Search

Biomedical image processing and decision making is a growing research demand under global pandemic situation. The quality of medical images plays a vital role in streamlining remote diagnosis and processing via telemedicine platform, in providing unambiguous results and decision supports. This paper presents an improved Medical Image Content Quality (MICQ) technique and it aims to enrich the Magnetic Resonance (MR) image content or pixels based on semi supervised clustering technique for the process of deeper analysis and investigation to identify the normal and abnormal portions. The proposed (IMICQ) system is containing three stages namely pre-processing, clustering and validation respectively. In the pre-processing stage, the MICQ divides the MR image into finite number of non-overlapping blocks or vectors with size (2*2). Next stage, the proposed MICQ system iteratively partitions the MR image dataset or vector set into optimum number of highly relative dissimilar clusters based on K-Means clustering technique. In the last stage, the proposed system measures the quality of clustering result which obtained in the previous stage based on Effective Cluster Validation Measure (ECVM). Experimental results show that the MICQ is better suitable to improve MR image content quality for telemedicine platform and to predict the normal and abnormal portions over the image with higher accuracy ratio.

Malware attacks have grown in prominence in recent years, posing severe security risks and resulting in significant financial losses. The ability to rapidly and reliably classify malware is vital to cybersecurity due to the exponential growth of malware variants. The role of artificial intelligence plays a significant role in cybersecurity industry. Recently, in the field of malware detection deep learning technique seeks more attention than the machine learning techniques due to the complexity of its behavior. Because the deep learning technique performs well than the machine learning techniques in terms of accuracy and it is well suited for large amount of data. The input attribute for the proposed model is windows-based system call sequence which is collected from NT mal detect project. In this work, the unsupervised deep learning technique used for text classification namely LSTM autoencoder and the performance of proposed model compares with existing DL methods such as CNN, RNN and LSTM with the performance parameters of accuracy, precision, recall and F1-measure.

Pneumonia is a significant global health issue, causing substantial morbidity and mortality. Early and accurate diagnosis of pneumonia is crucial for effective treatment and patient management. The model is trained on a large dataset of labeled chest X-ray images, consisting of both pneumonia-positive and pneumonia-negative cases. To train the MobileNetV2 model, a transfer learning approach is adopted, where the pre-trained weights from a large-scale dataset are used as a starting point. Fine-tuning is performed by training the model on the pneumonia dataset using a combination of deep learning techniques, including data augmentation and regularization, to improve generalization and reduce overfitting. To evaluate the performance of the proposed pneumonia detection system, extensive experiments are conducted on an independent test dataset. The model's performance is assessed using various evaluation metrics, including accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC-ROC). The results demonstrate the efficacy of the MobileNetV2 model in accurately detecting pneumonia from chest X-ray images, showcasing its potential for use in real-world clinical settings. In conclusion, this study presents a pneumonia detection system based on the MobileNetV2 model, offering a practical and efficient solution for automated pneumonia diagnosis. The proposed model's high accuracy and suitability for deployment on mobile devices make it a valuable tool for healthcare professionals, enabling timely and accurate diagnosis of pneumonia, thereby facilitating prompt treatment, and improving patient outcomes.&rdquo

The proposed system is to identify and detect ‘bots’, ‘spam’ or ‘fake accounts’, referred to as inauthentic accounts that mimic how people use Twitter. While some spam accounts are automated, others are run by real people, making their detection challenging. Among other things, bots can follow and be followed by other users, tweet at people, and share tweets. On Twitter, scamming spam bots are routinely observed enticing users to transmit cryptocurrency, or digital currency, to online wallets in exchange for fictitious prizes

Epileptic Seizure (ES) is an abnormality associated with discharging of continues electric impulses from the instance of normal activity. The period and time interval of occurrence is a challenging task to record and validate. In this article, a focus is made to classify and predict the occurrence ratio of seizer based on augmented learning and fuzzy rules. The Epileptic Seizure datasets are acquired from pre-trained and validated approaches further re-trained using interdependent attributes based on augmented learning and training approach. The outcome of training is further used by fuzzy rules to classify and categorize the Epileptic Seizure based on occurrences series of patterns and time. The proposed technique is a hybrid approach and novel as segmented based learning is used to predict the seizer. The technique has recorded 92.23% accuracy in seizure classification and 89.91% in reliable prediction.

