Search

Medium-frequency (mid-frequency) vibration analysis of complex structures plays a prominent role in automotive, aerospace, mechanical, and civil engineering. Flexible beam structures displayed by the traditional Euler–Bernoulli pillar hypothesis have been broadly utilized in different designing issues. A kinematic speculation made in the Euler–Bernoulli shaft hypothesis is that the plane areas of a pillar typical to its impartial pivot remain planes after the bar encounters twisting disfigurement, which ignores shear deformity. The expanded DTFM decides the recurrence reaction of a bar structure in a careful and insightful structure, in any recurrence locale covering low, center, or high frequencies. In the mean time, the proposed technique gives the nearby data of a bar structure, like uprooting, shear misshapening, bowing second and shear power at any area, which in any case would be extremely challenging with energy-based strategies. amplifiers by and large accept that a confined main thrust can spread energy uniformly across the outside of a board. Notwithstanding, examinations have shown that board vibrations stay confined around the driving point at high frequencies, Energy spreading will possibly happen when the board is incited in a recurrence district with a low thickness of modes, as numerous modes impelled together will join to frame a band-restricted delta work at the area of the main impetus. A quantitative proportion of limitation is presented, in light of the proportion of the energy contained in a little locale around the driving point to the energy contained in the whole board. Analyses show that the limitation transcends a basic recurrence, and this basic recurrence connects well with forecasts of the great recurrence plate district utilizing standard modular cross-over measurements.

Increase in energy requirement has lead to the development of new energy storage devices for the movement of vehicles. This paper categorises the various types of vehicles with different intensities of hybrids based on degree of hybridisation. The parameters which depend on the performance of the battery are shallow cycle life, high dynamic charge acceptance for sturdy for regenerative braking and sturdy life of the battery. Lead Acid battery either liquid or absorptive glass fibre is expected to work with 14 V systems for micro hybrids with low cost battery maintenance system for increasing applications.. Advanced latest batteries are considered for mild or medium hybrid once they have proven the worthiness under real operating conditions under dynamic responses. NiMH and Li-ion batteries are dominating current and potential battery for high functionality HEVs. Li -ion offers an attractive opportunity for improvement in performance and cost. Super capacitors are considered for power applications where in the variation of signal is in the pulse form. Specific attention is given to system integration of the battery and advance technology during vehicle integration. In this paper there are various opportunities and challenges in obtaining a potential battery pack.

Though DBS is suiting well to Parkinson`s still its current version being modified now and often as suggested by the neuroscientists. The technique has gained an effectual management surgical therapy concerned to PD, explicitly whilst growing idea as an effectual method to alleviate Parkinson’s disease and other movement disorders. The present ver.1 indirect DBS is an open loop based and its parameters are changeable manually and there is no provision to adjust automatically or online based on Parkinson diseased behavior and performance. Hence, supervised classification of patient behavior is a major and significant step towards the design of next generation DBS systems which are adaptive closed loop based. The work in this study demonstrates a supervised classification machine learning (CML, i.e., multiple kernel learning M-K-L) method to distinguish such cognitive behavioral tasks by using the subthalamic nuclei (STN) biomarkers, i.e., biomedical data of microelectrode recording (MER) bio signals (or local field potential LFP). We applied the time domain and frequency domain representation spectrograms of the raw data acquired from right and left hemisphere brain`s STNs as the feature vectors. Following the feature extractions, we combined those features via support vector machines (SVMs) with complex multifaceted root learning, i.e., C-M-L or multi kernel learning (M-K-L) formulation. The C-M-L based classification techniques were applied to a class and categorize different tasks such as switch (push-pull button), movement of jaws, vocalizations, plus movement of arm due to the tremor. Our experiments show that the l - n o r m C-M-L/M-K-L radically smash distinct kernel SVM-based classifiers in classifying behavioral tasks of five subjects even using signals acquired with a low sampling rate of 10 Hz. This leads to a lower computational cost. Keywords: Biomarker, Deep Brain Stimulation, Local Field Potential, Complex Root/Multi Kernel Learning, Time domain

Microelectrode recording (MER) or microelectrode signals recording of local field potentials by means of subthalamic-nuclei deep brain stimulation is highly successful for construing or deducing Parkinson disease (PD) signal analysis acquiescent to elucidation are fetching ever more germane. These signals are supposed to emulate STN neurons action-potential movement and, these potential frequency modulations are coupled to spiking-events. The method uses auto regression stochastic (random nature) model machine learning approach and, other standard techniques as of system identification field. A usual conventional local field potential implication involves computing spectral-densities, i.e., power (P S Ds) of these signals—waveforms, that confines power on different frequencies. But, P S D s is second-order statistics might not confine non trivial temporal-dependencies which subsist in unprocessed data. Hence, we suggest L F Ps technique which is valuable in support of relating or unfolding distinctive plus sole features of temporal-dependencies in L F P waveforms. This technique is derived as of auto-regression (A R) modeling originating as of the systems plus control theory in fastidious systems-identification. We have distinctively, built A R models of L F P movement activity plus inferred, and also verified statistically major differentiations in temporal-dependencies among damaged tissues of nuclei plus protective areas in Parkinson`s receiving innocuous microelectrodes via deep brain stimulator (D BS). Differentiations in spectral-densities of field-waveforms amid the two conglomerates were not statistically-significant. Keywords: Microelectrode recording (MER), Parkinson disease (PD).

