraction of the characteristics wave or intervals is carried out semi-automatically. Each segment of the extracted signal is then approximated by polynomials of low degrees, so that the error between the original signal and the approximation polynomial is minimum. Compression consists in keeping a small number of polynomial coefficients to store or transmit the signal, instead of the large amount of samples values. We have obtained very satisfactory results as regards the compression with very high compression ratio and low values of PRD (Percent Difference square root). We have also shown that the coefficients from the polynomial approximations can be used for the automatic analysis of the signal, especially in the estimation of peak values of the waves and determining their concavities. It is possible to establish correlations between the coefficients of the polynomial approximation and the energy of the signal, and also the spectral congestion of the signal. Similarly, tools of intelligent systems such as artificial neural networks and fuzzy logic may be associated with the polynomial coefficients for the automatic interpretation of the ECG signal.
Index Terms— Automated Analysis of ECG, Compression of ECG, ECG Features extraction, Optimal polynomial approximations.

In computer vision, traditional machine learning (TML) and deep learning (DL) methods have significantly contributed to the advancements of medical image analysis (MIA) by enhancing prediction accuracy, leading to appropriate planning and diagnosis. These methods substantially improved the diagnoses of automatic brain tumor and leukemia/blood cancer detection and can assist the hematologist and doctors by providing a second opinion. This review provides an in-depth analysis of available TML and DL techniques for MIA with a significant focus on leukocytes classification in blood smear images and other medical imaging domains, i.e., magnetic resonance imaging (MRI), CT images, X-ray, and ultrasounds. The proposed review’s main impact is to find the most suitable TML and DL techniques in MIA, especially for leukocyte classification in blood smear images. The advanced DL techniques, particularly the evolving convolutional neural networks-based models in the MIA domain, are deeply invest...

Heart Rate Variability (HRV) analysis is the
ability to assess overall cardiac health and the state of the
autonomic nervous system (ANS), responsible for regulating
cardiac activity. ST-change due to ischemia and their HRV
analysis have not been well discussed in the previous works.
The proposed simple and time efficient TBC algorithm has
been tested in four sets of standard databases with selected
patient’s data having ischemic conditions (i.e.MIT-BIH
Normal-Sinus Rhythm Database (NSRDB), European ST-T
Database (EDB), MIT-BIH ST Change Database (STDB) &
Long-Term ST Database (LTSTDB))for the detection of R-peak
& HRV analysis. The pre-processing is done by MAF and DWT
to remove the baseline drift and noise induced in the ECG
signal. The mean/average of HR is calculated for each set of
databases and in case of EDB it is of 57 BPM (subjected to
bradycardia). The Probability with normal distribution is
analyzed by comparing the NSRDB data with the ischemic data sets. The performance of this algorithm is found to be 98.5%.

Speech recognition of disorder people is a difficult task due to the lack of motor-control of the speech articulators. Multimodal speech recognition can be used to enhance the robustness of disordered speech. This paper introduces an automatic speech recognition system for people with dysarthria speech disorder based on both speech and visual components. The Mel-Frequency Cepestral Coefficients (MFCC) is used as features representing the acoustic speech signal. For the visual counterpart, the Discrete Cosine Transform (DCT) Coefficients are extracted from the speaker's mouth region. Face and mouth regions are detected using the Viola-Jones algorithm. The acoustic and visual input features are then concatenated on one feature vector. Then, the Hidden Markov Model (HMM) classifier is applied on the combined feature vector of acoustic and visual components. The system is tested on isolated English words spoken by disorder speakers from UA-Speech data. Results of the proposed system indicate that visual features are highly effective and can improve the accuracy to reach 7.91% for speaker dependent experiments and 3% for speaker independent experiments.

The disorders identified with retina of the eye such as: Diabetic Retinopathy (DR), age-related Macular Degeneration (AMD), Glaucoma, etc. can cause visual impairments. The retinal fundus pictures of the patients can be acquired with a computerized fundus camera. It can then be used as machine learning application against the manual technique for the detection and prevention of DR. In this research, the retinal fundus image obtained from Himalaya Eye Hospital Pokhara, MESSIDOR database and EyePACS are used for the detection of DR along with its severity label. In the preprocessing phase, the images are first converted to grayscale images. Histogram equalization technique is performed to adjust image intensity in order to enhance contrast. Median filtering, a non-linear digital filtering technique, is implemented to remove noise present in the image. It too helps to improve the result for further processing. Blood vessels are detected and optic discs are identified and removed. Exuda...

