Browse Items (16488 total)
Sort by:
-
Rectifying Whole Brain Segmentation Errors Using a Novel Under-Segmentation Correction Method
Pre-processing is a critical step in any data-driven study, particularly in the field of medical imaging, where it significantly enhances the reliability of disease and disorder diagnosis. In this context, medical image segmentation allows for more precise data analysis by isolating the regions of interest. Accurate segmentation of these regions can reveal influential variabilities in analysis, potentially leading to unique scientific findings. This article presents a novel under-segmentation error correction technique specifically designed for whole-brain segmentation. Additionally, it performs a set of pre-processing steps for the structural magnetic resonance imaging (sMRI) images, which are necessary to maintain the structural integrity and uniformity of MRI scans across different subjects. The proposed algorithm effectively eliminates under-segmentation errors, thereby improving the accuracy of whole-brain segmentation, particularly for structurally intact brain images. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025. -
FaithfulNet: An explainable deep learning framework for autism diagnosis using structural MRI
Explainable Artificial Intelligence (XAI) can decode the black box models, enhancing trust in clinical decision-making. XAI makes the predictions of deep learning models interpretable, transparent, and trustworthy. This study employed XAI techniques to explain the predictions made by a deep learning-based model for diagnosing autism and identifying the memory regions responsible for children's academic performance. This study utilized publicly available sMRI data from the ABIDE-II repository. First, a deep learning model, FaithfulNet, was developed to aid in the diagnosis of autism. Next, gradient-based class activation maps and the SHAP gradient explainer were employed to generate explanations for the model's predictions. These explanations were integrated to develop a novel and faithful visual explanation, Faith_CAM. Finally, this faithful explanation was quantified using the pointing game score and analyzed with cortical and subcortical structure masks to identify the impaired brain regions in the autistic brain. This study achieved a classification accuracy of 99.74% with an AUC value of 1. In addition to facilitating autism diagnosis, this study assesses the degree of impairment in memory regions responsible for the children's academic performance, thus contributing to the development of personalized treatment plans. 2025 Elsevier B.V. -
The effect of spatial and intensity level augmentation of structural magnetic resonance images on autism diagnosis model
In deep learning, the robustness and generalizability of models significantly depend on diverse and heterogeneous training data. Acquiring such an extensive dataset is challenging in fields like disorder prediction due to data scarcity, which can be attributed to factors such as privacy concerns, limited patient population, or inadequate facilities. Data augmentation can be an ideal solution to this problem, particularly in the field of disorder prediction, like autism, using medical imaging. Data augmentation can expand and balance datasets by generating high-quality and varied data, thereby improving the generalizability of deep learning models. This study proposed two types of augmentation methods: 1. Spatial level 2. Intensity level augmentation techniques. Eight different levels of augmentations were experimented with across these categories. This study found that the combination of spatial and intensity level augmentations enhanced the model's generalizability and robustness, achieving an AUC value of 0.7433. Additionally, it was observed that the Left to Right flip method, under spatial augmentation, diminished the model's performance, whereas random noise injection, under intensity level augmentation, improved prediction accuracy. 2026 Elsevier B.V. -
Diagnosis of Autism Spectrum Disorder: A Review of Three Focused Interventions
Autism is a neurological developmental disorder that impacts a persons physical, social, and emotional behavior. This disorder develops over time and is characterized by social deficits and repetitive behavior. Although there is no cure for this disorder, an early diagnosis and intervention can do significant wonders and can help the subject to become active functioning members of the family and society. The aim of this study is to minimize the diagnostic period by finding an optimal diagnosis procedure from the existing diagnosis tools. The diagnosis of autism can be done in three ways: 1. clinical evaluation; 2. screening tools; 3. brain images. In this review paper, we have thoroughly gone through all three types of diagnostic procedures and found that there was no single diagnostic tool to confirm the disorder. We also found that the diagnosis period was too long. As the result of this review, we found an ASD diagnosis triad which helps to choose the right diagnosis procedure based on the subjects age which reduces the diagnostic period and helps to aid early diagnosis by eliminating the chaos in choosing the diagnostic tools. 2022, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. -
Explaining Autism Diagnosis Model Through Local Interpretability Techniques - A Post-hoc Approach