For any industry, an important part of quality control is the detection and identification of defects of the products. During the manufacturing process, a wide range of defects occur on or in the p...

The main objective of image fusion for multimodal medical images is to retrieve valuable information by combining multiple images obtained from various sources into a single image suitable for better diagnosis. In this paper, a detailed survey on various existing medical image fusion algorithms, with a comparative discussion is presented. Image fusion algorithms available in the current literature are categorized into various methods known as (1) morphological methods, (2) human value system operator based methods, (3) sub-band decomposition methods, (4) neural n++32w1etwork based methods, and (5) fuzzy logic based methods. This research concludes that even though there exist a few open-ended creative and logical difficulties, the fusion of medical images in many combinations assists in utilizing medical image fusion for medicinal diagnostics and examination. There is a tremendous progress in the fields of deep learning, artificial intelligence and bio-inspired optimization techniques. Effective utilization of these techniques can be used to further improve the efficiency of image fusion algorithms.

This paper aims to develop a novel deep learning concept to deal with video inpainting. Initially, motion tracking is performed, which is the process of determining motion vectors that describe the transformation from adjacent frames in a video sequence. Further, the regions or patches of each frame are categorized using the optimized recurrent neural network (RNN), in which the region is split into a smooth and structure region. It is performed using the texture feature called grey-level co-occurrence matrix. The filling of the smooth region is accomplished by replacing with the mean pixels of unmasked region, and the structure region is done by optimized patch matching approach based on scale-invariant feature transform (SIFT). The main objective optimized patch matching is based on the minimized Euclidean distance between the extracted SIFT features of the original patch and reference patch, and the certain patch is inpainted by the optimized patch. Here, the hybridization of two meta-heuristic algorithms like cuckoo search algorithm and multi-verse optimization (MVO) called Cuckoo Search-based MVO is used to optimize the RNN and patch matching. Finally, the experimental results verify the reliability of the proposed algorithm over existing algorithms.

Background/Objectives: There is only limited research work is going on in the field of facial expression recognition on low resolution images. Mostly, all the images in the real world will be in low resolution and might also contain noise, so this study is to design a novel convolutional neural network model (FERConvNet), which can perform better on low resolution images. Methods: We proposed a model and then compared with state-of-art models on FER2013 dataset. There is no publicly available dataset, which contains low resolution images for facial expression recognition (Anger, Sad, Disgust, Happy, Surprise, Neutral, Fear), so we created a Low Resolution Facial Expression (LRFE) dataset, which contains more than 6000 images of seven types of facial expressions. The existing FER2013 dataset and LRFE dataset were used. These datasets were divided in the ratio 80:20 for training and testing and validation purpose. A HDM is proposed, which is a combination of Gaussian Filter, Bilateral Filter and Non local means denoising Filter. This hybrid denoising method helps us to increase the performance of the convolutional neural network. The proposed model was then compared with VGG16 and VGG19 models. Findings: The experimental results show that the proposed FERConvNet_HDM approach is effective than VGG16 and VGG19 in facial expression recognition on both FER2013 and LRFE dataset. The proposed FERConvNet_HDM approach achieved 85% accuracy on Fer2013 dataset, outperforming the VGG16 and VGG19 models, whose accuracies are 60% and 53% on Fer2013 dataset respectively. The same FERConvNet_HDM approach when applied on LRFE dataset achieved 95% accuracy. After analyzing the results, our FERConvNet_HDM approach performs better than VGG16 and VGG19 on both Fer2013 and LRFE dataset. Novelty/Applications: HDM with convolutional neural networks, helps in increasing the performance of convolutional neural networks in Facial expression recognition. Keywords: Facial expression recognition; facial emotion; convolutional neural network; deep learning; computer vision

