Browse Items (16481 total)
Sort by:
-
Detection of Alzheimers Disease Stages Based on Deep Learning Architectures from MRI Images
Acquiring, utilizing and storing information of any sort is known as memory. The power of memory makes the life of mankind to be more alive and reasonable. Thus, the loss of one such great capability is a rather painful phase of human life which can be destructed by multiple reasons such as diseases and disorders. One such disease is Alzheimers disease (AD). Alzheimers disease progressively damages brain cells and degrades mental activity that leads to mental illness. The accurate diagnosis of AD at earlier stages will help to prevent the disease before the brain gets damaged completely. In analyzing neurodegenerative disorders, neuroimaging plays an important role in diagnosing subjects with AD, mild cognitive impairment (MCI), and cognitively normal (CN). In this study, advanced deep learning (DL) architectures with brain imaging techniques were employed to maximize the diagnostic accuracy of the model developed. The proposed method works with convolutional neural networks (CNNs) to analyze the MRI input-output modalities. The method is evaluated using Alzheimers Disease Neuroimaging Initiative (ADNI) dataset. Binary classification is done on AD and MCI subjects from CN. This method is efficient to analyze multiple classes with a less amount of training data. 2023 selection and editorial matter, Jyotismita Chaki; individual chapters, the contributors. -
Detection of a new sample of Galactic white dwarfs in the direction of the Small Magellanic Cloud
Aims. In this study, we demonstrate the efficacy of the Ultraviolet Imaging Telescope (UVIT) in identifying and characterizing white dwarfs (WDs) within the Milky Way Galaxy. Methods. Leveraging the UVIT point-source catalogue towards the Small Magellanic Cloud and cross-matching it with Gaia DR3 data, we identified 43 single WDs (37 new detections), 13 new WD+main-sequence candidates, and 161 UV bright main-sequence stars by analysing their spectral energy distributions. Using the WD evolutionary models, we determined the masses, effective temperatures, and cooling ages of these identified WDs. Results. The masses of these WDs range from 0.2 to 1.3 M? and the effective temperatures (Teff) lie between 10 000 K to 15 000 K, with cooling ages spanning 0.1-2 Gyr. Notably, we detect WDs that are hotter than reported in the literature, which we attribute to the sensitivity of UVIT. Furthermore, we report the detection of 20 new extremely low-mass candidates from our analysis. Future spectroscopic studies of the extremely low-mass candidates will help us understand the formation scenarios of these exotic objects. Despite limitations in Gaia DR3 distance measurements for optically faint WDs, we provide a crude estimate of the WD space density within 1kpc of 1.3 10-3 pc-3, which is higher than previous estimates in the literature. Conclusions. Our results underscore the instrumental capabilities of UVIT and anticipate forthcoming UV missions such as INSIST for systematic WD discovery. Our method sets a precedent for future analyses in other UVIT fields to find more WDs and perform spectroscopic studies to verify their candidacy. The Authors 2024. -
Detection andDealing ofMulticollinearity Using aNew RidgeEstimator
The Ordinary Least Square (OLS) method is a widely used technique for estimating regression coefficients that describe the relationship between the independent variables and the dependent variable. The regression estimates obtained using this approach give a poor result in the presence of multicollinearity. Ridge regression is employed to address this problem. A ridge estimator, also called a biasing constant, plays a notable role in the parameter estimation of ridge regression. In this paper, we introduced a new ridge estimator and studied its properties. A comparative study of the proposed estimator with some of the existing ridge estimators under extreme multicollinearity has been illustrated by using a simulation technique. The estimator developed in this paper seems to perform better than all the other estimators because of the smaller ratio of average mean square error. The performance of the proposed estimator is analysed and verified using real-life data. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2026. -
Detection and Sensing of Human Body Micro-Motions Using 24GHz mm-Waves: A Case Study
The paper presents a 24 GHz millimeter-wave (mm-Wave) system for real-time human presence and distance detection, leveraging the 24 GHz band's balance of range, resolution, and power efficiency for applications in smart environments. The system uses Doppler-based reflections from signals to accurately detect presence and estimate distances with 12cm accuracy up to 5 meters with little needed infrastructure. The challenges presented by signal attenuation and multipath interference are addressed using beamforming and Massive MIMO with 5G-enabled IoT as the framework. The use of Ultra-Reliable Low Latency Communication and Massive-Machine-Type Communication allows a framework for rapid data processing and scalability. This system enables automated smart home lighting, healthcare occupancy detection, and security intrusion alert applications, with very high detection accuracies. The limits are reduced detection performance beyond a distance of 6 meters and interference from reflective surfaces. Future work includes investigating the 60 GHz bands, which will yield higher resolution in the detection, and using machine learning techniques to develop adaptive detection, as a scalable and cost effective method of real-time automation in a range of settings. 2025 IEEE. -
