Browse Items (14421 total)
Sort by:
-
A Novel Approach towards Key-point based Real-time Children Emotion Prediction
Emotion prediction is crucial in mental healthcare. It is vital in children as it aids in managing behavioral issues and early identification of emotional distress that can lead to helpful mental health support. The research uniquely centres on children's emotional expressions, addressing a gap in existing emotion detection studies, which often focus on adults. By specifically tailoring the model to recognize subtle expressions unique to children, the study contributes valuable insights into child psychology and emotion recognition. To address this gap, this work attempts to establish a comprehensive children's emotion dataset that can facilitate the study of emotions across various pose orientation. The approach introduces advanced key-point detection techniques that capture a higher density of facial landmarks, allowing for more nuanced analysis of emotional expressions. This fine-grained detection enables the identification of subtle changes that are critical in interpreting children's emotions. An effective face detector with deep architecture is designed to handle all pose orientations from key image frames. Optimal features are then chosen by re-ranking the features using a hybrid feature selection mechanism. The emotion category is revealed by careful analysis of sequences of emotion identification from these features and is not based on a single frame. This framework holds promise for educational institutions and healthcare facilities, offering insights into children's behavior through emotion analysis. Through experimental analysis and comparisons with three existing SOTA emotion prediction models, it is observed that the proposed system consistently outperforms existing models by exhibiting an accuracy of 77.7 on average. Overall, this study recommended that the proposed model is suitable for children's emotion prediction. 2025 The Author(s). -
A novel approach using steganography and cryptography in business intelligence
In the information technology community, communication is a vital issue. And image transfer creates a major role in the communication of data through various insecure channels. Security concerns may forestall the direct sharing of information and how these different gatherings cooperatively direct data mining without penetrating information security presents a challenge. Cryptography includes changing over a message text into an unintelligible figure and steganography inserts message into a spread media and shroud its reality. Both these plans are successfully actualized in images. To facilitate a safer transfer of image, many cryptosystems have been proposed for the image encryption scheme. This chapter proposes an innovative image encryption method that is quicker than the current researches. The secret key is encrypted using an asymmetric cryptographic algorithm and it is embedded in the ciphered image using the LSB technique. Statistical analysis of the proposed approach shows that the researcher's approach is faster and has optimal accuracy. 2021, IGI Global. -
A novel approach with matrix based public key crypto systems
Here in this model, a new mechanism is used for Public Key Cryptography. A generator matrix is used to generate a field with a large prime number. The generator matrix, prime number and quaternary vector are used as global variables. The Generator Matrix is powered by a private key to generate Public Key. Since the model is based on Discrete Logarithm Problem, which is Hard problem, the proposed algorithm supports the features like Authenticity of users, Security & Confidentiality of data transmitted. Going by the construction of the algorithm, Encryption is being done on blocks of data for which it consumes less computing resources. Going by complexity of the algorithm, the key length needed is about 72 bit lengths to provide sufficient strengths against crypto analysis. 2017 Taru Publications. -
A Novel Architecture for a Medical Image Recognition System Using Deep Learning-Based Multiple Regression Evaluation
Models based on machine learning are optimization models that collect data, assess it, and deliver the reports required by specialists and management to make the best decisions. The application of contemporary machine learning allows the organization to quickly analyze photographs, differentiate voices assist in providing customer service, assess the information that is at hand, and uncover connections to aid in decision-making processes. The results of this investigation use quantitative methodologies to collect data and analyze it using mathematical procedures such as regression modeling as well as analysis of variance. Deep learning techniques applied to digital imaging, particularly in medical treatment, can increase picture quality, aid in modeling, aid in making the best possible diagnosis, and successfully address demands from patients. To analyze the hypothesis, investigators intend to utilize statistical approaches such as descriptive data analysis, regression evaluation, and analysis of