Alzheimer’s disease is associated with a fragmental protein deposits termed as biomarkers. These biomarkers are studied and researched with various techniques in improving the performance and accuracy of diagnosis. In this research article, a technique is proposed to extract the attribute of brain MRI datasets. The attributes are processed and computed using a neural networking technique to categorize attribute mapping based on Interdependent Attribute Interference (IAI). The categorized data is teamed with a fuzzy logic to provide a reliable computation rule in decision making. The proposed technique has outperformed the accuracy of disease evaluation and diagnosis with a categorization sensitivity of 89.27% and an accuracy of 93.91%.

In India, the resources availability with respect to demand has a larger gap in reality. For data analytics and resource scheduling for optimized access using the Internet of Things (IoT) as a data frontend collector, a machine learning technique is discussed in this paper. Telemedicine was initially considered "futuristic" and "experimental," but telemedicine is the need of this time. Telemedicine applies in a number of fields: patient care, education, science, management, and public health. The present study describes launchpad's for patient compliance. The proposed approach is aimed at achieving an integrated IoT Unit, Machine Learning Unit, and Broadcasting Unit treatment platform using an interface medium. Methods: The proposed system is designed for data collection using MSP430 and RL200 Launchpads at the patient side under sensory operations. The processed data is extracted to servers for processing and thus providing an optimized machine learning technique for resource sharing, such as drug distribution, instrumental enhancement, and appointment allocation. Novelty: The proposed system architecture utilizes user group from various locations under both rural and urban categories installing the monitoring application as discussed under the IoT unit of for data acquisition and transmission. The hardware kit is aided with MSP430 and supporting Wi-Fi for connectivity to cloud servers. Findings: In the future stage of experiment and setup, MSP430 was replaced by RL200 with a built-in Wi-Fi module for easy connectivity and achieve compactness. The proposed technique has achieved satisfactory results for open datasets under the Hadoop environment.

This paper presents an improved cluster validation scheme called two phase cluster validation (TPCV) and aims to estimate the inter closeness and inter separation among the clusters in the cluster set of unsupervised clustering schemes based on probability measure for validating the cluster quality without prior identification. First phase, the TPCV computes the representative cluster centroid of each individual cluster in the cluster set based on standard mean operation and then it estimates the probability of inter closeness of each cluster with other clusters in the cluster set based on cluster centroid. Next phase, it calculates the probability of separation among the clusters in the cluster set based on cluster centroid by distance measure. Experimental results show that the TPCV scheme is simple and effective to estimate the cluster quality by measuring the probability of closeness and separation between the clusters in the result of unsupervised clustering scheme.

Telemedicine is a blooming field with inter-disciplinary research and wide area of application refinement. Various techniques are proposed since last decade with a primary focus of improving telemedicine. The algorithms are either data dependent or application centric. In this paper, a multi-objective optimal medical (MooM) data processing technique is proposed under multi-dimensional data types of medial samples such as text files, image files, log files, electronic health records (EHR), audio signal files and graphic files. The technique proposes a dedicated methodology for independent data-type processing to retrieve on a standard protocol platform for transmission of data via telemedicine channel. The technique uses unsupervised and hybrid clustering approaches of machine learning to predict data-types attributes for processing, thus resulting in higher-order accuracy and data scalability on transmission channel of telemedicine environment. The MooM technique processed on medical images retrieve the compressed stream of data frames with QoS recorded 9.23, for medical textural data the QoS is 9.87 and audio signal pattern data, the QoS is recorded 9.76 on a scale of 10.