Detection and Robust Classification of Lung Cancer Disease Using Hybrid Deep Learning Approach
Effective lung cancer diagnosis and treatment hinge on the early detection of lung nodules. Various techniques, such as thresholding, pattern recognition, computer-aided diagnostics, and backpropagation calculations, have been explored by scientists. Convolutional neural networks (CNNs) have emerged as powerful tools in recent times, revolutionizing many aspects of this field. However, traditional computer-aided detection systems face challenges when categorizing lung nodule detection. Excessive reliance on classifiers at every stage of the process results in diminished recognition rates and an increased occurrence of false positives. To address these issues, we present a novel approach based on deep hybrid learning for classifying lung lesions. In this study, we explore multiple memory-efficient and hybrid deep neural network (DNN) architectures for image processing. Our proposed hybrid DNN significantly outperforms the current state-of-the-art, achieving an impressive accuracy of 95.21%, all while maintaining a balanced trade-off between specificity and sensitivity. The primary focus of this research is to differentiate between CT scans of patients who have early-stage lung cancer and those who do not. This is achieved by utilizing binary classification networks, including standard CNN, SqueezeNet, and MobileNet. 2023 IEEE. -
Detection and management of sucralose pollution: innovations in sustainable monitoring, mitigation, and energy utilization
Sucralose, a widely utilized artificial sweetener, has emerged as a potential environmental contaminant, persisting in aquatic environments and causing ecological risks. The use of suitable methods and sustainable materials can be applied for the detection of sucralose in wastewater, sludge, and other waste matrices Using sustainable materials and novel techniques for discerning sucralose. Leveraging waste-derived components, particularly in electrochemical sensing systems with competent nanomaterials can be a promising avenue for sucralose detection. Moreover, integrating waste-derived carbon materials like activated carbon or carbon black as electrodes enhances sensor sensitivity by augmenting organic molecule adsorption capacities, thus amplifying detection capabilities. Furthermore, waste-derived substrates such as cellulose fibers or agricultural residues can be engineered to serve as substrates or immobilization matrices in enzyme-based sensors for sucralose detection. While the electrochemical systems current output may not directly serve energy applications, its potential utilization for energy generation or storage, through technologies like electrochemical capacitors or batteries, underscores the broader sustainability objectives by repurposing energy from the detection process for additional beneficial applications. The amalgamation of sustainable methodologies in sucralose pollution detection and management not only contributes to environmental preservation but also aligns with the broader trajectory toward sustainable resource utilization and conservation. 2026 Elsevier Inc. All rights reserved. -
Detection and localization for watermarking technique using LSB encryption for DICOM Image
Watermarking is an effective way of transferring hidden data from one place to another, or proving ownership of digital content. The hidden data can be text, audio, images GIF etc., the data is embedded in a cover object usually an image or a video sequence. Usually the watermarking system(s) rely on their hidden aspect, as their primary security measure, once this is established that the cover object is counting some hidden data, then it is generally possible to recover the hidden information. The author proposed an in-genuine technique for DICOM color image water marking by combining Multi Quadrant LSB with truly random mixed key cryptography. This system provides a high level of security by just the water marking technique, as it breaks the cover image into up to four quadrants, & does LSB replacement of two bytes each quadrant. The bit sequence as the quadrant sequence can be randomized to increase the randomness, use of truly random mixed key cryptography, by using a pre shared, variable length, truly random, private key, turns hidden data into noise, which can only be recovered by having the private key. Thus, the proposed technique truly diminishes the probability of recovering hidden data, even if it is detected that something is hidden in cover object. 2022 Taru Publications. -
Detection and identification of un-uniformed shape text from blurred video frames
The identification and recognition of text from video frames have received a lot of attention recently, that makes many computer vision-based applications conceivable. In this study, we modify the picture mask and the original identification of the mask region convolution neural network and permit detection in three levels, including holistic, sequence, and at the level of pixels. To identify the texts and determine the text forms, semantics at the pixel and holistic levels can be used. With masking and detection, existences of the character and the word are separated and recognised. In addition, text detection using the results of 2-D feature space instance segmentation is done. Moreover, we explore text recognition using an attention-based optical character recognition (OCR) method with mask region convolution neural networks (R-CNN) to address and detect the problem of smaller and blurrier texts at the sequential level. Using attribute maps of the word occurrences in sequence to seq, the OCR method calculates the character sequence. At last, a fine-grained learning strategy is proposed to constructs models at word level using the annotated datasets, resulting in the training of a more precise and reliable model. The well-known benchmark datasets ICDAR 2013 and ICDAR 2015 are used to test our suggested methodology. 2024, Institute of Advanced Engineering and Science. All rights reserved. -