variance (ANOVA). The authors employ the purposive sample approach to choose respondents from the healthcare industry. Purpose sampling is a non-probability sampling approach. Researchers collected data from 193 respondents working at hospitals that are privately owned in Southern Asia. As stated by the study, all factors, including efficiently meeting patient needs, have a probability value of under 0.05, indicating that they are statistically noteworthy. Following the study, the coefficient of variance (R squared) is 0.744, or 74.4%. According to the study, there is a high association between better image quality and ML-based digital picture identification systems. The recognition of patterns and the application of artificial intelligence to computerized recognition of pictures also have a close link. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
A Novel Artificial Intelligence System for the Prediction of Interstitial Lung Diseases
Interstitial lung disease (ILD) encompasses a spectrum of more than 200 fatal lung disorders affecting the interstitium, contributing to substantial mortality rates. The intricate process of diagnosing ILDs is compounded by their diverse symptomatology and resemblance to other pulmonary conditions. High-resolution computed tomography (HRCT) assumes the role of the primary diagnostic tool for ILD, playing a pivotal role in the medical landscape. In response, this study introduces a computational framework powered by artificial intelligence (AI) to support medical professionals in the identification and classification of ILD from HRCT images. Our dataset comprises 3045 HRCT images sourced from distinct patient cases. The proposed framework presents a novel approach to predicting ILD categories using a two-tier ensemble strategy that integrates outcomes from convolutional neural networks (CNNs), transfer learning, and machine learning (ML) models. This approach outperforms existing methods when evaluated on previously unseen data. Initially, ML models, including Logistic Regression, BayesNet, Stochastic Gradient Descent (SGD), RandomForest, and J48, are deployed to detect ILD based on statistical measures derived from HRCT images. Notably, the J48 model achieves a notable accuracy of 93.08%, with the diagnostic significance of diagonal-wise standard deviation emphasized through feature analysis. Further refinement is achieved through the application of Marker-controlled Watershed Transformation Segmentation and Morphological Masking techniques to HRCT images, elevating accuracy to 95.73% with the J48 model. The computational framework also embraces deep learning techniques, introducing three innovative CNN models that achieve test accuracies of 94.08%, 92.04%, and 93.72%. Additionally, we evaluate five full-training and transfer learning models (InceptionV3, VGG16, MobileNetV2, VGG19, and ResNet50), with the InceptionV3 model achieving peak accuracy at 78.41% for full training and 92.48% for transfer learning. In the concluding phase, a soft-voting ensemble mechanism amplifies training outcomes, yielding ensemble test accuracies of 76.56% for full-training models and 92.81% for transfer learning models. Notably, the ensemble comprising the three newly introduced CNN models attains the pinnacle of test accuracy at 97.42%. This research is poised to drive advancements in ILD diagnosis, presenting a resilient computational framework that enhances accuracy and ultimately betters patient outcomes within the medical domain. 2024, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. -
A novel assessment of bio-medical waste disposal methods using integrating weighting approach and hesitant fuzzy MOOSRA
Bio-medical waste (BMW) management is highly important precaution for human health and environmental concern. There are several disposal treatment followed by medical practitioners in medical waste management. Here, a few disposal treatment is considered to be an alternatives. When assessing, it is necessary to evaluate and assume that all disposal treatment methods are safe and hygienic. In this way, every alternative assessment is evaluated based on the social acceptance, technology and operation, environmental protection, cost, noise and health risk. Finally the best alternative is chosen. When BMW is disposed and we select the best treatment method in BMW management, it can lead to multi-criteria decision making (MCDM) processes related to uncertain critical assessments. When making a decision, the decision makers having some hesitation to give their suggestions. Therefore, here we use hesitant MCDM method. In today's practice we have choose five methods of BMW disposal methods used in the medical world and we have its alternatives. One of these alternative is sorted by six criteria weights for selecting the best method. The main aim of this research paper is propose a new methodology of hesitant fuzzy weight finding technique, it is named as Hesitant Fuzzy Subjective and Objective Weight Integrated Approach (HF-SOWIA) and also propose a new hesitant fuzzy rank finding methodology, it is named as Hesitant Fuzzy Multi-Objective Optimization on the basis of Simple Ratio Analysis (HF-MOOSRA). After evaluation, the result shows that autoclaving is the best alternative for BMW disposal treatment methods. Furthermore, sensitivity analysis is make in order to observe the difference of alternative ranking when the importance of subjective and objective weights changes. 2020 Elsevier Ltd -