A Knight Tour problem is an ancient puzzle which remains as a focus of current researcher. The objective of a Knight Tour algorithm is to find out a solution with construct a closed moving path by a knight for visiting each square of a chessboard by only once. This paper proposes a new structural Knight Tour algorithm by four knights moving all together with heuristic preset rules. This new proposed method fulfills any board size of 4n+2 where n>1 with lower execution time and without the needs of any brute force method and backtracking method.

—This paper presents a novel method to detect three types of abnormal Red Blood Cells (RBCs) called Poikilocytes in Iron deficient blood smears. Classification and counting the number of Poikilocyte cells is considered as an important step for the automatic detection of Iron Deficiency Anemia (IDA) disease. Dacrocyte, Elliptocyte and Schistocyte cells are three essential Poikilocyte cells that are prevalent in IDA. The suggested cell recognition approach includes preprocessing, segmentation, feature extraction and classification steps. Classification is done by using three distinct classifiers including Neural Network (NNET), Support Vector Machine (SVM) and K-Nearest Neighbor (KNN) classifiers. Finally, the output of all of the three classifiers are used via Maximum Voting theory to choose the proper class. In maximum voting theory, the class that receives the maximum number of votes is chosen as the final predicted class of a sample cell. In this paper, the accuracy of the proposed method is %99, %97 and %100 for detecting Dacrocyte cells, Elliptocyte cells and Schistocyte cells, respectively.

International Journal of Electrical and communication engineering (IJECE) is an open access peer-reviewed journal that publishes quality articles contributing new results in all areas of Applied Control Systems, Electrical Engineering and Electronics Engineering. The Journal of Electrical and communication engineering (IJECE) is an abstracted and indexed international journal of high quality devoted to the publication of original research papers from Control Systems, Electrical Engineering, Power Systems, Automation, Electronics Engineering, Networks and Communication Systems and their broad range of applications.

Digital images often contain "noise" which takes away their clarity and sharpness. Most of the existing denoising algorithms do not offer the best solution because there are difficulties such as removing strong noise while leaving the features and other details of the image intact. Faced with the problem of denoising, we tried solving it with a Convolutional Neural Network architecture called the "U-Net". This paper deals with the training of a U-Net to remove 3 different kinds of noise: Gaussian, Blockiness, and Camera shake. Our results indicate the effectiveness of U-Net in denoising images while leaving their features and other details intact

Parkinson’s disease (PD) is a neurodegenerative brain disorder that occurs when approximately 60% to 80% of the dopamine-producing cells are damaged. PD is the second common neurodegenerative disorder after Alzheimer. PD could be diagnosed by various signals such as EEG, gait and speech. Approximately, 90 percent of people with PD suffer from speech disorder, thus it might be considered as the easiest way to this aim. This paper investigates a new method for detection of Parkinson form speech signals at which PCA combines extracted features form the data and the classification is done using SVM network. The classification accuracy percent of 91.5 per 3 optimized features is obtained.

Classification of skull fracture is a challenging task for both radiologists and researchers. Skull fractures result in broken pieces of bone, which can cut into the brain and cause bleeding and other injury types. So it is vital to detect and classify the fracture very early. In real world, often fractures occur at multiple sites. This makes it harder to detect the fracture type where many fracture types might summarize a skull fracture. Unfortunately, manual detection of skull fracture and the classification process is time-consuming, threatening a patient's life. Because of the emergence of deep learning, this process could be automated. Convolutional Neural Networks (CNNs) are the most widely used deep learning models for image categorization because they deliver high accuracy and outstanding outcomes compared to other models. We propose a new model called SkullNetV1 comprising a novel CNN by taking advantage of CNN for feature extraction and lazy learning approach which acts as a classifier for classification of skull fractures from brain CT images to classify five fracture types. Our suggested model achieved a subset accuracy of 88%, an F1 score of 93%, the Area Under the Curve (AUC) of 0.89 to 0.98, a Hamming score of 92% and a Hamming loss of 0.04 for this seven-class multi-labeled classification.