Replicate and near-replicate images are variants derived from original images that are common among the large number of images on the cloud, desktop and mobile storage areas respectively. The existence of such near-replicate an image in web search indicates the presence of redundancy and lowers the system performance. The detection of replication is challenging due to the multitude of possible variations of images. In this paper, a new system called Multilevel Clustering for Image Replication Identification (MCIRI) is presented. The proposed MCIRI system aims to identify the replication level of images over the large storage area based on multilevel unsupervised clustering techniques. The MCIRI consists of four stages: feature extraction, first level clustering, second level clustering and replication measurement. In the first stage, the MCIRI extract the three features over each individual image in the image set based on standard statistical operations such as mean, standard deviation and variance. Next stage, the MCIRI split the image feature vector of image-set into ‘K’ dissimilar groups based on K-means clustering technique. In the third stage, the MCIRI partially identifies the dissimilar cluster over each individual cluster in the cluster set of previous level clustering result. In the final stage, the MCIRI estimate the replication level of each individual cluster in the result of second level clustering based on the cluster validation measure. Experimental results show that the proposed MCIRI is better suitable to estimate the replication level in the image set.

Telemedicine in India is expanding and thus it is essential for application to reduce storage and transmission bandwidth. The prime objective in this paper is to demonstrate an investigative protocol for medical image compression in telemedicine. The method is based on Integer Wavelet Transform (IWT) and Huffman’s Coding for image compression with an added novelty of machine learning terminology for extracting the region of image based on pixel density and optimized layering for neighboring pixel ratio analysis. The experimental results are evaluated and demonstrated on open medical datasets and has acquired minimal predictive error, improvised Signal to Noise Ratio (SNR) with respect to Compression Ratio (CR) ratio and optimized bandwidth under transmission.

In this paper, an improved sentimental text analysis system called Probability Based Text Classifier (PBTC) is presented. It aims to train the existing unstructured text command set and to classify the sampled text command belongs into positive or negative polarity based on probability theory and supervised concepts. It consists of three stages pre-processed, training and classification. In the first stage, the proposed (PBTC) system identifies the relevant and irrelevant words in the unstructured text command set based on pre-determined text pattern model. In the second stage it identifies two dissimilar classes over the preprocessed text command set based on predetermined text pattern model and simple probability theory concepts. Next stage, the PBTC identifies the sample test text command without class label belong on which class based on Naive Bayer scheme and trained existing text command set. Experimental result shows that the proposed (PBTC) system is well suitable to train the unstructured text command set and classify the new text command belongs into positive or negative polarity with higher accuracy

Telemedicine is venturing towards development, resulting in larger data evaluation online. Medical signals are major Para-types in telemedicine transmission channel line. Since, biomedical signals are sensitive and acute in nature, a smallest ratio of hampering causes false positive prediction on resultant diagnosis. In this paper, a machine learning algorithm is discussed to regenerate the signals under transmission. The signals are decomposed in four layer DWT prior transmission to achieve signal optimization through channel. The methodology uses Real-Time Signal Re-Generator and Validator (RTSRV) Algorithm designed on neural networking model for training and validating incoming signals. The results demonstrate a performance consistency of 1.16 over 767 EEG processed samples with 0.65sec as an average processing time for regeneration and training.

In telemedicine and tele-consultation, physiological signals such as ECG, EEG, and PCG play a vital role on eHealth. On low line transmission channels, achieving higher Quality of Service (QoS) of signals is a major research challenge. In this paper, a compression technique is proposed using machine learning algorithm for most alike pattern extraction from signal segments called motifs. These motifs are clustered using K-nearest neighbor (KNN) technique, and resultant dataset is compressed by a native discrete wavelet transformers (DWT). The sample signals considered under experiment is phonocardiograhic (PCG) signal as majority of population cover under smartphones and thus acquiring heart signals via microphones to process in telemedicine environment becomes easier. The experiments retrieves lower SNR ratio with higher order of compression ratio and QoS.

An increased pressure of fluid in optic nerve can subsequently leads to permanent blindness are known as Glaucoma. The normal pressure of eye is 15mmHg or even lower, once it is higher than 30mmHg then there is risk in vi-sion loss. There are many existing technique that require experienced clinicians and cost effective. These systems use higher order spectra and discrete wavelet transform features for extracting the values and fed to classifier for normaliza-tion and ranking the feature. In this paper presenting a new methodology for di-agnosis of glaucoma based on EWT. Empirical wavelet transform is applied on image to format the sub band which is also called as decomposed image. These features are sustained into neural network system that produces ne value from n iteration and classify images into mild, intermediate and heavily affected eye using Fundus images.

