Ganesan M.

Patient monitors with arrhythmia detection will enhance the quality of living of human by aiding in prediction of diseases in much early stage. In this work we have developed an algorithm for classifying the ECG signals into normal and arrhythmic signal. Here we have detected the R peaks from denoised ECG signal with an accuracy of 97.56%. Extracted features from the signal in both time and frequency domain and the signals are classified into normal and abnormal signals using support vector algorithm. The accuracy of the algorithm is tested by applying on MIT-BIH arrhythmia database and we obtained an overall 80% classifier accuracy.

More »»

Food production techniques have to be improved because of rapid demand in food. Since India has agriculture as the main source of production, proper irrigation schemes are to be employed for an efficient outcome. This design automates the irrigation process thereby reducing the manual intervention and the water losses. It is more helpful in the places where water scarcity is seen more. It consists of 2 sensors which takes the values of temperature of surroundings and moisture level of soil. Output of these sensors are given to ADC and then to microcontroller. Microcontroller compares the values with the threshold values and drives the relay which controls the motor. LCD display is used to display data in the field. GSM modem is also a part of design which helps in transmitting SMS to farmer's phone number which contains the status of motor in field. The design is cost effective and also affordable.

More »»

Unwanted signal contents always degrades the quality of ECG signal. Since ECG is a non-stationary signal, noise removal always is a complicated task. ECG signal is a raw material for diagnosis and analysis of almost all heart diseases and hence demands a good quality. A survey of various types of various denoising approaches emerged over recent years has been presented in this paper. FIR and IIR filtering, low and high frequency noise removal techniques, Quadrature filtering, Adaptive noise cancellation techniques, Non-Local Means denoising techniques (NLM), Empirical Mode Decomposition(EMD), Variational Mode Decomposition (VMD), Wavelet transform denoising methods and recent developments are discussed along with comparatives studies.

More »»

An ECG is a sensitive diagnostic tool used to detect various cardio-vascular diseases by measuring and recording the electrical activity of the heart. The efficiency and speed of the feature extraction scheme has a major role in the ECG diagnostic system. The proposed work tries to develop an ECG feature extraction system based on the multi-resolution wavelet transform. This system tries to improve the performance of ECG analysis system by extracting highly accurate ECG features. The Daubechies wavelet filter is used here for extracting ECG features. MIT-BIH Arrhythmia database is used in this work to obtain the digitized ECG input signal.

More »»

Non-gated images (in the absence of external trigger) cannot reconstruct 3D volumes directly due to mismatch of acquisition intervals. Such a type of image needs synchronization for making direct reconstruction into 3D volume from 2D data set. Since, Spatio-temporal resolution is more important when dealing with reconstruction methods and noise reduction techniques, often wavelets are used. To reduce motion artifacts, wavelet denoising is performed. Wavelet correlation is performed on image sequences for synchronization. After correlating all image sequences, reconstruction will be performed to get 4D imaging. Comparison is made for different synchronization methods with and without wavelets by finding their root mean square error and peak signal-to-noise ratio

More »»

Various problems associated with epilepsy detection is that the epileptic spike essentially change from one patient to the other and we are in need of trained professional to classify normal brain activity, where the non-pathological events that resemble pathological ones. The aim of this work is the automatic detection of Epileptic and non-Epileptic spike in EEG which plays a vital role in the determination of epilepsy. The present study proposes a system that integrates wavelet transform, Feature extraction and Artificial Neural Network for the detection and classification of Epilepsy. The system was evaluated on testing data from 25 patients of which 86.0% of the epileptic spikes and 80% of the non-epileptic spikes were detected correctly. This system has good performance in detecting epileptic activities and it is found that the Wavelet based Artificial Neural Network approach is an appropriate way of detecting the epileptic spikes in EEG.

More »»

Ballistocardiogram (BCG) a promising technique that records the vibrational movement of the body related to heartbeat of a person. In this paper, a electromechanical film (Emfi) sensor and Electrocardiogram (ECG) set up was used to measure the force of contraction by which the health condition of a person was indexed. The experimental setup was carried out with a base Emfi sensor and a standard Lead I ECG electrode system on a flat chair. Wavelet denoising was performed for both ECG and BCG to evaluate the QRS complex of ECG relevant to IJK complex of BCG. Force of contraction is the amount of blood that heart can pump. IJ peak of BCG corresponds to force of contraction. Based on physical activity, 20 subjects were selected for experimental setup, 10 normal adult subjects and 10 athlets. From the BCG readings, the IJ amplitude level, which is related to the average force of contraction, was calculated as 3.98mV for normal subject and 6.21mV for athletic subject. © 2016 IEEE. More »»

This paper aims at providing efficient and accurate analysis of the cardiac activity of patient acquired from Electrocardiogram and Ballistocardiogram (BCG) signals using discrete multiwavelets de-noising analysis. The functioning of heart is studied from the Electrocardiogram (ECG), a noninvasive technique recorded by using skin electrodes and from BCG, an electrode-less technique that calculates the fitness of the heart. The setup of BCG and ECG was successfully made. The corresponding signals acquired, to calculate the cardiac output with respect to the parameters like R peak of ECG, J peak of BCG and R-J interval. Efficient analysis of ECG and BCG were done after removing the baseline drift and power line interference. Coiflet, Daubechies wavelets and multiwavelets were used to study the performance for better denoising and pre- processing to extract the features. Multiwavelet was found to be a better choice for finding the optimum performance of heart activity. 20 subjects (Normal life style adults and athletes) recordings have been taken for study and analysis on heart rate and cardiac output were made for each subject. The analyses were made 2 times a day (Morning and Night) to determine the index in evaluating the fitness of the heart. Cardiac output of athletes was found to be better than normal subjects. More »»

Ballistocardiogram (BCG) is a non-invasive technique to measure cardiac parameters. It was popularized by Dr. Isaac Staar in 1940. The Ballistocardiogram signal is generated due to the vibrational activity of the heart. BCG was considered as a promising technique but was replaced by Electrocardiogram due to the difficulty involved in detecting and analysing the BCG waveforms. With the increase in processing power and better signal processing techniques over the last few decades, BCG has regained its prominence and is being considered to be used as a continuous patient monitoring system. The usability of BCG was limited in the earlier days due to the large size of the equipment and the lack of signal processing systems to analyse this complicated signal. Cardiac output is defined as the amount of blood pumped out by the heart in a minute. This parameter can be utilized to determine the state of the heart. One method to determine the cardiac output from BCG waveform has been discussed in section II of this paper. The sensor used for our experiment is a lightweight and flexible sheet type electromechanical film which is placed on the seat of the chair. The setup used has a two-stage amplifier which is connected to a data acquisition card which is in turn connected to a laptop. The signal processing is done using NI’s software LabView. The BCG setup was made and the signal was successfully validated with ECG. The R-J interval, which is the interval between the R peak of the ECG signal and J peak of the BCG signal, was determined. Echocardiogram, another cardiac measurement instrument, was kept as a standard basis for determining the cardiac output from the BCG signal. Recording of 14 different subjects have been taken and the cardiac output has been determined for each case. More »»