Transarterial radioembolization (TARE) with yttrium-90 (90Y) microspheres is a liver-directed treatment for primary and secondary hepatic malignancies. Personalized dosimetry aims for maximum treatment effect and reduced toxicity. We aimed to compare pre-treatment voxel-based dosimetry from 99mTc macroaggregated albumin (MAA) SPECT/CT with post-treatment 90Y PET/CT for absorbed dose values, and to evaluate image quality of 90Y SiPM-based PET/CT. Forty-two patients (28 men, 14 women, mean age: 67 ± 11 years) with advanced hepatic malignancies were prospectively enrolled. Twenty patients were treated with glass and 22 with resin microspheres. Radiation absorbed doses from planning 99mTc-MAA SPECT/CT and post-therapy 90Y PET/CT were assessed. 90Y PET/CT images were acquired for 20 min and reconstructed to produce 5-, 10-, 15-, and 20-min datasets, then evaluated using the 5-point Likert scale. The mean administered activity was 3.44 ± 1.5 GBq for glass and 1.62 ± 0.7 GBq for resin microspheres. The mean tumor absorbed doses calculated from 99mTc-MAA SPECT/CT and 90Y PET/CT were 175.69 ± 113.76 Gy and 193.58 ± 111.09 Gy (P = 0.61), respectively for glass microspheres; they were 60.18 ± 42.20 Gy and 70.98 ± 49.65 Gy (P = 0.37), respectively for resin microspheres. The mean normal liver absorbed doses from 99mTc-MAA SPECT/CT and 90Y PET/CT were 32.70 ± 22.25 Gy and 30.62 ± 20.09 Gy (P = 0.77), respectively for glass microspheres; they were 18.33 ± 11.08 Gy and 24.32 ± 15.58 Gy (P = 0.17), respectively for resin microspheres. Image quality of 90Y PET/CT at 5-, 10-, 15-, and 20-min scan time showed a Likert score of 3.6 ± 0.54, 4.57 ± 0.58, 4.84 ± 0.37, and 4.9 ± 0.3, respectively. 99mTc-MAA SPECT/CT demonstrated great accuracy for treatment planning dosimetry. SiPM-based PET/CT scanner showed good image quality at 10-min scan time, acquired in one bed position. A PET/CT scan time of 5 min showed acceptable image quality and suffices for dosimetry and treatment verification. This allows for inclusion of 90Y PET/CT in busy routine clinical workflows. Studies with larger patient cohorts are needed to confirm these findings.

Abrasive water jet machining is a non-conventional machining process that is used in automobile, aircraft, chemical processing equipment, marine, spacecraft and sporting goods industries related applications. There has been a growing interest in using composite materials in place of conventional materials due to their unique properties, such as extremely high strength to weight ratio and corrosion resistance. As a result, these materials are being used in aerospace and aeronautical structural applications. Abrasive water machine has capability to cut almost plethora of materials owing certain advantages such as no additional tools are required, no cleaning is required as it self-cleaned, fast set-up, etc. because of this characteristic’s, the attention of AWJM has attracted industries over unconventional machines. Glass fibre which covers under polymer matrix composite is considered in this experimental work. This paper evaluates an experimental study on three different thickness of laminated plates, pressures, stand-off distance and traverse rate are different parameters used for design of experiments. Top kerf width, bottom kerf width and kerf taper are the machining process responses which has been evaluated using analysis of variance (ANOVA) technique. Delamination studies have been done using Scanning Electron Microscope (SEM).

Heat dissipation plays progressively more significant role in several fields such as high-density gadgets, airplane microelectronics devices, advanced space crafts, thermal power plants, turbo-charger, due to the swiftly increasing thermal loads affects the reliability of components and limits its life span. With high stiffness, strength and low coefficient of thermal expansion (CTE), carbon fiber strengthened polymer composites have been broadly utilized in numerous engineering applications. Therefore, in the current material investigation, thermal characterizations were made on polyethersulfone (PES) material compounded with short carbon fiber (SCF) by four different weight fraction via 0, 10, 20 and 30 wt%. Specific heat (cp) was determined through differential scanning calorimeter (DSC) and thermal conductivity was measured by transient line heater method. Specific heat of PES/SCF composites decreased and conductivity was increased with the higher loading of SCF. Thermal diffusivity was in turn computed by using the relation between specific heat and thermal conductivity. In addition, numerical analysis was performed in Abaqus finite element package to simulate the transient line heater method. Temperature profiles of the composites extracted by numerical simulation matched with that of experimental temperature profile.

The acoustic levels generated by ship propeller are very much important in the design and development of the propeller from stealth technology point of view. The KCS SVA 1193 model of propeller geo...

Congestion is an important concern in both wired and wireless communication systems. Several technologies are in use for the data link, network, and transport layer to manage congestion. Conventional congestion control approaches are unreliable for transport layer protocols by means of congestion control mechanisms utilized for multimedia streaming. Packets that schedule flow control to employ multiple network interfaces to smoothly transmit streaming data play an important role in congestion control by overwhelming the intermediate node’s buffer and causing packet loss. In this paper, we propose a Multipath Packet Scheduling (MP-PS) approach to efficiently stream the data packets in multiple paths. It consists of a Multi-Path Packet Scheduler (MPS) to perform the path scheduling based on the path scheduling delay probability, and a Multipath Congestion Control (MCC) to regulate the congestion control through an explicit acknowledgment for the rate of transmission control based on packet delivery rate and buffer availability. It aims to reduce packet loss and improve the throughput utilizing the MP-PS approach. The proposed approach has been extensively simulated by means of a simulator and it showed improvised performance in provisions of throughput and buffer utilization when evaluated with the conventional approach.