Detection and Classification of Thoracic Diseases in Medical Images using Artificial Intelligence Techniques: A Systematic Review
Artificial Intelligence is at the leading edge of innovation and is developing very fast. In recent studies, it has played a progressive and vital role in Computer-Aided Diagnosis. The chest is one of the large body parts of human anatomy and contains several vital organs inside the thoracic cavity. Furthermore, chest radiographs are the most commonly ordered and globally used by physicians for diagnosis. An automated, fast, and reliable detection of diseases based on chest radiography can be a critical step in radiology workflow. This study presents the conduction and results of a systematic review investigating Artificial Intelligence Techniques to identify Thoracic Diseases in Medical Images. The systematic review was performed according to PRISMA guidelines. The research articles published in English were filtered based on defined inclusion and exclusion criteria. The Electrochemical Society -
Detection and Classification of Potholes in Indian Roads Using Wavelet Based Energy Modules
Maintenance of roads is one the major challenge in the developed countries. The well maintained roads always indicates the economy of the whole country. The heavy use of roads, environmental conditions and maintenance is not performed regularly that leads the formation of potholes which causes the accidents and unwanted traffics. The paper discuss about the detection of potholes based on wavelet energy field. The proposed method mainly includes three phases (A)Wavelet energy filed is constructed in order to detect the image by using geometric criteria and morphological processing (B)Extracting Region of intersect by edge based segmentation technique (C)Classifying the potholes using Neural Network. 2019 IEEE. -
Detection and classification of lung cancer using deep neural network
Lung cancers hold a critical spot among the reasons for most cancer deaths among humans. The better way to maximise the survival rate is the detection of cancer at the earliest. But existing traditional techniques are time-consuming and error-prone. This study is a significant and efficient method for the detection and classification of lung cancer into large cell carcinomas, small cell, adenocarcinoma, squamous cell carcinomas, or benign respectively. In the proposed technique, a novel algorithm is implemented to generate the Image patches from whole slide histopathological images. Then, histogram normalisation is carried out to remove noise and enhance the image. Then a novel extended Mobius transformation technique is used for image augmentation. Finally, Dense EfficientNetB7 is trained to extract the features for the detection and classification of lung cancer. The performance of the proposed technique is proved more efficient and par with histologists attaining an accuracy of 98.87%. Copyright 2025 Inderscience Enterprises Ltd. -
Detection and Classification of Colorectal Polyp Using Deep Learning
Colorectal Cancer (CRC) is the third most dangerous cancer in the world and also increasing day by day. So, timely and accurate diagnosis is required to save the life of patients. Cancer grows from polyps which can be either cancerous or noncancerous. So, if the cancerous polyps are detected accurately and removed on time, then the dangerous consequences of cancer can be reduced to a large extent. The colonoscopy is used to detect the presence of colorectal polyps. However, manual examinations performed by experts are prone to various errors. Therefore, some researchers have utilized machine and deep learning-based models to automate the diagnosis process. However, existing models suffer from overfitting and gradient vanishing problems. To overcome these problems, a convolutional neural network- (CNN-) based deep learning model is proposed. Initially, guided image filter and dynamic histogram equalization approaches are used to filter and enhance the colonoscopy images. Thereafter, Single Shot MultiBox Detector (SSD) is used to efficiently detect and classify colorectal polyps from colonoscopy images. Finally, fully connected layers with dropouts are used to classify the polyp classes. Extensive experimental results on benchmark dataset show that the proposed model achieves significantly better results than the competitive models. The proposed model can detect and classify colorectal polyps from the colonoscopy images with 92% accuracy. 2022 Sushama Tanwar et al. -
Detection and Behavioral Analysis of Preschoolers with Dyscalculia
