Category Archives: Uncategorized

Authors: Sujit Kumar, Dr. Sanjay Kumar Singh, Dr. Niranjan Kumar Mandal

Abstract: Coal mining can influence riverine water quality through runoff, sediment transport, and the movement of dissolved constituents. This comparative study examines seasonal water-quality patterns reported in two related investigations of the Ajay River system: one covering two canals and the other covering three barrages. Both studies considered the rainy, winter, and summer seasons and reported physicochemical parameters, selected metals, chemical oxygen demand (COD), biochemical oxygen demand (BOD), dissolved oxygen (DO), and total coliform counts. The canal results showed rainy-season turbidity of 20 NTU in both canals, while winter turbidity differed sharply between Canal 1 (0 NTU) and Canal 2 (25 NTU). In the barrage study, turbidity ranged from 17–20 NTU during the rainy season, fell to 0 NTU in winter, and rose to 22–35 NTU in summer. The barrage sites also showed higher summer COD (8–13 mg/L), BOD (3–4 mg/L), and coliform counts (2.3–2.7 × 102 CFU/100 mL) than in the rainy season. Across both studies, reported concentrations of most measured metals were low or below the stated detection limits, although iron varied in the barrage samples. The comparison indicates that seasonal conditions and the type of water body may shape the observed water-quality profile. However, the studies do not provide sufficient site-level information or statistical analyses to establish that coal mining alone caused the observed differences. Consistent sampling, clearly documented analytical methods, and expanded metal and microbial monitoring are recommended.

DOI: https://doi.org/10.5281/zenodo.23417929

Published by: vikaspatanker

Authors: Pradeepa S, Anamika S S, Mariammal alias Madhu M, Oviya S

Abstract: Facial expressions provide useful visual cues for human–computer interaction, but automatic interpretation remains challenging because expressions vary across individuals, lighting conditions, head pose, and image quality. This paper presents a face emotion recognition system that combines computer vision with a convolutional neural network (CNN) to classify facial expressions from images and live video. The system detects a face using OpenCV’s Haar Cascade face detector, extracts and preprocesses the facial region, and passes a 48 × 48 grayscale image to a CNN trained using the FER-2013 dataset. The network uses convolutional blocks, batch normalization, max pooling, dropout, and a softmax output layer to classify seven expression categories: angry, disgust, fear, happy, neutral, sad, and surprise. The implementation also integrates separate pretrained age-group and gender models, allowing the application to display multiple facial attributes in a unified interface. Emotion confidence scores and bounding boxes are overlaid on the detected face, while temporal averaging across recent frames is used to reduce rapid changes in predictions. The project demonstrates an end-to-end facial analysis pipeline using Python, TensorFlow/Keras, and OpenCV. Quantitative performance should be reported from the final held-out test run; the project report supplied for this paper does not provide a verified final test-results table, so no unverified performance claim is made here.

DOI: https://doi.org/10.5281/zenodo.23417701

Published by: vikaspatanker

Authors: Anbazhagan R, Dr. Susitha N

Abstract: Pregnancy-related complications such as gestational diabetes, preeclampsia, preterm birth, and fetal distress remain major causes of maternal and perinatal morbidity, and timely detection depends on regular, accurate monitoring that many women cannot easily access. Advances in wearable sensing, mobile health, and artificial intelligence (AI) have made continuous, data-driven pregnancy monitoring technically feasible. This paper presents a structured narrative review of AI-based pregnancy health monitoring, drawing mainly on recent systematic, scoping, and narrative reviews together with selected primary studies. It examines the data sources and sensing modalities used, the machine learning and deep learning techniques applied, and the main application areas: gestational diabetes and preeclampsia prediction, fetal monitoring through cardiotocography, preterm birth and imaging, and wearable and IoT-based remote monitoring.The review finds that AI methods show consistent promise across these areas, but that the evidence base is limited by small single-centre datasets, geographic concentration of research, class imbalance, scarce external validation, and an unresolved tension between accuracy and interpretability. Privacy, sensor accuracy, and user compliance further constrain real-world deployment. The paper proposes a five-layer reference architecture for an AI-based pregnancy health monitoring system and outlines research directions relevant to resource-constrained settings such as India.