Anomaly detection is a vital research problem among the different domains intrusion detection, fraud detection, device health monitoring, fault data detection, event detection in sensor networks. Anomalies mean an outlier, noise, novelties, exceptions which do not match the expected behavior of the system. Machine learning techniques work well in identifying these abnormal patterns. In this paper, the unsupervised clustering technique K-means, and its variation K-medoids partitioning are applied to detect anomalies. Sensor-embedded wearable devices are allowing smart healthcare services for people even in remote areas. These devices support continuous monitoring of people's health and allow the caregivers to provide better health assistance. Early-stage anomaly detection in such types of smart healthcare practices increases the efficiency of health services. In experimental discussion, K-means, and K-medoids partitioning clustering algorithms are assessed, and their performance is addressed.

More widespread use of positron emission tomography (PET) imaging is limited by its high cost and radiation dose. Reductions in PET scan time or radiotracer dosage typically degrade diagnostic image quality (DIQ). Deep-learning-based reconstruction may improve DIQ, but such methods have not been clinically evaluated in a realistic multicenter, multivendor environment. In this study, we evaluated the performance and generalizability of a deep-learning-based image-quality enhancement algorithm applied to fourfold reduced-count whole-body PET in a realistic clinical oncologic imaging environment with multiple blinded readers, institutions, and scanner types. We demonstrate that the low-count-enhanced scans were noninferior to the standard scans in DIQ (p 0.05) and overall diagnostic confidence (p 0.001) independent of the underlying PET scanner used. Lesion detection for the low-count-enhanced scans had a high patient-level sensitivity of 0.94 (0.83-0.99) and specificity of 0.98 (0.95-0.99). Interscan kappa agreement of 0.85 was comparable to intrareader (0.88) and pairwise inter-reader agreements (maximum of 0.72). SUV quantification was comparable in the reference regions and lesions (lowest p-value=0.59) and had high correlation (lowest CCC = 0.94). Thus, we demonstrated that deep learning can be used to restore diagnostic image quality and maintain SUV accuracy for fourfold reduced-count PET scans, with interscan variations in lesion depiction, lower than intra- and interreader variations. This method generalized to an external validation set of clinical patients from multiple institutions and scanner types. Overall, this method may enable either dose or exam-duration reduction, increasing safety and lowering the cost of PET imaging.

More widespread use of positron emission tomography (PET) imaging is limited by its high cost and radiation dose. Reductions in PET scan time or radiotracer dosage typically degrade diagnostic image quality (DIQ). Deep-learning-based reconstruction may improve DIQ, but such methods have not been clinically evaluated in a realistic multicenter, multivendor environment. In this study, we evaluated the performance and generalizability of a deep-learning-based image-quality enhancement algorithm applied to fourfold reduced-count whole-body PET in a realistic clinical oncologic imaging environment with multiple blinded readers, institutions, and scanner types. We demonstrate that the low-count-enhanced scans were noninferior to the standard scans in DIQ (p p p-value=0.59) and had high correlation (lowest CCC = 0.94). Thus, we demonstrated that deep learning can be used to restore diagnostic image quality and maintain SUV accuracy for fourfold reduced-count PET scans, with interscan variations in lesion depiction, lower than intra- and interreader variations. This method generalized to an external validation set of clinical patients from multiple institutions and scanner types. Overall, this method may enable either dose or exam-duration reduction, increasing safety and lowering the cost of PET imaging.

Cis and trans-interactions in cadherins are the foundations of cellular adhesions in multicellular organisms. While the trans-interactions mediate the intercellular attachment, the cis-interaction is presumed as reinforcement to trans. Thus, trans precedes cis has been the well-accepted model in cadherin adhesion. The stronger affinity of trans-binding over cis has been the decisive influence in the trans first model. Here we show that cadherin-23, a non-classical cadherin with an extended extracellular region, can undergo cis-clustering in solution independent of trans and phase separate as liquid droplets. Using single-molecule measurements, we decipher that weaker cis-interactions favor the cis-clustering. In-cellulo, the cis-clustering is manifested as puncta, a common feature in non-classical cadherin junctions, and accelerates the cell adhesion. The cis-clustering thus kinetically controls cell-adhesion before trans-binding. Notably, M2-macrophages predominantly express cadherin-23 and rapidly attach to circulatory tumor cells during metastatic migration. However, the relation of cis-clustering with rapid cell-cell adhesion in physiology is not yet established

Convolutional neural networks (CNNs) are similar to “ordinary” neural networks in the sense that they are made up of hidden layers consisting of neurons with “learnable” parameters. These neurons receive inputs, perform a dot product, and then follows it with a non-linearity. The whole network expresses the mapping between raw image pixels and their class scores. Conventionally, the Softmax function is the classifier used at the last layer of this network. However, there have been studies conducted to challenge this norm. Empirical data has shown that the CNN model was able to achieve a test accuracy of ≈99.04% using the MNIST dataset. MNIST dataset consists of 60,000 training images and 10,000 testing images. This experiment was inspired by following the experiment on MNIST dataset. The dataset we used in this experiment is collection of images consisting of cats and dogs. These images are gathered from different sources over internet. This dataset consists of 10,000 images of each class i.e., Cats and Dogs. The overall accuracy of training and validation set is 96.85%. The said results may be improved if data pre-processing techniques were employed on the datasets, and if the base CNN model was alternatively more sophisticated than the one used in this study.