Human behaviours are influenced by various factors that might impact their thought process. The way human beings response in situations have a strong connection with genetic makeup, cultural values and experiences from the past. Behaviour Analysis discusses the effect of human response to external/internal stimuli. This study helps in understanding behaviour changes among individuals suffering from various psychological disorders. Dyscalculia is one similar type of learning disorder [LD] which is commonly found among individuals and goes undetected for years. It is a lifelong condition which causes difficulty for people to perform mathematics-related tasks. Dyscalculia is quite eminent at every age. Since the symptoms are prominent from a young age, it can be detected at the earliest. Dyscalculia has no medical treatment but can be minimized by getting involved in some brain exercises especially created for children with Learning Disabilities. The chapter deals with minor research and the behaviour analysis for the above-mentioned disorder among pre-schoolers. In this chapter, a study of the behavioural patterns of pre-schoolers with dyscalculia is performed. This chapter also attempts to propose a model that can detect and predict the possibility of a child suffering from dyscalculia. It also includes a number of brain training activities that can help them to improve and enhance their confidence in mathematics. 2020, Springer Nature Switzerland AG. -
Detection and analysis of android malwares using hybrid dual Path bi-LSTM Kepler dynamic graph convolutional network
In past decade, the android malware threats have been rapidly increasing with the widespread usage of internet applications. In respect of security purpose, there are several machine learning techniques attempted to detect the malwares effectively, but failed to achieve the accurate detection due to increasing number of features, more time consumption decreases in detection efficiency. To overcome these limitations, in this research work an innovative Hybrid dual path Bidirectional long short-term memory Kepler dynamic graph Convolutional Network (HBKCN) is proposed to analyze and detect android malwares effectively. First, the augmented abstract syntax tree is applied for pre-processing and extracts the string function from each malware. Second, the adaptive aphid ant optimization is utilized to choose the most appropriate features and remove irrelevant features. Finally, the proposed HBKCN classifies benign and malware apps based on their specifications. Four benchmark datasets, namely Drebin, VirusShare, Malgenome -215, and MaMaDroid datasets, are employed to estimate the effectiveness of the technique. The result demonstrates that the HBKCN technique achieved excellent performance with respect to a few important metrics compared to existing methods. Moreover, detection accuracies of 99.2%, 99.1%,99.8% and 99.8% are achieved for the considered datasets, respectively. Also, the computation time is greatly reduced, illustrating the efficiency of the proposed model in identifying android malwares. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2024. -
Detection and analysis of android malwares using hybrid dual Path bi-LSTM Kepler dynamic graph convolutional network
In past decade, the android malware threats have been rapidly increasing with the widespread usage of internet applications. In respect of security purpose, there are several machine learning techniques attempted to detect the malwares effectively, but failed to achieve the accurate detection due to increasing number of features, more time consumption decreases in detection efficiency. To overcome these limitations, in this research work an innovative Hybrid dual path Bidirectional long short-term memory Kepler dynamic graph Convolutional Network (HBKCN) is proposed to analyze and detect android malwares effectively. First, the augmented abstract syntax tree is applied for pre-processing and extracts the string function from each malware. Second, the adaptive aphid ant optimization is utilized to choose the most appropriate features and remove irrelevant features. Finally, the proposed HBKCN classifies benign and malware apps based on their specifications. Four benchmark datasets, namely Drebin, VirusShare, Malgenome -215, and MaMaDroid datasets, are employed to estimate the effectiveness of the technique. The result demonstrates that the HBKCN technique achieved excellent performance with respect to a few important metrics compared to existing methods. Moreover, detection accuracies of 99.2%, 99.1%,99.8% and 99.8% are achieved for the considered datasets, respectively. Also, the computation time is greatly reduced, illustrating the efficiency of the proposed model in identifying android malwares. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2024. -
Detecting the magnitude of depression in Twitter users using sentiment analysis
Today the different social networking sites have enabled everyone to easily express and share their feelings with people around the world. A lot of people use text for communicating, which can be done through different social media messaging platforms available today such as Twitter, Facebook etc, as they find it easier to express their feelings through text instead of speaking them out. Many people who also suffer from stress find it easier to express their feelings on online platform, as over there they can express themselves very easily. So if they are alerted beforehand, there are ways to overcome the mental problems and stress they are suffering from. Depression stands out to be one of the most well known mental health disorders and a major issue for medical and mental health practitioners. Legitimate checking can help in its discovery, which could be useful to anticipate and prevent depression all-together.Hence there is a need for a system, which can cater to such issues and help the user. The purpose of this paper is to propose an efficient method that can detect the level of depression in Twitter users. Sentiment scores calculated can be combined with different emotions to provide a better method to calculate depression scores. This process will help underscore various aspects of depression that have not been understood previously. The main aim is to provide a sense of understanding regarding depression levels in different users and how the scores can be correlated to the main data. 2019 Institute of Advanced Engineering and Science. All rights reserved. -