A Novel Assessment of Healthcare Waste Disposal Methods: Intuitionistic Hesitant Fuzzy MULTIMOORA Decision Making Approach
Waste produced from medical facilities systems incorporates a blend of dangerous waste which can posture dangers to humans and ecological receptors. Lacking administration of healthcare waste can prompt hazard to medicinal service specialists, patients, public health, communities and the wider environment. Hence, proper management of healthcare waste is imperative to reduce the associated health and environment risk. In this paper, we extend the MULTIMOORA decision making method with intuitionistic hesitant fuzzy set to evaluate the healthcare waste treatment methods. Intuitionistic hesitant fuzzy set is a generalized form of a hesitant fuzzy set. Intuitionistic hesitant fuzzy set considers the uncertainty of data in a single framework and take more information into account. The MULTIMOORA method consists of three parts namely the ratio system, reference point approach and the full multiplicative form. In the optimal ranking methods, the IHF-MULTIMOORA method is uncomplicated it is able to be used practically with high dimension intuitionistic hesitant fuzzy sets. For pathological, pharmaceutical, sharp, solid and chemical wastes, the preferred waste disposal methods are deep burial, incineration, autoclave, deep burial, and chemical disinfection, respectively. 2013 IEEE. -
A Novel Auto Encoder- Network- Based Ensemble Technique for Sentiment Analysis Using Tweets on COVID- 19 Data
The advances in digitalization have resulted in social media sites like Twitter and Facebook becoming very popular. People are able to express their opinions on any subject matter freely across the social media networking sites. Sentiment analysis, also termed emotion artificial intelligence or opinion mining, can be considered a technique for analyzing the mood of the general public on any subject matter. Twitter sentiment analysis can be carried out by considering tweets on any subject matter. The objective of this research is to implement a novel algorithm to classify the tweets as positive or negative, based on machine learning, deep learning, the nature inspired algorithm and artificial neural networks. The proposed novel algorithm is an ensemble of the decision tree algorithm, gradient boosting, Logistic Regression and a genetic algorithm based on the auto-encoder technique. The dataset under consideration is tweets on COVID-19 in May 2021. 2024 Taylor & Francis Group, LLC. -
A novel automated method for coconut grading based on audioception
The quality of the coconuts used for various purposes is of utmost importance. Demand for better quality products is constantly on the rise due to the improvements in the standard of living of people. There is a possibility that a bad coconut goes unnoticed by the traders, as it is hard to decide if a coconut is good or bad by relying only on its external appearance. Traditionally, quality assessment is carried out manually with the help of three senses; sight, hearing and smell. In the proposed work, a sound processing technique is used in an attempt to automate this process which overcomes the drawbacks of manual processing, which can be used in large godowns and warehouses. This proposed method provides the quality assessment of the coconut purely based on audioception. While creating the database, coconuts varying in size, shape, color and water content were taken from several places as a source for the dataset. Features are extracted from the sound pattern produced by the dropped coconut, which forms the basis for classification. Sequential Minimal Optimization (SMO), Dagging and Naive Bayes classifiers were used and the results obtained were found to be encouraging. 2005 ongoing JATIT & LLS. -
A novel automated method for the detection of strangers at home using parrot sound
The sound produced by parrots is used to gather information about their behavior. The study of sound variation is important to obtain indirect information about the characteristics of birds. This paper is the first of a series in analyzing bird sounds, and establishing the adequate relation of bird's sound. The paper proposes a probabilistic method for audio feature classification in a short interval of time. It proposes an application of digital sound processing to check whether the parrots behave strangely when a stranger comes. The sound is classified into different classes and the emotions of the birds are analyzed. The time frequency of the signal is checked using spectrogram. It helps to analyze the parrot vocalization. The mechanical origin of the sound and the modulation are deduced from spectrogram. The spectrogram is also used to check the amplitude and frequency modulation of sound and the frequency of the sound are detected and analyzed. This research and its findings will help the bird lovers to know the bird behavior and plan according to that. The greater understanding of birds will help the bird lovers to feed and care for birds. BEIESP. -