Image segmentation is a vital step in image processing and has attracted many researchers towards its potential applications like object recognition, pattern recognition, computer vision and artifi...

Image inpainting, image deblurring, and noise removal are influential concepts in the field of digital image processing. Second-order diffusion-based image restoration models suffer from staircase effects and connectivity principle, while fourth-order models suffer from speckle artifacts. In this article, a robust image inpainting model using fractional-order nonlinear diffusion driven by difference curvature is proposed and fractional-order variational model is utilized to remove the noise and blur. Fractional-order derivatives can deal well with edges and attain good trade-off between edges preservation and elimination of staircase and speckle artifacts of an image. Difference curvature is a feature descriptor which can effectively characterize the intensity variations in the image. In this work, difference curvature-based a bi-weight, adaptive conductance coefficient is proposed to restore the image and fractional-order derivative is implemented by using discrete Fourier transform. Simulation results validate that the proposed model can adequately complete the damaged regions, solve the connectivity principle, and also avoid the staircase and speckle artifacts.

Transradial approach has become preferable to conventional femoral artery access for both diagnostic coronary angiography and percutaneous coronary intervention. A rare but recognizable complication of radial access is radial artery pseudoaneurysm (RAp), which represents a potentially catastrophic complication. Treatment options comprise ultrasound (USG)-guided manual compression or thrombin injection or surgical correction. In this case series, we report four cases of RAps that we encountered at a single tertiary care center from July 2015 to January 2018. We compressed the radial artery using a sphygmomanometer cuff differentially rather than a TR band proximal to the pseudoaneurysm to treat three of them. One of the patients underwent surgical repair of the pseudoaneurysm as the location of the aneurysm was not suitable for compression or thrombin injection. In our series of cases, we conclude that RAp, a rare complication of radial catheterization, was seen more commonly in elderly female patients and can be easily treated by the USG-guided differential compression, a simple and readily available method. Prevention and early diagnosis is the key to avoid serious consequences.

BackgroundATLAS evaluated the efficacy and safety of the PARP inhibitor rucaparib in patients with previously treated locally advanced/unresectable or metastatic urothelial carcinoma (UC).MethodsPatients with UC were enrolled independent of tumor homologous recombination deficiency (HRD) status and received rucaparib 600 mg BID. The primary endpoint was investigator-assessed objective response rate (RECIST v1.1) in the intent-to-treat and HRD-positive (loss of genome-wide heterozygosity ≥10%) populations. Key secondary endpoints were progression-free survival (PFS) and safety. Disease control rate (DCR) was defined post-hoc as the proportion of patients with a confirmed complete or partial response (PR), or stable disease lasting ≥16 weeks.ResultsOf 97 enrolled patients, 20 (20.6%) were HRD-positive, 30 (30.9%) HRD-negative, and 47 (48.5%) HRD-indeterminate. Among 95 evaluable patients, there were no confirmed responses. However, reductions in the sum of target lesions were observed, including 6 (6.3%) patients with unconfirmed PR. DCR was 11.6%; median PFS was 1.8 months (95% CI, 1.6–1.9). No relationship was observed between HRD status and efficacy endpoints. Median treatment duration was 1.8 months (range, 0.1–10.1). Most frequent any-grade treatment-emergent adverse events were asthenia/fatigue (57.7%), nausea (42.3%), and anemia (36.1%). Of 64 patients with data from tumor tissue samples, 10 (15.6%) had a deleterious alteration in a DNA damage repair pathway gene, including four with a deleteriousBRCA1orBRCA2alteration.ConclusionsRucaparib did not show significant activity in unselected patients with advanced UC regardless of HRD status. The safety profile was consistent with that observed in patients with ovarian or prostate cancer.Trial registrationThis trial was registered inClinicalTrials.gov(NCT03397394). Date of registration: 12 January 2018. This trial was registered in EudraCT (2017–004166-10).

Neural network models are a big segment of machine learning which is highly useful to segregate based on its feature. They have the immense capability to extract and learn the features. Predicting cancer well before the patient shows symptoms can help avoid deaths drastically. Since deep learning can do the task of feature extraction, we designed a DL model which can classify various cancer images into cancerous or non-cancerous. The proposed model was separately trained over breast cancer (BC) and brain tumor (BT) dataset and has achieved an overall accuracy of 0.95 and 0.72 respectively. For this reason, we compared the proposed model with Resnet-50, VGG-19 and VGG-16. The proposed model achieved 0.95 accuracy on brain tumor dataset. Where as vgg16 obtained 0.94 accuracy and resnet-50 obtained only 0.81 accuracy respectively on brain tumor dataset. Clearly, our model surpassed the rest not only in terms of accuracy but also speed. The proposed model performed far better than Resnet-50.

