Search

Diabetes mellitus is becoming a major health concern causing kidney failure, retinopathy, and other neural diseases and thus becoming a major research area for advancement in machine learning. Analysis of diabetes data is very difficult due to inconsistency and poor data quality, Correlational Mmisodels, and the complex nature of most clinical data. Both diagnosis and treatment are unsuitable algorithms based on machine learning. This prompted us to review the various diabetes standards. Diagnosing diabetes at an early stage necessitates a methodological departure from past practices. The compiled information from all pertinent, independent qualitative investigations with the purpose of fixing the issue of diabetes prediction is discussed in the proposed work. This article discusses the research undertaken in detection of diabetes. The analysis of samples tested for diabetes from different state-of-the-art studies and recommending the best practice is conveyed in this work. Using risk stratification based on machine learning, diabetes-related individuals can be separated from non-diabetic patients. [ABSTRACT FROM AUTHOR]

Hotel industry is the one which has confronted the unprecedented effect of the coronavirus disease 2019 (COVID-19) pandemic to significant social and economic risks. The COVID-19 pandemic has challenged the tourism across the globe and impacted hospitality in hotel industry severely. This study aims to assess customer satisfaction by carrying sentiment analysis and topic modelling over customer reviews on the hospitality provided by hotels in different continents during January to September 2020, i.e. the COVID-19 pandemic. We formulate an improved new scale of metrics to categorize customer satisfaction assessed by sentiment analysis in an elaborate way. Topic modelling was deployed to understand various topics most often discussed by customers. We find that North America and Europe could perform up to customer expectation. In Asia, Sri Lanka did well, Indonesia could maintain its customer satisfaction, while India consistently improved the satisfaction level. We identified 12 most discussed topics, and main reasons of dissatisfaction appear in staff, service, room, cleanliness, slow booking, and pandemic response by hotel. Findings of this study will help senior managers of hotels of developed as well as developing countries in providing new and effective services that can satisfy customers and restore their confidence. [ABSTRACT FROM AUTHOR]

Many strains of corona virus such as alpha, beta, gamma, delta, and omicron are still prevalent in various parts of the world. The new type of corona virus is called a variant when it is caused by more than one genetic mutation from the previous type of corona virus. Various strains around the world have come so far. The cough may persist for more than an hour or three or four times in 24 hours and body heat is high. You may not be able to feel the smell or taste. Researchers say that some people may have symptoms similar to those of a severe cold. In this paper, a multi threshold model was proposed to find the initial infection detection of COVID disease. Based on the initial health symptoms these methods observe the inputs of the patients. Then the observed symptoms are compared with the existing database and identify the spreading of the disease. This report was directly monitored by the patient and doctor. This model was helpful to provide the periodical monitoring and perfect treatments to the infected patients. [ABSTRACT FROM AUTHOR]

• Fairness-aware Collaborative Filtering (FCF). • Adaptive and Concise Empirical Wavelet Transform (ACEWT). • Attention Induced Multi-head Convolutional Neural Network with MobileNetv1. • Classifies the COVID-19, like Normal, COVID-19, SARS, Pneumocystis. • The proposed C19D-AIMCNN-MNet-JSOA model experimentally authenticated. In this research, Attention Induced Multi-head Convolutional Neural Network Organization using MobileNetv1 Transfer Learning and COVID-19 Diagnosis using Jellyfish Search Optimization Process on Chest X-ray Images (C19D-AIMCNN-MNet-JSOA) is proposed. Initially, the input images are taken from chest X-ray dataset. Fairness-aware Collaborative Filtering (FCF) is utilized for eliminating the noise and also improves the X-ray image quality. Next, these pre-processed images are given to Adaptive and Concise Empirical Wavelet Transform (ACEWT) for extracting Grayscale statistic and Haralick Texture features. The extracted features are given into the Attention Induced Multi-head Convolutional Neural Network with MobileNetv1 (AIMCNN-MNet) which classifies the COVID-19, like Normal, COVID-19, SARS, Pneumocystis. In general, AIMCNN-MNet does not show any optimization adaption techniques to determine the ideal parameter to provide precise COVID-19 categorization. The proposed C19D-AIMCNN-MNet-JSOA model experimentally authenticated utilizing chest X-ray dataset in MATLAB and performance metrics including sensitivity, precision, F-Score, specificity, accuracy, Kappa, computation time, error rate used to examine the efficiency of proposed method. The performance of the C19D-AIMCNN-MNet-JSOA approach attains 25.99%, 20.34%, 30%, 19% and 20.35% high Precision, 25.43%, 29.53%, 22%, 28% and 25.31% lower computation Time and 15.249%, 25.491%, 10%, 31% and 13.98% higher RoC comparing with existing methods like novel hand-crafted fusion model founded on deep learning features COVID-19 diagnosis and organization using X-ray pictures of the chest (C19D-CNN-MLP), Multi-modal fusion of deep transfer learning founded COVID-19 diagnosis and organization utilizing chest x-ray images (C19D-MMF-DTL), Recognition and organization of lung diseases for pneumonia and Covid-19 utilizing machine along deep learning methods (C19D-RNN-LSTM). [ABSTRACT FROM AUTHOR]