In this era of machine learning and deep learning algorithms dominating the Artificial Intelligence (AI) world, the trustworthiness of these black box models is still questionable. Life-caring sectors like healthcare and banking make use of these black box models as assistance in critical decision-making processes, but the degree of reliability of these decisions is still uncertain. This is because these black box models will not reveal the causation of the predicted outcome. However, creating an interpretable model that can explain the internal workings of these black box models can provide some reliable insights and trustable justifications for the predicted outcome. This study aimed to create an interpretable model for autism diagnosis which can give some trustable explanations for its predicted outcome. Using local interpretability methods such as LIME, SHAP, and Anchors the predicted outcome for each instance is explained well with some standard visual representations. As a result, this study developed an interpretable autism diagnosis model with an accuracy rate of 91.37% and with good local model explanations. 2023 IEEE. -
Practical Benefits of Using AI for More Accurate Forecasting in Mental Health Care
Artificial Intelligence (AI) is the general term for being able to make computers do things that require human-like intelligence. AI is the novel idea of the computer pioneers like Alan Turning and John von Neumann in the 1940s. Their novel intuition towards making machines think is the key start for this AI technology evolution. As shown in Fig. 1, the first milestone of AI happened in the year 1956 when it was proved by a group of researchers that a machine could solve any problem with the use of an unlimited amount of memory. Here they named this program General Problem Solver (GPS). 2024 Sachi Nandan Mohanty, Preethi Nanjundan and Tejaswini Kar. -
Vocational training for women empowerment: Saint Kuriakose Elias Chavaras vision
The present study attempted to describe the initiatives of Saint (St.) Kuriakose Elias Chavara on vocational training and womens empowerment. The study narrated the present condition of one of his vocational training initiatives known as rosary-making, which is in vogue at Koonammavu Village in Kerala state, India. The study employed a multi-method research design approach to carry out the present study. It included historical, qualitative, and quantitative methods in sequence. In the historical method, the study employed document analysis of primary and secondary sources. As a part of the qualitative method, the study conducted a semi-structured interview with 10 rosary-making entrepreneurs in Koonammavu. As a part of the quantitative method, the study administered a questionnaire to 100 families who are actively involved in the rosary-making business. Document analysis revealed that St. Chavaras initiative on rosary-making vocational training for nuns was a contribution to womens empowerment. Narrative thematic analysis revealed 5 main themes and 10 subthemes in Chavaras contribution. Quantitative data revealed that rosary-making emerged as a livelihood, business, and source of income for many families. The study recommends future researchers focus on all the initiatives of St. Chavara in the realm of vocational training and womens empowerment. 2025, Intelektual Pustaka Media Utama. All rights reserved. -
Legacy of St. Kuriakose Elias Chavara and CMI and CMC congregations: educational modernity in Kerala
Education plays a predominant role in modernising society. It has the power to transform society by dismantling out-dated superstitions and promoting progressive social change. The present study explores the educational initiatives of St. Chavara and his followers, the CMI and CMC, and their role in shaping the modernity of nineteenth century Kerala, India. The study highlights how St. Chavaras visionary leadership in education contributed to modernisation in Kerala. The study focuses on his key initiatives such as secular education, inclusive education, women education, universal education and new educational support programmes which led to modernity of Kerala. The study reveals that St. Chavara's educational initiatives had a profound impact on both individuals and society of nineteenth century Kerala. The study emphasises the continued impact of St. Chavaras educational legacy through the institutions led by his followers, the CMI and CMC, which played a significant role in driving the modernisation of Kerala. 2025 Informa UK Limited, trading as Taylor & Francis Group. -
Geometric aspects of noncommutative wormholes with conformal symmetry
This paper investigates traversable wormhole solutions within the framework of (Formula presented) f(Q,T) gravity by incorporating conformal symmetry and employing a Lorentzian distribution to model the matter sources. The study considers different equations of state, such as traceless and barotropic forms, to explore their impact on the viability of wormhole solutions. A comprehensive analysis of the effects of the model parameters on the wormhole geometry and its physical properties is carried out. The findings reveal that the resulting shape function satisfies all the necessary wormhole conditions. Importantly, certain scenarios are identified where the wormhole can be supported by non-exotic matter, highlighting the physical plausibility of such solutions within modified gravity. 2026 IOP Publishing Ltd. All rights, including for text and data mining, AI training, and similar technologies, are reserved. This article is available under the terms of the https://publishingsupport.iopscience.iop.org/iop-standard/v1. -
IoT-Based Emergency Vehicle Detection Using YOLOv8