Coronavirus pandemic disease is caused by severe acute respiratory syndrome coronavirus 2. Generally RT-PCR or other Nucleic testing is used in order to detect covid19. In computed Tomography Scans, it can be clearly viewed that to how much extent the virus has damaged the Lungs. Computed Tomography gives the result in 15 minutes, whereas RT PCR takes 24 hours. PCR only checks whether virus is in nose or throat but the proposed model checks in lungs which is most accurate. The utilization of computed Tomography Scans will give us better and accurate results. The proposed novel model helps to recognize the corona virus in Lungs Computed Tomography Scans and achieved an accuracy of 0.93 with Gabor filter and 0.85 without Gabor filter. The existing models VGG16, VGG19, ResNet50 and Mobile Net achieves an accuracy of 0.89,0.91,0.91,0.91 respectively using Gabor filter and 0.78,0.71,0.81 and 0.89 without using Gabor Filter. Gabor filter will help to remove the noise from the data, it is linear filter and orientation sensitive. Our model achieves an accuracy 0.93 which is better than VGG16, VGG19, ResNet50, Mobile Net models using Gabor Filter.

Trauma is a major cause of death in the United States, particularly in the younger population. Many traumatic deaths, as well as major morbidity, occur secondary to uncontrolled hemorrhage and eventual exsanguination. Interventional radiology plays a major role in treating these patients, and interventional techniques have evolved to the point where they are an integral part of treatment in these critically ill patients. This article reviews the role of interventional radiology in the treatment algorithms for traumatic injury sponsored by major societies and associations.

In recent years, an increased fatal rate has been observed due to water injury/accidents and lack of emergency life savers. The major water accidents occur as a result of the improper maintenance of water boats. To overcome these challenges, this research work aims to monitor the conditions of water boat and track its location continuously. If any abnormal condition occurs in the boat, the proposed system will send the information to the concerned authority via SMS to save human lives. The main objective of the proposed system is to track the location of the ship and alert the rescue team as well as the captain when the ship/boat is in danger or in an emergency condition. This system mainly uses ARDUINO, GPS, and GSM for tracking and communication. The water sensors present at the bottom portion of the ship helps to monitor the ship’s condition and it is simply used to verify whether the boat is present above the safety level. If the ship gets overloaded, an indication will be displayed on the LCD display and also an SMS will be sent to the mobile phone of captain and also to the control room. Simultaneously the engine will be turned off to stop the ship/boat. If the ship is going to sink into the water, then the coordinates of the ship will be sent to the rescue teams for ensuring an immediate and fast rescue. There is also an option for sending the location to control room in case of any help with a message "Help". Some more additional features like finding the wind speed in that area and detecting the nearer objects to the ship using ultra sonic sensor are also included.

This work aims to detect the signs of the Coronavirus, also known as Covid-19. A dry cough, sore throat, and fever are the most common symptoms of Covid-19. For Covid-19, it is important to find fast and precise results at the time to stop it in the early stages and to avoid it from the vast spread. To interpret and identify the symptoms from X-rays and CT scan images, the machine learning and computer vision principles were applied. The works are usually performed with the CSV datasets. However, the analysis is performed to compare the images of patients with Covid and Non-Covid. To enhance the classification performance, it is feasible to use feature extraction techniques such as CNN, local directional pattern (LDP), gray-level run length matrix (GLRLM), gray-level scale zone matrix (GLSZM), and discrete wavelet transform (DWT) algorithms (Barstugan in Coronavirus (Covid-19) classification using CT images by machine learning methods [1]). In this paper, the convolution neural network model is selected as the classifier. Softmax is used during the classification process to classify the images given, whether they belong to Covid or Non-Covid. This implementation is carried out on both the X-ray images dataset and the dataset of CT scan images which are obtained from the repository that is publicly accessible.

Facial expression detection has long been regarded as both verbal and nonverbal communication. The muscular expression on a person's face reflects their physical and mental state. Using computer programming to integrate all face curves into a categorization class is significantly more important than doing so manually. Convolutional Neural Networks, an Artificial Intelligence approach, was recently developed to improve the task with more acceptance. Due to overfit during the learning step, the model performance may be lowered and regarded underperforming. There is a method dropout uses to reduce testing error. The influence of dropout is applied at convolutional layers and dense layers to classify face emotions into a distinct category of Happy, Angry, Sad, Surprise, Neutral, Disgust, and Fear and is represented as an improved convolutional neural network model. The experimental setup used the datasets namely JAFEE, CK48, FER2013, RVDSR, CREMAD and a self-prepared dataset of 36,153 facial images for observing train and test accuracy in presence and absence of dropout. Test accuracies of 92.33, 96.50, 97.78, 99.44, and 98.68 are obtained on Fer2013, RVDSR, CREMA-D, CK48, and JAFFE datasets are obtained in presence of dropout. The used features are countably large in the computation as a result the higher computation support of NVDIA with the capacity of GPU 16GB, CPU 13GB and memory 73.1 GB are used for the experimental purposes.