A Novel Back-Propagation Neural Network for Intelligent Cyber-Physical Systems for Wireless Communications
Wireless sensor networks, which play a significant role in monitoring complex environments that change rapidly over time, were used in the Artificial Intelligence method. External factors or the device designers themselves are both responsible for this complex behavior. Sensor networks often use machine learning techniques to adapt to such conditions, eliminating the need for excessive redesign. Cyber-physical systems (CPS) appeared as the promising option for improving physical-virtual interactions. The quality of the system containing processing information is primarily determined by the system function. There are many benefits obtained while combining Artificial Intelligence (AI) and Cyber-Physical Systems (CPSs) in buildings. In CPS-based indoor environment has various design schemes containing measurement and intelligent buildings in the control system consisting of detection, tracking, execution, and communication modules. The Multi-Agent System (MAS) is the smallest control unit that simulates among neurons and it flexibly provides the information. To mimic the interactions between human neurons, multi-agents are used. In this paper, the CPSs information world is built on the fundamental principle of granular formal concepts and the theory of granular computing is investigated. The calculation module is used by Back-Propagation Neural Network (BPNN) for pattern recognition and classification by environmental information. Various parameters namely the normalized root mean square error, peak signal-to-noise ratio, mean square error, and the mean absolute error are chosen as the objective assessment criteria to assess the benefits of the proposed method and the effectiveness of the proposed system is proven. 2024 IETE. -
A Novel Blockchain-Integrated Deep Learning Framework for Securing Smart Healthcare Communication Networks
With the rapid expansion of intelligent medical equipment and their interconnectedness through the Internet of Things (IoT), addressing safety issues in the communicating system has become increasingly critical. A learning mechanism is proposed for an intelligent healthcare-based communication system that uses blockchain for secure network communication and incorporates a data evaluation layer based on cloud which actively segregates and ranks transactions into three main categories: Good, Moderate, and Malware. Fog servers are utilized to route the communicating nodes via Rician and Rayleigh channels. The learning mechanism employs a deep neural network to instruct and classify categories, thereby improving the blockchain layer's decision-making process. This paper introduces several significant contributions, such as the development of a secure blockchain framework for user authentication and a protected digital ledger for communication. Additionally, it incorporates a cloud-driven data analysis layer combined with a neural network to improve training accuracy and category classification. The developed algorithm surpassed the existing works in terms of quality of service (QoS) parameters with low latency, bit error rate (BER), higher signal to inference plus noise ratio (SINR), packet delivery ratio (PDR), true detection rate (TDR), false detection rate (FDR), and throughput. Also, a thorough comparison of consensus mechanisms like practical Byzantine fault tolerance (pBFT), proof of work (PoW), Raft, and Paxos is done to ensure which consensus helps optimize the proposed system in terms of security and fault tolerance with low latency and energy-efficient operations. It also establishes a secure and efficient communication network for smart healthcare, aimed at enhancing the overall quality of life for individuals. 2025 Wiley Periodicals LLC. -
A novel chemical route for low-temperature curing of natural rubber using 2,4 dihydroxybenzaldehyde: improved thermal and tensile properties
A novel method for chemically curing natural rubber (NR) using 2,4-dihydroxybenzaldehyde (DHB) at low temperatures has been discovered. Adding varying amounts of DHB to NR increases the crosslinking between the NR molecular chains. The chemical reaction between NR molecular chains and DHB was confirmed through Fourier transform infrared (FTIR) and proton nuclear magnetic resonance (NMR) spectra. From the thermogravimetric analysis (TGA), the thermal stability and activation energy of degradation were determined. The variation in glass transition temperature (Tg), as an indication of increased crosslink density, reducing the mobility of rubber chains, has been confirmed through differential scanning calorimetry (DSC). The addition of DHB to latex significantly enhanced the thermal stability of the rubber. An increase in the activation energy of 5.52% was observed upon the addition of 80mL DHB into NRL when