• To introduce a pre-processor using a cascaded Variable step size NLMS and Sparse Low-Rank filter (CVSS-NLMS-LRF) method to remove noise from ECG signals. • To perform Higher-order spectral energy distribution image (HSDI) method using two-dimensional Fourier transform of the third order cumulant function. • To classify cardiac arrhythmia detection using the multi-stage dual-swin transformer method with attention learning (MS-DSwin-AL). • To propose integrated average subtraction and standard deviation-based optimizer (IASSD) Method for hyperparameter optimization. • To compare the performances of the proposed model with the existing methods in terms of different performance metrics to prove the performance superiority of the proposed method. In this article, a new framework for arrhythmia identification using higher-order spectral distributions and deep learning approaches has been proposed. The input signal is first pre-processed using the Sparse Low-Rank filter (CVSS-NLMS-LRF) and Cascaded Variable step size-Normalized least mean square algorithm (CVSS-NLMS-LRF) techniques. This method eliminates various types of noise signals from the Electrocardiogram (ECG) signals, such as power line noise, baseline wander noise (BW), and high-frequency muscle artefacts. After pre-processing the signal, the features are selected using a Higher-order spectral energy distribution image (HSDI) obtained using a two-dimensional Fourier transform from the third-order cumulant process. Finally, the multi-stage dual swin transformer with attention learning (MS-DSwin-AL) is proposed to classify cardiac arrhythmias in the input ECG spectral. The Dual Swin Transformer, the Channel and Element-wise Attention Mechanism (CEAM) and the Transitional Module (TM) form the framework of the proposed classification. Additionally, the classification parameters are fine-tuned using an Integrated Average Subtraction and Standard Deviation based Optimizer (IASSD) algorithm. As a result, the proposed model is executed in the Python platform using the MIT-BIH dataset, and the performance is considered in terms of various evaluation metrics. Furthermore, the performance of a proposed model is compared with existing classifiers. The proposed model achieves an accuracy (96.01%), specificity (94%), recall (93.02%), and F1-score (89.14%) higher than the existing classifiers for cardiac arrhythmia classification. As a result, it can be said that the proposed model has a strong chance of identifying cardiac arrhythmias from the provided data. [ABSTRACT FROM AUTHOR]

In modern companies, the functions of divisions, departments and staff are provided by telecommunication transmitting analog and digital unit information via SCS. Such cable system refers to the use of copper or optical cable networks, passive and active switching devices. Structured cabling system or abbreviated SCS is a complex set of cable trunks and switching equipment that provide the transfer of various types of media data (audio, video, computer data) and is the basis for the operation and integration of telephone, local computer networks, security systems and other services. Many modern systems of security or communications today integrate a wide variety of interfaces into their arsenal, greatly expanding their capabilities and performance. In this paper a smart model based on high-speed communication network with the help of structured cabling system (SCS). Here the speed and bandwidth play the major role. The proposed system focused the highspeed communication between sender and receiver with some higher bandwidth optimization. [ABSTRACT FROM AUTHOR]

Here we are reporting a case of capillary haemangioma in a new born female child born to non consanguineous parents. Capillary Haemangioma is a very common angiomatous lesion that occurs in infancy or in childhood. It may occur either superficially in the skin or at a deeper level. At the deeper level, intramuscularly or very rarely within the osseous tissue. On examination, child had a well-defined nonpulsatile swelling over the upper one-third of the right forearm on the flexor aspect very near the elbow. Skin was stretched and shinny with haemorrhagic spots and pale yellow border all round. There was a painful limitation of movement in the right elbow. All the peripheral pulses were normal, except the right radial pulse which was feeble. This study has profound embryological, as well as, clinical importance. Hence, it has been studied and reported.