Real-time vision is central to many embedded applications (e.g. vehicle guidance). It is a computationally intensive task well beyond current general purpose computing platforms such as PCs and workstations. Thus, most real time vision systems need special high performance computing platforms, commonly provided in the form of parallel processing engines or dedicated hardware. The proposed architecture uses new generations of re-programmable logic devices and modularised hardware, thereby gaining the performance advantage of hard-wired logic with the flexibility and associated economies of programmable systems. The architecture takes the form of an extensive processing hierarchy consisting of a set of tightly coupled parallel processors, each processing a portion of the image using a classic pipeline arrangement. A programmable image splitting (and reconstruction) engine feeds this array and offers the potential of further enhancing the performance of the engines by restructuring the...

The ability of Minkowski Functionals to characterize local structure in different biological tissue types has been demonstrated in a variety of medical image processing tasks. We introduce anisotropic Minkowski Functionals (AMFs) as a novel variant that captures the inherent anisotropy of the underlying gray-level structures. To quantify the anisotropy characterized by our approach, we further introduce a method to compute a quantitative measure motivated by a technique utilized in MR diffusion tensor imaging, namely fractional anisotropy. We showcase the applicability of our method in the research context of characterizing the local structure properties of trabecular bone micro-architecture in the proximal femur as visualized on multi-detector CT. To this end, AMFs were computed locally for each pixel of ROIs extracted from the head, neck and trochanter regions. Fractional anisotropy was then used to quantify the local anisotropy of the trabecular structures found in these ROIs and to compare its distribution in different anatomical regions. Our results suggest a significantly greater concentration of anisotropic trabecular structures in the head and neck regions when compared to the trochanter region (p < 10-4). We also evaluated the ability of such AMFs to predict bone strength in the femoral head of proximal femur specimens obtained from 50 donors. Our results suggest that such AMFs, when used in conjunction with multi-regression models, can outperform more conventional features such as BMD in predicting failure load. We conclude that such anisotropic Minkowski Functionals can capture valuable information regarding directional attributes of local structure, which may be useful in a wide scope of biomedical imaging applications.

ECG signal is a biological non-stationary signal which contains valuable information about rhythms of heart.
Amplitude and duration  of ECG signal can be buried by various types of noise, especially, power line
interference, which sometimes leads to misdiagnosis. This paper presents a ECG signal denoising based on  the
wavelet packet transform and two adaptive filters, such as, normalized least-mean-square (NLMS) and
recursive-least-square (RLS), and the results are compared with a conventional notch filter both in time and
frequency domain. The results are evaluated by examining power spectral density (PSD), spectrogram and
magnitude squared coherence (MSC) analyses. SNR, %PRD, MSE and ESD performance parameter are also
estimated. The simulation is done by using Signal Processing Toolbox built in MATLAB
®
and, the simulation
result shows that wavelet packet transform is an excellent method for denoising the ECG signal.

El número, tamaño y distribución de células sanguíneas proporciona información importante para el diagnóstico médico. Los tres tipos de células presentes en la sangre son: eritrocitos (glóbulos rojos), leucocitos (glóbulos blancos) y trombocitos (plaquetas).
Entre las técnicas existentes para el análisis de estas células se encuentra el método manual o convencional y los equipos automatizados. A pesar de su utilidad, estos equipos representan un alto costo en mantenimiento y consumo de reactivos para los centros de diagnóstico, además, no permiten la observación celular. Para ayudar a solucionar parte de este problema, se hizo una revisión del estado del arte de las técnicas de procesamiento digital de imágenes utilizadas en microscopía, para aplicarlas en la detección de eritrocitos, ya que éstos, en conjunto con la hemoglobina y el hematócrito, se utilizan para diagnosticar normalidad, anemia o policitemia en pacientes. Se presentan los algoritmos encontrados en la literatura, y el método propuesto a partir de las técnicas seleccionadas para finalmente llegar a la detección y conteo de eritrocitos.

Edge detection places a vital role in image processing. It is one of the techniques used to detect the intensity value and sharpness of the image. Variety of edge detection techniques is available. In this paper we use Canny „sedge detection technique. This technique is more advanced and optimal when compared to other techniques. Canny‟s method gives better output in noise presence also. This method is used for finding detection and extraction from different images. This method works on the basis of an algorithm named as Canny‟s algorithm. The method gives low error, minimal response and good localization.

— This paper is a design of non-invasive instrument for measuring the glucose level in the body using saliva acetone. In this method, the electric current which is equivalent to the acetone level present in the body is converted into voltage using the sensor. This voltage has a negative correlation with the blood glucose level. Thus, it is converted into glucose level by programming done using ATMEGA328 processor which is displayed in LCD. This is a painless procedure with low energy consumption. It is a handheld device and it is cost effective.

