An optimized design of seizure detection system using joint feature extraction of multichannel EEG signals

The detection of seizure onset and events using electroencephalogram(EEG) signals are important tasks in epilepsy research.The literature available on seizure detection has discussed the implementation of advanced signal processing algorithms using tools accessed over the cloud.However,seizure monitoring application needs near sensor processing due to privacy and latency issues.In this paper,a real time seizure detection system has been implemented using an embedded system.The proposed system is based on ensemble empirical mode decomposition(EEMD) and tunable-Q wavelet transform(TQWT) algorithms.The analysis and classification of non-stationary EEG signals require the wavelet transform with high Q-factor.However,direct use of TQWT increases the computational complexity of feature extraction from multivariate EEG signals.In this paper,the first step is to process the signal by using EEMD to obtain 8 intrinsic mode functions(IMFs).The Kraskov(KraEn),sample(SampEn),and permutation(PermEn) entropy features of IMFs are extracted and based on optimum values,and 4 IMFs are decomposed using TQWT.Secondly,centered correntropy(CenCorrEn) features of the 1^(st)and 16^(th) sub-band of TQWT have been used as classifier inputs.The performance of multilayer perceptron neural networks(MLPNN),least squares support vector machine(LSSVM),and random forest(RF) classifiers has been tested on the multichannel EEG data recorded from a local hospital.The RF classifier has produced the highest accuracy of 96.2% in classifying the signals.The proposed scheme has been employed in developing an embedded seizure detection system to assist neurologists in making seizure diagnostic decisions.

Performance analysis of cow dung as an eco-friendly additive material for sustainable moulding and casting

In the present work,an attempt is made to partially replace high cost silica sand with sustainable eco-friendly cow-dung.The practical utility of cow dung as an additive material that facilitates binding in sand moulds has been tested.The Taguchi method is used to plan and conduct nine experiments with three replicates each.Pareto analysis of variance study is performed to understand the practical significance of moulding sand variables,that is,percentage of cow dung,clay,water,and degree of ramming on sand mould properties.The conflicting multiple objective functions(maximize mould hardness,and minimize collapsibility and gas evolution)are optimized by utilizing the Data Envelopment Analysis Ranking(DEAR)method.The optimal parameters i.e.6%of clay,4%of water,5%of cow-dung and 4 numbers of ramming strokes are obtained by applying a hybrid Taguchi-DEAR method.These parameters show the best moulding properties are:mould hardness 55,gas evolution 5.9 ml·g-1,and collapsibility 470 g·cm-2.Thereafter,a Lovejoy coupling made of aluminium is cast in the sand mould prepared with cow dung and without cow dung,with the optimum set of parameters.The mechanical properties,surface finish,and microstructure of the casting cast with cow-dung are better than that without cowdung.

Surface Water Quality Assessment of Panchagnaga River and Development of DO-BOD Relationship Using Empirical Approach

Surface water samples were collected from selected locations along river Panchaganga,from Kolhapur to Narsobawadi during April 2019.Physicochemical parameters were determined in the laboratory and chemical mass balance approach was adopted to estimate the individual ionic loads in the river water.Streeter-Phelps equation was applied to derive a relationship between DO and BOD_(5).Model parameters such as De-oxygenation Rate(K_(d))and Re-aeration Rates(K_(r))were optimized using different empirical methods.The result of chemical mass balance showed an increase in the loading of various ions from upstream to downstream which could be attributed to agricultural and industrial wastes that enter the main stream.De-oxygenation rate and re-aeration constants were calculated using various empirical methods.DO sag curve was developed using Streeter Phelp’s model and compared with the observed parameters which showed a significant correlation.DO-BOD concentration observed along the course of the river indicated that the self-purification capacity of the river is high due to which the river regains the lost DO level at a distance less than 50 meters.