The current study employs biclustering to alleviate some of the drawbacks associated with gene expression data grouping. Different biclustering algorithms are used in this study to detect unique gene activity in various contexts and reduce the duplication of broad gene information. Furthermore, machine learning or heuristic algorithms have become widely utilised for biclustering due to their suitability in problems where populations of potential solutions allow examination of a larger percentage of the research area. To begin with, gene expression data biclusters frequently contain data that is the same under a variety of different situations of gene expression. Therefore, the biclustering technique is particularly effective if the matrix lines and columns are merged immediately. Submatrices can be identified using the Large Average Sub matrix. A Fuzzy C-Means algorithm is also used to ensure that the sub-matrix can be expanded to include more rows and columns for further analysis. The sub-matrices and component precision and strength are factored into the system design. It uses biclustering techniques to differentiate gene expression information. On the Garber dataset, the simulation is run in Java. Using the average match score for non-overlapping modules, the influence of noise on overlapping modules using constant bicluster and additive bicluster, and the overall run duration, the study is assessed. [ABSTRACT FROM AUTHOR]

In this paper, we develop a liveness detection of iris present in the study to reduce various spoofing attacks using gray-level co-occurrence matrix (GLCM) and Deep Learning (DL). The input images of iris are been given to this technique for the extraction of texture and colour features with fine details. The details are fused finally and given to a DL classifier for the classification of liveness detection. The simulation is conducted to test the efficacy of the model and the results of simulation shows that the proposed method achieves higher level of accuracy than existing methods. [ABSTRACT FROM AUTHOR]

Cervical cancer is regarded as a serious threats to humanity, globally and this is a vital disease with huge spreading of virus that affects the health of humans. The virus is spreading at a rapid rate through mosquitoes that even may kill the one who is affected with cervical cancer. In this paper, we develop a quick response system that certainly finds the disease through a faster validation process. The study uses Convolutional Neural Network (CNN) as a deep learning model that classifies and predicts the condition or the infection status of a patient. The study uses a pre-processing model and a feature extraction model to prepare the image datasets for classification. The simulation is conducted to validate the effectiveness of the model over cervical cancer image datasets i.e. the blood samples of humans. The validation shows that the proposed method effectively classifies the patients in a faster manner than the other deep learning models. [ABSTRACT FROM AUTHOR]

Energy conservation is a key element in MANET and so the architecture of the protocol demands particular consideration. Several MANET designs explored processes for the storage of renewable resources. Energy dissipation is a key coordinating factor among MANET nodes. Power utilization is ensured in monitoring applications that battery depletions are lowered to avoid regular substitution. MANET primarily aims to relay data by using energy-efficient routing protocols and an increasing lifespan of the network. The results of the simulation show that the approach suggested improves energy efficiency compared to other approaches. The findings are checked and provide decreased power usage, better packet distribution, lower packet error rate, higher network lifetime and lower end-to-end latency compared to conventional methods. [ABSTRACT FROM AUTHOR]

The evidence is conflicting regarding the effect of delays from the time of diagnosis to surgery on the survival in patients with hepatocellular carcinoma. We sought to investigate the impact of time to surgery on overall survival for patients who underwent curative resection for primary hepatocellular carcinoma. We performed a retrospective review of all patients who underwent liver resection for primary hepatocellular carcinoma between the years 2000 and 2015. Using 30-, 60-, and 90-day cutoffs, we investigated the effect of time to surgery on survival outcomes by dichotomizing the patients and using inverse probability of treatment weighting to ensure comparability. We also investigated time to surgery in prognostic subgroups by modeling the statistical interaction between time to surgery and the relevant prognostic variable in multivariable Cox models. A total of 863 patients underwent liver resection for primary hepatocellular carcinoma during the study period. Using 30-, 60-, and 90-day cutoffs, time to surgery did not have a significant bearing on overall survival. For elderly patients (>70 years), patients with Child-Pugh B liver disease, American Society of Anesthesiologists status 2/3, tumor size >5cm, tumor size ≥10cm and presence of extrahepatic invasion, hazard ratio decreased and overall survival improved as time to surgery increased. However, for patients with liver cirrhosis or portal hypertension, increasing time to surgery was found to portend higher risks of death. Time to surgery does not have a significant bearing on overall survival, and modest delays even appear to be associated with improved survival in specific subsets of patients. The importance of these findings is that patients with hepatocellular carcinoma should be fully optimized before and not rushed to surgery because of concerns of tumor progression and a diminished survival. [ABSTRACT FROM AUTHOR]