A facial expression is a natural reflection of human feelings, It is the nature of the human to reciprocate through the facial expression to the living world from where the inputs are perceived. The human science measures the emotion, feeling and sentiment by seeing the human face and face curves, but the recognition of emotion through artificial means with high accuracy and less computing resources is more challenging. In this research work, we developed a state-of-the-art procedure that recognizes the emotion of seven categories, namely Happy, Anger, Sad, Disgust, Neutral, Surprise, and Fear efficiently using deep learning. In this work, the model is trained using the fer2013 data set consists of 35887, and the CK48+ dataset consists of 3540 images. We proposed a hybrid model of feature selection that is used before feeding to the proposed computing model of CNN architecture. We claim through the use of both the models one after the other the emotions is correctly recognized with high accuracy during both training and testing phases, which the conventional method doesn’t have.

Deep learning models can assist dermatologist in classification of cancer into benign or malignant with high success rate. So, we proposed a new convolutional neural network model for classification of skin cancer, brain tumor, breast cancer, and colon cancer into benign or malignant. The proposed model and ResNet-50 model are trained on training set and compared by evaluating them on testing set. The insights of the results show that proposed model performs better than ResNet-50 in every category. The proposed model achieved 95% accuracy for brain tumor classification; the ResNet-50 model obtained 82% accuracy for brain tumor classification. The proposed model clearly outperformed ResNet-50 model. The average time taken per each epoch for proposed model is 4 s and for ResNet-50 model is 9 s.

Emotions have a key role in Feedback analysis to provide a good customer service, the main seven emotions are Anger, Disgust, Fear, Happy, Neutral, Sad and Surprise. There are several advantages, an efficient Facial Emotion Recognition model can help us in self-discipline and control over the drivers, while they are driving the vehicle. Low resolution and Low-reliable images are main problems in this field. We proposed a new model which can efficiently perform on Low resolution and Low-reliable images. We created a low resolution facial expression dataset (LRFE) by collecting various images from different resources, which contains low resolution images. We also proposed a new hybrid filtering method, which is a combination of Gaussian, Bilateral, Non local means filtering techniques. Densenet-121 achieves 0.60 0.68 accuracy on fer2013 and LRFE respectively. When hybrid filtering method is combined with Densenet-121, it achieved 0.95 accuracy. Similarly Resnet-50, MobileNet, Xception models performed effectively when combined with the hybrid filtering method. The proposed convolutional neural network(CNN) model achieved 0.65 accuracy on fer2013 dataset, while the existing models like Resnet-50, MobileNet, Densenet-121 and Xception obtained 0.60 0.57 0.60 0.52 accuracies on fer2013 respectively. The proposed model when combined with hybrid filtering method achieved 0.85 accuracy. Clearly the proposed model outperforms the traditional methods. When the hybrid filtering method is combined with the CNN models, there is significant increase in the accuracy.

Asymmetric modifications in the stern of KVLCC2 vessel are made in order to improve propulsion efficiency. The purpose of the modifications is to create a pre-swirl in a direction opposite to that of the propeller rotation and hence reduce losses. Stern asymmetry is designed by geometric modifications of the stern frames through an angle in the direction opposite to that of propeller rotation. The nominal wake distribution is investigated using CFD analyses along with the distribution of axial and tangential velocity components. A significant difference in the tangential velocity component is observed between the original symmetric KVLCC2 and the modified asymmetric case. The resistance of the asymmetric variant is found to be marginally higher. However, the rpm in self-propelled condition and the power requirement is found to be reduced by the asymmetric modification. For the asymmetric variant, model scale self-propulsion experiments indicate a reduction of power by about 8%, whereas full scale CFD calculations estimate a reduction of 4% corresponding to the design speed of 15.5 knots.

Amid the deals on Black Friday, all the retailers are packed. Many items are discounted with limits and clients surge in purchase of the items. It is hard for clients to purchase the items even with a strong arrangement. In any case, the shop proprietors face significantly more trouble on controlling the group with constrained staff and in focusing on imminent clients. A few methods have been utilized to handle this issue, yet they are not unreasonably effective. A model for predicting is a system that demonstrated promise in taking care of this issue. I focus on predicting models to build an exact and proficient algorithm to analyze the clients spending before and yield the future spending of the client with similar features. Distinctive data analyzing techniques such as regressors, NN and classifiers to develop a model for prediction are actualized and an examination is implemented dependent on their exhibition and exactness of predicting. These machine learning techniques are executed using various algorithms to find the optimal prediction. I applied six distinct ML algorithms. Furthermore, I apply the visuals and data preprocess to obtain the optimum results.