The rapid response of emergency services plays a critical role in saving lives and minimizing the impact of emergencies. However, identifying and locating emergency vehicles in real-Time can be challenging, especially in congested urban areas. This paper focuses on the emergency vehicle identification using the You Only Look Once version 8 (YOLOv8) algorithm and is focused on Internet of Things (IoT). The goal of this research is to develop a real-Time and precise emergency vehicle detection system using You Only Look Once version 8 (YOLOv8) algorithm, trained and tested with a dataset from a camera placed on a busy road, to enhance emergency service response times. The findings demonstrate the suggested system's ability to recognize emergency vehicles at a speed of 31 frames per second and with a 95% accuracy rate. Modern object identification algorithms include the You Only Look Once version 8 (YOLOv8) algorithm, which has shown promising results in various applications. The proposed system is built on a Raspberry Pi, which acts as an edge device and processes the video stream in realtime. The system consists of an Internet of Things (IoT) device with a camera that captures the live video stream, which is then fed into the algorithm for object detection. Once an emergency vehicle is detected, the system sends an email notification to the nearby emergency services, like a police station, using Simple Mail Transfer Protocol (SMTP), who can then take appropriate action. The results of this investigation show that the Internet of Things and You Only Look Once version 8 (YOLOv8) algorithms have great promise for creating effective and dependable emergency vehicle detection systems. The proposed system possesses the capacity to save lives and improve the effectiveness of emergency response by speeding up response times for emergency services. The suggested solution is also inexpensive, simple to implement, and adaptable to existing infrastructure. Through the development of intelligent transportation systems, emergency services can operate more safely and effectively. More sophisticated machine learning algorithms may be incorporated into the proposed system, and further sensors can be added to utilize alternative methods beyond camera-based detection to identify emergency vehicles. Overall, this research shows the potential of Internet of Things (IoT) and machine learning in creating creative emergency services solutions. 2025 Syed Suhana et al.Published by Sciendo. -
Smart Street Lighting System Using AI, Computer Vision, and IoT for Energy Efficiency and Sustainability
In pursuit of smart cities, optimum utilization of energy without any compromise on public security is a requirement. Traditional lighting systems work with minimal flexibility but result in enormous wastage of energy. The proposed model dynamically controls lighting intensity based on real-time pedestrian and environmental information. The model is capable of saving wasteful utilization of electricity by integrating innovative technologies without any compromise on public security and visibility levels. The system requires the deployment of IoT-enabled cameras at strategic points on the highways to monitor the movement and traffic of pedestrians and automobiles. The system is capable of detecting and recognizing moving objects in real-time through computer vision technologies. Based on movement and density patterns, the system determines the optimal level of brightness to be used over each street lighting segment. This adaptive lighting method ensures the utilization of energy only where it is required and minimizes the energy consumption substantially without any compromise on security. Overall, this AI-driven smart lighting solution is a cost-effective and scalable solution for cities to become intelligent and resource-efficient. 2025 IEEE. -
A novel approach to optimize power utilization and scheduling in dynamic networks through generative adversarial network-based prediction of network parameters
The infrastructure-less network communication has been in an ever-increasing demand to cater to the needs of effective communication while the network dynamism exists. The quality of service (QoS)quality of service (QoS) demands increasing the efficiency of network by reducing the time taken for a data packet to reach the destination, increasing the probability of successful data transmissiondata transmission, minimizing packet loss,packet loss and optimizing power utilizationpower utilization. In this study, a generative adversarial network-based learning modelgenerative adversarial network-based learning model has been developed that considers the previous network statistics, as realized data, to predict future network patterns by the generatorgenerator to make such predictions, called as unrealized data, as near to the realized data. Further, the proposed model uses penalty-award criteria by the discriminatordiscriminator, to fine-tune the predicted network parameters. Now, having the set of realized and unrealized data, the model uses Markov decision processMarkov decision process to perform power scheduling and effective utilization of buffer space. The buffer utilization in the intermediate nodes necessitates the model to stochastically schedule the data transmission, depending on the percentage of utilization of buffer. Simulation results denote the effective utilization of buffer that makes continued transmission of data, whenever possible, without having data packet lossdata packet loss. Also, power scheduling, by the use of goodput function and increased transmission probability improves the power utilization that ultimately increases the lifetime of the network. 2026 Walter de Gruyter GmbH, Berlin/Boston, Genthiner Stra 13, 10785 Berlin. -