compared to the uncured one. Furthermore, the tensile properties, in terms of tensile strength and modulus of elasticity of rubber, were drastically increased through DHB crosslinking. Tensile strength values of rubber were found to increase by reducing its elongation at break due to the formation of crosslinks between the macromolecular chains. NR cured with 80mL DHB exhibited superior tensile and thermal properties among the series of cured samples. By adding 80mL of DHB, the tensile strength increased by 390% and the elongation at break decreased by 10%. The advantage of this curing method is that, it is an effective technique for crosslinking NR directly from NR latex at comparatively low temperature. Graphical abstract: (Figure presented.) Iran Polymer and Petrochemical Institute 2024. -
A Novel CNN Approach for Condition Monitoring of Hydraulic Systems
In the dynamic landscape of Industry 4.0, the ascendancy of predictive analytics methods is a pivotal paradigm shift. The persistent challenge of machine failures poses a substantial hurdle to the seamless functioning of factories, compelling the need for strategic solutions. Traditional reactive maintenance checks, though effective, fall short in the face of contemporary demands. Forward-thinking leaders recognize the significance of integrating data-driven techniques to not only minimize disruptions but also enhance overall operational productivity while mitigating redundant costs. The innovative model proposed herein harnesses the robust capabilities of Convolutional Neural Networks (CNN) for predictive analytics. Distinctively, it selectively incorporates the most influential variables linked to each of the four target conditions, optimizing the model's predictive precision. The methodology involves a meticulous process of variable extraction based on a predetermined threshold, seamlessly integrated with the CNN framework. This nuanced and refined approach epitomizes a forward-looking strategy, empowering the model to discern intricate failure patterns with a high degree of accuracy. 2024 IEEE. -
A novel congestion-aware approach for ECC based secured WSN multicasting
--Multicasting in Wireless Sensor Networks greatly reduces the communication complexity between The Base station and set of sensor nodes deployed in a given region. It reduces the number of packets to be sent thus minimizing the chance of congestion. Still the existence of congestion appears due to improper channel utilization resulting in low throughput. In this paper, we have addressed the issue of congestion with reference to WSN multicasting. The Simulation results have shown that our approach is better in terms of throughput and delay compared with existing approaches. 2018, Institute of Advanced Scientific Research, Inc.. All rights reserved. -
A Novel Cross-Validation Fusion Model Combining Vision Transformer and DenseNet161 for Enhanced Cervical Lesion Classification
Cervical cancer is fourth most common cancer in women across the world with highest impact in low- and middle-income countries. World Health Organization sent out a call for all UN nations to work toward the elimination of cervical cancer. Deep learning and artificial intelligence have been the go-to solutions for medical image analysis for diagnosis and prognosis. This paper aims to classify lesions in a colposcope captured cervix image with help of artificial intelligence models. To further advance automated cervical lesion classification, the study proposes a novel hybrid model that combines the complementary strengths of a vision transformer and DenseNet architecture. The paper also addresses ongoing challenges, such as interference from specular reflection areas and the difficulty in distinguishing between different lesion grades due to subtle visual differences. The proposed cross-validation decision fusion strategy aims to improve the reliability and robustness of the classification process. The results of the study affirm that deep learning and fusion technologies will steer the future direction of research in medical image analysis. DenseNet model has performed with an accuracy score of 0.695, sensitivity of 0.912, specificity of 0.979 and F1 score of 0.9100. These metrics are significantly improved versions of state of the art used in this study for comparative analysis. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
A NOVEL DATA SECURE MODEL FOR INTERNET OF HEALTH THINGS WITH A NEW LIGHTWEIGHT CRYPTOGRAPHY ALGORITHM AND STEGANOGRAPHY TECHNIQUE
Ensuring the security of data in Internet of Things (IoT) based healthcare systems (HS) presents considerable challenges due to the limitations of traditional embedding methods and cryptography techniques, leading to more memory consumption, more execution time, less security, inadequate payload capacity, and performance inefficiencies. To address these issues, the Bernoulli Fish-based Stego Algorithm (BFBSA) is introduced as an innovative solution. Specifically designed for IoT healthcare data, this algorithm is validated through the encryption and embedding of healthcare data. The process involves initializing IoT healthcare data, encrypting it using the BFBSA algorithm, and embedding the encrypted data within steganographic images. Performance analysis is conducted using key metrics such as payload capacity, encryption time, memory usage, PSNR, and MSE. Comparative analysis with existing approaches highlights the BFBSA models efficiency and its effectiveness in ensuring secure and optimized data management in IoT healthcare environments. Little Lion Scientific -