Detecting Student Depression Using Non-Clinical Measures with Explainable Predictive Modeling
Depression among students is a serious global mental health concern, affecting academic performance, emotional wellbeing, and long-term development. While traditional diagnostic tools like self-reported questionnaires and clinical interviews are useful, they often suffer from subjectivity, recall bias, and limited scalability. This study introduces a data driven, interpretable machine learning approach to predict student depression using both academic and nonacademic factors, without relying on clinical indicators. The dataset comprises student information on demographics, academic workload, lifestyle habits, social interactions, financial stress, and emotional state. Following thorough preprocessing including handling missing values, encoding variables, correlation-based feature reduction, and SMOTE to address class imbalance, ten supervised machine learning models were trained and assessed. Among them, a SMOTE enhanced XGBoost model achieved the highest test ROC AUC score of 0.95. To maintain transparency, SHAP (Shapley Additive Explanations) was employed to interpret the model's predictions, highlighting key risk factors such as academic pressure, poor sleep quality, financial difficulties, and low social support. These findings can help guide early interventions and build trust with stakeholders. Future work may involve incorporating longitudinal and multimodal data, deploying real time solutions in educational settings, and addressing ethical considerations around privacy, fairness, and consent in AI based mental health systems. 2025 IEEE. -
Detecting Infectious Disease Based on Social Media Data Using BERT Model
Seasonal diseases are those diseases that are widespread during a particular time of the year including monsoons, winter etc. In the absence of preventative measures, the human race remains vulnerable to the hazardous effects of seasonal diseases following regular patterns of increased inci- dence and transmission which remains a global concern. Dengue, Influenza, etc. are such types of diseases where every year many people get affected globally. The primary focus of this research paper is to understand the opinion of people regarding the seasonal diseases. The research paper covers sentiment analysis on textual data from social media where people have vocalized their sentiments or thinking regarding seasonal diseases and seasonal infectious diseases. Influenza, Dengue, Malaria, Japanese Encephalitis, and Chikungunya are the seasonal diseases that have been covered in this research paper. To achieve this, the language model Bidirectional Encoder Representations from Transformers (BERT) was used to verify the sentiments about the seasonal diseases. The result of the investigation hold the potential to significantly enhance our comprehension of societal sentiments, discerning between states of tranquility and concern among individuals. The outcome of the study will help healthcare department to plan the necessary actions. 2024 IEEE. -
Detecting Fake Information Dissemination using Leveraging Machine Learning and DRIMUX with B-LSTM
Information integrity and public confidence are seriously threatened by the rapid expansion of fake news and misinformation that has resulted from the online broadcast of information. This work focuses on the detection of fraudulent information propagation utilizing machine learning techniques and the Digital Reputation and Influence Measurement Unit (DRIMUX) in order to address this problem. The use of Bidirectional Long Short-Term Memory (B-LSTM) networks into the detection process is something we really advocate. B-LSTM enables the capture of contextual dependencies from both past and future time steps, enhancing the understanding of sequential data. Additionally, DRIMUX provides reputation and influence measurements to assess the credibility of information sources. Experimental analyses on various datasets reveal the promising performance of the suggested methodology, highlighting its potential in preventing the spread of false information and protecting the veracity of digital information. 2024, Ismail Saritas. All rights reserved. -
Detecting Deepfake Voices Using a Novel Method for Authenticity Verification in Voice-Based Communication
The widespread use of deepfake technology in recent years has given rise to grave worries about the alteration of audio-visual material. The integrity of voice-based communication is particularly vulnerable to the threat posed by deepfake voice synthesis. The development of cutting-edge methods for the identification of deepfake voices is examined in this paper, which also offers a thorough analysis of current approaches, their advantages, and disadvantages. The research presents a novel method for detecting deepfakes in voice recordings that uses signal processing, machine learning, and audio analysis to separate synthetic voices from authentic voices. By achieving superior accuracy in differentiating between real and deepfake voices, and proposed method supplies a strong barrier against the misuse of voice synthesis technology for malicious purposes, also go over the research some of the possible uses for this technology, like voice authentication system security and social media platform content moderation. The paper's insights will support continued efforts to strengthen the authenticity of voice communication in the digital age and reduce the risks associated with deepfake voice synthesis. The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.