The increase in the use of metal additive manufacturing (AM) processes in major industries like aerospace, defense, and electronics indicates the need for maintaining a tight quality control. A quick, low-cost, and reliable online surface texture measurement and verification system are required to improve its industrial adoption. In this paper, a comprehensive investigation of the surface characteristics of Ti-6Al-4V selective laser melted (SLM) parts using image texture parameters is discussed. The image texture parameters extracted from the surface images using first-order and second-order statistical methods, and measured 3D surface roughness parameters are used for characterizing the SLM surfaces. A comparative study of roughness prediction models developed using various machine learning approaches is also presented. Among the models, the Gaussian process regression (GPR) model gives an accurate prediction of roughness values with an R2 value of more than 0.9. The test data results of all models are presented.

Background: Gastric outlet obstruction (GOO) mechanically impedes gastric emptying, normal emptying of the stomach. It is a diagnostic and therapeutic challenge for general surgeons in their daily practice. This paper highlights the etiology, clinical presentation and treatment outcome of GOO.Methods: A Prospective study was conducted on patients with GOO treated at Rajarajeswari medical college and hospital, Bangalore during September 2015 to august 2017. Data was tabulated and analyzed using descriptive statistical methodology.Results: Carcinoma stomach with antral growth and cicatrized duodenal ulcer (both 41.5%) were the most common cause of gastric outlet obstruction. Male were more affected than females (2.5:1). Most common symptom was vomiting and abdominal pain (noted among all), followed by loss of appetite (83%) and loss of weight (82.35%). 94.1% patients of Cicatrized duodenal ulcer underwent truncal vagotomy with posterior gastrojejunostomy and 5.9% underwent truncal vagotomy with antrectomy. 58.8% patients of carcinoma stomach, underwent distal gastrectomy with ante-colic Roux-en-Y gastro- jejunostomy, 12.1% patients underwent subtotal gastrectomy with ante-colic and Roux-en-Y gastro jejunostomy and 4.8% patients underwent palliative gastrojejunostomy. In corrosive antral stricture Billroth I gastrectomy was done. Patients of pancreatic malignancy underwent palliative anterior gastrojejunostomy and pseudo- pancreatic cyst was treated by cystojejunostomy. The average hospital stay was 10-14 days and an overall mortality of 5.8% for malignant patients was noted.Conclusions: Study concludes that gastric outlet obstruction is an important and a common surgical condition in tertiary hospital. Malignancy and benign cicatrized duodenal ulcer being the most common cause.

The image inpainting quality depends on the method of patch priority calculation and best matching patch selection method. Image inpainting can be applied in repairing the damaged images and removi...

Introduction: Reconstruction of complex abdominal wall defects is both challenging and technically demanding for plastic surgeon. Objectives in abdominal wall reconstruction are consistent and include restoration of abdominal wall integrity, protection of intra abdominal viscera and prevention of herniation. Materials: We conducted a retrospective study on five patients in whom lateral thigh flaps such as anterolateral thigh (ALT) flaps and tensor fascia lata (TFL) myocutaneous flaps as pedicled or free flaps were used for complex abdominal wall Type II defects over a 5-years period between 2007 and 2012. Results: In two patients, free flaps were used for reconstruction of the upper abdomen and both were ALT. In three patients of lower abdominal defects, one patient had bilateral pedicled ALT flaps, one pedicled TFL myocutaneous and one free TFL myocutaneous in view of ipsilateral electric burn scars. There were no flap losses. Patients were followed up beyond 6 months and found to have a good abdominal contour and only one of five had clinical evidence of herniation. Conclusion: It can be concluded that flap from the Lateral thigh (ALT or TFL) is flap of choice for large Type II abdominal defects. Including vascularised fascia in the flap maintains abdominal wall integrity and use of synthetic mesh is not necessary. Upper abdominal defects need free flaps and in lower abdominal defects a pedicled flap suffices.

The facial emotion recognition by the machine is a challenging task. From decades, researchers applied different methods to classify facial emotion into the different classes. The expansion of artificial intelligence in a form of deep convolutional neural network (CNN) changed the direction of the research. The facial emotion recognition using deep CNN is powerful in terms of taking bulk input images for processing and classify with high accuracy. It has been noticed in a few cases the classification model does not judge the facial images into appropriate classes due to the influence of noises. So, it is highly recommended to apply a noiseless image to the facial emotion recognition model for classification. We adopted a mechanism and proposed a model for classifying facial image into one of the seven classes with high accuracy. The images are smoothed before applying to the model by different smoothing process as part of image preprocessing. We claim facial emotion recognition with image smoothing by different filters or a mixture of filter are more robust than without preprocessing. The detail is explained in the subsequent sections.