A Novel Decision Tree and LSTM Powered Intelligent Agent System for Early Detection of Vegetable Plant Diseases
The early and precise diagnosis of vegetable plant diseases is crucial for sustainable agriculture since these diseases have a major impact on crop output and quality. Disease identification performance is examined in this work through a robust detection pipeline that examines the effects of several preprocessing techniques, class imbalance handling strategies, and deep learning classifiers. To better represent data and increase model knowledge of illness characteristics, the GLCM was used to extract texture information. By combining XGBoost with LSTM networks, a new hybrid model was created. While XGBoost is great at classifying structured data, the LSTM component is great at evaluating sequential data, which allows it to capture patterns and trends in the evolution of plant diseases over time. Better and more meaningful forecasts are made possible by this supplementary integration. By surpassing more conventional methods of illness classification, the suggested LSTM-XG model attained a remarkable prediction accuracy of 99.34%. An important factor in achieving this outcome was the use of hybrid modeling in conjunction with thorough preprocessing and correction of class imbalance. Finally, the LSTM-XG model shows great promise for practical use in precision farming. Its precision and efficiency in identifying illnesses in vegetable plants might facilitate prompt action, lessen crop loss, and encourage better farming methods. 2025 IEEE. -
A Novel Deep Learning Approach for Identifying Interstitial Lung Diseases from HRCT Images
Interstitial lung diseases (ILDs) are defined as a group of lung diseases that affect the interstitium and cause death among humans worldwide. It is more serious in underdeveloped countries as it is hard to diagnose due to the absence of specialists. Detecting and classifying ILD is a challenging task and many research activities are still ongoing. High-resolution computed tomography (HRCT) images have essentially been utilized in the diagnosis of this disease. Examining HRCT images is a difficult task, even for an experienced doctor. Information Technology, especially Artificial Intelligence, has started contributing to the accurate diagnosis of ILD from HRCT images. Similar patterns of different categories of ILD confuse doctors in making quick decisions. Recent studies have shown that corona patients with ILD also go on to sudden death. Therefore, the diagnosis of ILD is more critical today. Different deep learning approaches have positively impacted various image classification problems recently. The main objective of this proposed research work was to develop a deep learning model to classify the ILD categories from HRCT images. This proposed work aims to perform binary and multi-label classification of ILD using HRCT images on a customized VGG architecture. The proposed model achieved a high test accuracy of 95.18% on untrained data. 2022, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. -
A Novel Deep Learning Approach for Retinopathy Prediction Using Multimodal Data Fusion
In contemporary research on mild cognitive disorders (MCI) and Alzheimer's disease (AD), the predominant approach involves the utilization of double data modalities for making predictions related to AD stages. However, there is a growing recognition of the potential benefits that could be derived from the fusion of multiple data modalities to obtain a more comprehensive perspective in the analysis of AD staging. To address this, we have employed deep learning techniques to holistically assess data from various sources, including, genetic (single nucleotide polymorphisms (SNPs)), imaging (magnetic resonance imaging (MRI)), and clinical tests, with the objective of categorizing patients into distinct groups: AD, MCI, and controls (CN). For the analysis of imaging data, convolutional neural networks have been employed. Moreover, we have introduced a novel approach for data interpretation, enabling the identification of the most influential features learned by these deep models. This interpretation process incorporates clustering and perturbation analysis, shedding light on the crucial aspects of the data contributing to our classification results. Our experimentation, conducted on the dataset (i.e., ADNI), has yielded compelling results. Furthermore, our findings have underscored the significant advantage of integrating multi-modality data over solely relying on double modality models, as it has led to improvements in terms of accuracy, precision, recall, and mean F1 scores. 2024, Ismail Saritas. All rights reserved.