Document classification is the task to split the document set into dis-tinct highly relative classes or groups based on nature of the document con-tents.Here, an improved approach of document classification called keyword-based document classification (KBDC) is introduced. It focuses on splitting the unstructured text document set into K number of dissimilar classes based on K predetermined keywords text models by improved probability technique. This new system comprises of the following stages. Namely, pre-processing, classi-fication and classifier stage respectively. Initial, the proposed system (KBDC) recognizes all the immaterial existing contents in the input text document through constructed Predetermined Irrelevant Text Pattern Model (PITPM). Next, it divides the pre-processed document set into ‘K’ different groups or classes by K number of Pre-determined Keyword Text Pattern Models (PKTPM) through probability technique, where K denotes the number of groups or classes or models. Finally, the KBDC system classifies the trial test text document without any class label that belongs to either of the existing group based on the K different class models (PKTPs). Experimentation results show that the KBDC is appropriate to split and identifies the unstructured text document set into K distinct extremely comparative classes.

As a proverb goes, “A Picture speaks more than a word”. The influence of a picture on any given information is predominant and has extra mileage in representing the information. In this paper, a linear model of word recognition is proposed using the image datasets. The proposed model is evaluated on primary datasets of UCL repository, the results are processed using MATLAB and the recognition ratio is effective compared to the previous approaches. Typically, the proposed technique is an optimized solution for word string extraction from a given image data-samples.

Data Mining is the procedure of extracting information from a data set and transforms information into comprehensible structure for processing. Clustering is data mining technique used to process data elements into their related groups or partition. Thus, the process of partitioning data objects into subclasses is term as ‘cluster’. It consists of data objects with un-unified proposition of high inter similarity and low intra similarity ratios. Thus reflecting the quality of cluster depends on the methods used. Clustering also called data segmentation, divides huge data sets into several groups based upon their similarities. This paper discusses a literature study of various clustering techniques and their comparison on key issues to give guidance for choosing clustering algorithm for a specific research application. The comparison is based on computing performance and clustering accuracy.

Telemedicine research improves the connectivity of remote patients and doctors. Researchers are focused on data optimization and processing over a predefined channel of communication under a depictive low QoS. In this paper a consolidated representation of telemedicine infrastructure of modern topological arrangement is represented and validated. The infrastructure is aided with Multiple Objective Optimized Medical dataset (MooM) processing and a channel optimizing TelMED protocol designed exclusively for remote medicine dataset transmission and processing. The proposed infrastructure provides an application oriented approach towards Electronics health records (EHR) creation and updating over edge computation. The focus of this article is to achieve higher order of Quality of Service (QoS) and Quality of Data (QoD) compared to typical communication channels algorithms for processing of medical data sample. Typically the proposed technique results are achieved to discuss in MooM dataset processing and TelMED channel optimization sessions and a resulting improvement is discussed with a comparison of each MooM dataset in reverse processing towards server end of diagnosis and a consolidated QoS is retrieved for proposed infrastructure.

The India is a home for 1.2 billion people and recognized as fastest growing economy on globe. In India, 68% (8.84 million) population contribute towards rural community. Information and communication technology (ICT) plays a vital role in development of telemedicine in India. In this study, a detailed analysis and observation reviews are conducted on the role of ICT in building telemedicine environment in India with major challenges and scenarios on implementation. Article also tracks major setbacks in storage systems, and need for the design of economic infrastructure and optimistic usage of available resources as India is a developing country.

As the energy demand is drastically increasing day by day all over the world. The reserves of fossil fuels like coal, oil and natural gas are depleting. To cope up with this unbalancing situation, we should keep more optimistic view about the renewable energy source. In India there is an abundant source of solar energy, this can be utilized in fulfilling the lower demands of the community. In this paper, solar energy is converted into electrical energy by photo voltaic effect hence a voltage is created appended to the load with an economical hardware and cost.

Data mining or Knowledge mining is researcher's selective area in current trend. Clustering amongst which is retained as most challenging issue. In this paper, a study is conducted on various techniques proposed in clustering with its limitations and remarks followed by possible outcome discussions using Grey wolf optimization approach.