Dual-tree complex wavelet transform and super-resolution based video inpainting application to object removal and error concealment

Video inpainting is a technique that fills in the missing regions or gaps in a video by using its known pixels.The existing video inpainting algorithms are computationally expensive and introduce seam in the target region that arises due to variation in brightness or contrast of the patches.To overcome these drawbacks,the authors propose a novel two-stage framework.In the first step,sub-bands of wavelets of a low-resolution image are obtained using the dualtree complex wavelet transform.Criminisi algorithm and auto-regression technique are then applied to these subbands to inpaint the missing regions.The fuzzy logic-based histogram equalisation is used to further enhance the image by preserving the image brightness and improve the local contrast.In the second step,the image is enhanced using super-resolution technique.The process of down-sampling,inpainting and subsequently enhancing the video using the super-resolution technique reduces the video inpainting time.The framework is tested on video sequences by comparing and analysing the structural similarity index matrix,peak-signal-to-noise ratio,visual information fidelity in pixel domain and execution time with the state-of-the-art algorithms.The experimental analysis gives visually pleasing results for object removal and error concealment.

Self-health Monitoring and Reporting System for COVID-19 Patients Using CAN Data Logger

In the evolving situation of highly infectious coronavirus,the number of confirmed cases in India has largely increased,which has resulted in a shortage of health care resources.Thus,the Ministry of Health and Family Welfare-Government of India issued guidelines for the‘Home isolation of COVID-19 positive patients’methodology for asymptomatic patients or with mild symptoms.During home isolation,the patients are required to monitor and record the pulse rate,body temperature,and oxygen saturation three times a day.This paper proposes a system that can request data from the required sensor to measure the pulse rate,body temperature,or oxygen saturation.The requested data is sensed by the respective sensor placed near the patients’body and sent to the CAN data logger over the CAN bus.The CAN data logger live streams the sensor values and stores the same to an excel sheet along with details like the patient’s name,patient’s age,and date.The physicians can then access this information.

A Study on Automatic Latent Fingerprint Identification System

Latent fingerprints are the unintentional impressions found at the crime scenes and are considered crucial evidence in criminal identification.Law enforcement and forensic agencies have been using latent fingerprints as testimony in courts.However,since the latent fingerprints are accidentally leftover on different surfaces,the lifted prints look inferior.Therefore,a tremendous amount of research is being carried out in automatic latent fingerprint identification to improve the overall fingerprint recognition performance.As a result,there is an ever-growing demand to develop reliable and robust systems.In this regard,we present a comprehensive literature review of the existing methods utilized in latent fingerprint acquisition,segmentation,quality assessment,enhancement,feature extraction,and matching steps.Later,we provide insight into different benchmark latent datasets available to perform research in this area.Our study highlights various research challenges and gaps by performing detailed analysis on the existing state-of-the-art segmentation,enhancement,extraction,and matching approaches to strengthen the research.

Unmanned Drug Delivery Vehicle for COVID-19 Wards in Hospitals

The prime reason for proposing the work is designing and developing a low-cost guided wireless Unmanned Ground Vehicle(UGV)for use in hospitals for assistance in contactless drug delivery in COVID-19 wards.The Robot is designed as per the requirements and technical specifications required for the healthcare facility.After a detailed survey and tests of various mechanisms for steering and structure of UGV,the best mechanism preferred for steering articulated and for body structure is hexagonal as this approach provides decent performance and stability required to achieve the objective.The UGV has multiple sensors onboard,such as a Camera,GPS module,Hydrogen,and Carbon Gas sensor,Raindrop sensor,and an ultrasonic range finder on UGV for the end-user to understand the circumferential environment and status of UGV.The data and control options are displayed on any phone or computer present in the Wi-Fi zones only if the user login is validated.ESP-32 microcontroller is the prime component utilized to establish reliable wireless communication between the user and UGV.These days,the demand for robot vehicles in hospitals has increased rapidly due to pandemic outbreaks as using this makes a contactless delivery of the medicinal drug.These systems are designed specifically to assist humans in the current situation where life can be at risk for healthcare facilities.In addition,the robot vehicle is suitable for many other applications like supervision,sanitization,carrying medicines and medical equipment for delivery,delivery of food and used dishes,laundry,garbage,laboratory samples,and additional supply.