Electrocardiogram (ECG) is commonly used in medical field in order to diagnose different cardiovascular diseases and patient's general health. ECG Signal are often affected by artifacts from the environment including baseline wander, power line interference and electromyographic (EMG) noise. In this project, ECG Signal to be processed is collected from MIT/BIH (Massachusetts Institute of Technology-Beth Israel Hospital) Arrythmia Database [1]. This method is being tested on one dataset only and serves as a beginner guide to digital signal processing with MATLAB for an affordable, precise, and portable Electrocardiogram device that may be used at underserved hospitals. With MATLAB, to remove the baseline wander, Detrend function can be used to remove low frequency below 0.5 Hz. Powerline interference (50 Hz or 60 Hz) can be filtered out by using bandstop IIR filter. EMG and high frequency noise can be reduced by using IIR low-pass filter to cut off high frequency signal. The resulting signal is filtered out successfully with less noise and P wave, QRS complex, T wave can be inferred properly.

Elastography is a promising technique for detection
and classification of abnormal growth of tumors in soft tissues. In this paper we propose a simple yet effective optimization to speed up the time-domain indirect strain estimation. The proposed method exploits the displacement values of the neighbors to predict the displacement and utilizes this predicted value to define an adaptive search region for finding the best match between pre and post compression signals. This Proposed method is faster than other block matching algorithms such as SAD, SSD or cross-correlation. We have evaluated the method using performance parameters such as elastrographic Signal to Noise Ratio (SNRe), elastrographic Contrast to Noise Ratio (CNRe), elastrographic Peak-Signal to Noise Ratio (PSNRe) and Mean Structural Similarity (MSSIM). For lower percentage of strain, the proposed method superior in computational speed demonstrates similar performance to the other methods.

This research presents a pulse-by-pulse analysis of Oscillometric blood pressure waveform at systolic, diastolic and mean arterial pressure points.
Using a mathematical optimization technique, pulses are characterized into component harmonic by minimizing the least square error. The results at the important pressure points are analyzed and compared for different subject using different waveform extraction techniques.
Blood pressure is estimated using the harmonic parameters. The approach studies changes in the parameters as oscillometric blood pressure recording is done. 8 harmonic parameters are obtained from the pulse characterization and are used to estimate Systolic arterial Blood Pressure, Mean arterial Blood Pressure, and Diastolic arterial Blood Pressure. The estimates are compared with our reference value to determine which has the best agreement. The proposed method is further compared with Maximum Amplitude Algorithm and Pulse Morphology Algorithm.
The effect of oscillometric waveform extraction methods on the proposed method is observed. The experiment established the fact that the extraction technique can alter the shape of oscillometric pulses. The methods were compared and it was observed that the used extraction methods did not make any significant difference on the accuracy, using this technique.

Escherichia coli (abbreviated as E. coli) is a common bacterium both in daily life and scientific researches. In order to demonstrate the biophysical properties of Escherichia coli, the 3D model and simulation of E. coli are built based on Blender, helping with the visualization. First, a brief introduction to Blender explains why it's chosen as the modelling software and demonstrates the advantages of Blender by comparison; next, the whole process of how the model is built through basic shapes and modifiers is provided step by step. Lastly, several C# scripts in Unity are written to make the model move as real Escherichia coli in a low Reynolds number environment to simulate the behaviour of the cell in line with a given motion mode. The final result manifests the above methods have a great performance in the construction and simulation of E. coli, which could offer ideas for further study of the simulation of the microorganism.

In this paper, we present a new complex electronic system for facilitating communication with severely disabled patients and telemonitoring their physiological parameters. The proposed assistive system includes three subsystems (Patient, Server, and Caretaker) connected to each other via the Internet. The two-way communication function is based on keywords technology using a WEB application implemented at the server level, and the application is accessed remotely from the patient’s laptop/tablet PC. The patient’s needs can be detected by using different switch-type sensors that are adapted to the patient’s physical condition or by using eye-tracking interfaces. The telemonitoring function is based on a wearable wireless sensor network, organized around the Internet of Things concept, and the sensors acquire different physiological parameters of the patients according to their needs. The mobile Caretaker device is represented by a Smartphone, which uses an Android application for communicating with patients and performing real-time monitoring of their physiological parameters.