Millets Industry Dynamics: Leveraging Sales Projection and Customer Segmentation
Millets delves into the dynamics of the millets industry, with a particular focus on sales projection and customer segmentation as strategic levers for growth. The research commences with an in-depth analysis of the millets market, encompassing production patterns, consumption trends, and emerging market opportunities. It explores the diverse range of millets varieties, their nutritional profiles, and the factors driving consumer preference. By understanding the market landscape, the study identifies key trends and challenges shaping the industry. A core component of this research is the development of a robust sales projection model. Employing advanced statistical and data-driven techniques, the model forecasts future sales based on historical data, market trends, and relevant economic indicators. The model incorporates factors such as consumer demographics, purchasing behavior, and competitive landscape to provide accurate and actionable insights. Customer segmentation is another critical aspect of the study. By applying clustering and profiling methodologies, the research identifies distinct customer segments based on factors such as age, income, dietary preferences, and purchasing habits. This segmentation enables a deeper understanding of customer needs and preferences, facilitating targeted marketing strategies and product development. The integration of sales projection and customer segmentation empowers businesses to make informed decisions, optimize resource allocation, and enhance overall market performance. By aligning product offerings and marketing efforts with customer segments, companies can achieve higher customer satisfaction, increased market share, and improved profitability. This research contributes to the growing body of knowledge on the millets industry by providing valuable insights into market dynamics, sales forecasting, and customer segmentation. The findings offer practical guidance for industry stakeholders, including farmers, processors, retailers, and policymakers, in navigating the evolving millets landscape. By leveraging the potential of sales projection and customer segmentation, the millets industry can unlock new opportunities and achieve sustainable growth. The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024. -
Facial Emotion Recognition Augmented with CNNs and Face Detection: Toward Emotive Emoji Synthesis
Emotion recognition is a crucial component with broad applications in technology and healthcare industries specifically in humancomputer interaction. To improve emotion recognition accuracy, this research introduces an innovative technique that integrates face detection with Convolutional Neural Networks (CNNs). Using the Fer2013 dataset, the approach consists of carefully identifying faces in images as a preprocessing step, followed by training a CNN network to identify emotions and create corresponding emojis. After conducting extensive testing and assessment, it is determined that after employing a face detection algorithm the suggested framework is effective in both correctly identifying emotions and producing visually appealing emojis. This helps to create an interface for emotional communication that is more user-friendly and captivating. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025. -
Utilizing Machine Learning for Sport Data Analytics in Cricket: Score Prediction and Player Categorization
Cricket is a popular sport with complex gameplay and numerous variables that contribute to team performance. In recent years, sports analytics has gained significant attention, aiming to extract valuable insights from large volumes of cricket data. Cricket has many fans in India. With a strong fan following, many try to use their cricket intuition to predict the outcome of a match. A set of rules and a points system govern the game. The venue and the performance of each player greatly affect the outcome of the match. The game is difficult to predict accurately as the various components are closely related. The CRR (Current Run Rate) approach is used to predict the final score of the first innings of a cricket match. Total points are calculated by multiplying the average number of runs scored in each over by the total number of overs. For ODI cricket, these methods are useless as the game can change very quickly regardless of the current run rate. The game may be decided by 1 or 2 overs. For more accurate score predictions, a system is needed that can more accurately predict the outcome of an inning. This research paper explores the application of machine learning techniques to predict scores and classify players based on their roles in the squad. The study utilizes a comprehensive dataset comprising various attributes of cricket matches, including player statistics, match conditions, and historical performance. Linear Regression, Logistic Regression, Naive Bayes, Support Vector Machines (SVM), Decision Tree, and Random Forest regression models are employed to predict scores. Additionally, player categorization is performed using a classification approach. The results demonstrate the effectiveness of machine learning techniques in enhancing performance analysis and decision-making in the game of cricket. 2023 IEEE. -
High-performance 2D photonic crystal sensor for simultaneous detection of chemical and biological analytes
This work proposes a two-dimensional (2D) photonic crystal based sensor for multi-analyte detection across biochemical and biological domains. The proposed sensor can detect sulfuric acid, hydrogen peroxide, salt concentrations in seawater, alcohol detection, cancer cell detection, and aberrant bone tissue without any external modification to the sensor configuration. The intriguing aspect of the proposed sensor is that its structural parameters are optimised to detect refractive index values in the range of 12, and the resonant wavelength is therefore shifted to a longer wavelength. Based on the shift, the performance parameters of the analyte are observed. The proposed sensor offers excellent performance metrics, including a high transmission efficiency of 100%, a high-quality factor of 1,600, a sensitivity of 315nm/RIU, and a maximum detection limit of 0.09. The footprint of the proposed design is 16?m. This makes the sensor suitable for photonic integrated circuits and lab-on-chip applications. It offers a promising platform for next-generation optical sensing technologies. The Author(s), under exclusive licence to The Optical Society of India 2025. -
-
Enhancing the job scheduling procedure to develop an efficient cloud environment using near optimal clustering algorithm
In this internet era, cloud computing and there are various problems in the cloud computing, where the consumers as well as the service providers facing in their day to day cloud activities. Job scheduling problem plays a vital role in the cloud environment. To provide an efficient job scheduling environment, it is necessary to perform efficient resource clustering. In this regard, the proposed system, concentrated on the resource clustering methodology by proposing an efficient resource clustering algorithm named identicalness split up periodic node size (ISPNS) in the cloud environment. The algorithm proposed helps in forming resource clusters with the help of cloud environment. The proposed system is compared with the existing systems to justify the performance of the proposed resource clustering algorithm and it produces near optimal solution for the resource clustering problem which helps to provide an efficient job scheduling in cloud environment. Copyright 2023 Inderscience Enterprises Ltd. -
Skill sets required to meet a human-centered Industry 5.0: A systematic literature review and bibliometric analysis
The first industrial revolution, known as Industry 1.0, was primarily concerned with mechanical engineering and water and steam. Electric power systems and mass production assembly lines were established during the second industrial revolution (Industry 2.0). The third industrial revolution (Industry 3.0) was defined as automatic manufacturing and the incorporation of electronics, computers, and information technology into manufacturing. The fourth industrial revolution (Industry 4.0) is automating business operations and advancing manufacturing to a level based on connected devices, smart factories, cyberphysical systems (CPS), and the internet of things (IoT), where machines will change how they interact with one another and carry out specific tasks. Industry 5.0, with all modern technologies, is aimed to be a harmonious balance between human and machine interaction, and has an emphasis on sustainable growth. The present study uses an interpretive-qualitative research method to review the skill sets required to meet a human-centered Industry 5.0. 2024, IGI Global. All rights reserved. -
Enablers and Outcomes of Supply Chain Collaboration for Sustainable Growth
This study explores the intricate dynamics, challenges, and potential benefits of supply chain collaboration, emphasizing its pivotal role in achieving sustainability goals. Modern Supply Chain Collaboration (SCC) projects focus on sustainability-related activities, fostering interdependence between partners and driving sustained competitive advantage. The study introduces a comprehensive framework encompassing specific enablers (e.g., Joint Decision Making, Technology Integration) and outcomes (e.g., Social, Economic, and Environmental Sustainability) of supply chain collaboration. It contributes to practical guidelines for businesses seeking to enhance collaboration strategies and delves into theoretical paradigms such as the Cooperative Advantage concept, Triple Bottom Line Theory, Resource-Based View Theory, and Network Theory. The Triple Bottom Line Theory serves as an integrated theory of sustainability, emphasizing economic advantages, environmental impact minimization, and societal benefits. The Resource-Based View Theory underscores the role of internal resources in gaining competitive advantages, aligning with sustainability goals. Network Theory explores collaborative dynamics among competing entities, emphasizing resource sharing. The study's findings offer practical implications, enabling companies to assess and improve the sustainability of their supply chain management. It advocates for the integration of supply chain collaboration into organizational missions, emphasizing the importance of trust-building through standardized guidelines. The insights gained from this study are applicable across sectors, aiding legislators in developing flexible regulations and refining collaboration processes. Additionally, the study highlights the potential cultural variations in supply chain collaboration, paving the way for future research. 2024, Iquz Galaxy Publisher. All rights reserved.