DOI: https://doi.org/10.5281/zenodo.23417250

Published by: vikaspatanker

Authors: Dr. Anil Karadwal

Abstract: Distributed cloud–edge systems support modern applications by combining centralised cloud infrastructure with computing resources located near data sources. Despite their advantages, these environments remain vulnerable to hardware faults, software failures, network disruptions, resource contention, and changing workloads. Conventional fault-tolerance mechanisms often respond after a failure has occurred, potentially increasing service interruption and recovery costs. Artificial Intelligence (AI) and Digital Twin (DT) technology offer opportunities to anticipate system degradation, evaluate alternative responses, and support automated recovery. This paper proposes a conceptual framework integrating digital twins, AI-based fault prediction, adaptive resource management, and self-healing mechanisms for distributed cloud–edge environments. The framework comprises six components: real-time monitoring, digital representation, predictive analytics, risk-aware decision-making, self-healing execution, and continuous feedback. It also identifies challenges relating to data quality, prediction uncertainty, model drift, computational overhead, security, and coordination across distributed nodes. A proposed evaluation methodology outlines measurable indicators, including prediction performance, service availability, recovery time, false-alarm rate, and resource overhead. The paper contributes a structured design approach for future investigation of predictive and autonomous fault management. As a conceptual study, it does not claim experimental validation or quantified performance improvements. Its proposed framework requires implementation and empirical evaluation under representative cloud–edge workloads.

DOI: https://doi.org/10.5281/zenodo.23414685

Published by: vikaspatanker

Authors: Mrs. Rajeswari M, Dr. Monika G

Abstract: Ultrasound is the first-line imaging modality for assessing the ovary, both for monitoring follicles in fertility care and for characterising adnexal and ovarian tumours. Interpretation, however, depends on operator skill and is affected by speckle noise, low contrast, and ambiguous boundaries. Deep learning, and in particular the U-Net family of encoder-decoder networks, has become the dominant approach for automating segmentation of the ovary, follicles, and masses. This paper reviews the literature on automated follicular and ovarian tumour analysis from ultrasound using U-Net variants. It summarises the clinical background, the architectural variants (plain, attention, nested, self-configuring, recurrent, and transformer-hybrid U-Nets), their application to follicle segmentation and adnexal mass segmentation, and the downstream tasks of quantification and classification. The reviewed studies report high overlap with expert outlines, for example Dice coefficients of 0.912 for the ovary and 0.858 for follicles with a composite U-Net, and 0.91 for adnexal masses using a bounding-box-guided U-Net. Evidence nevertheless remains constrained by small, single-centre, and retrospective datasets, limited external validation, device variability, and the scarcity of Indian data. The paper presents a conceptual analysis pipeline, compares key studies, and outlines research directions toward robust, clinically integrated tools.

DOI: https://doi.org/10.5281/zenodo.23411233

Published by: vikaspatanker

Authors: Janani A, Nivetha D, Prabhavathi U, Sumathi S

Abstract: The rapid circulation of news through websites and social-media platforms makes it difficult for readers to distinguish trustworthy reporting from fabricated or misleading content. This paper presents an AI-based fake news detection system that applies Natural Language Processing (NLP) and machine learning to classify textual news content into predefined categories such as fake and real. The workflow includes data preparation, text cleaning, feature extraction using Term Frequency–Inverse Document Frequency (TF-IDF), model training, prediction, and presentation of the result through an application interface. Logistic Regression is considered as the baseline classifier because it is efficient for sparse text features and provides a practical reference for text-classification tasks. The proposed system is intended to support early screening of suspicious news, not to replace source verification or professional fact-checking. The paper describes the design, implementation workflow, evaluation metrics, risks, and future improvements. Since verified experimental logs and final test measurements were not supplied for this manuscript, numerical performance is not fabricated; the results section specifies the measurements that the team should insert after running the final model. The work provides a structured, reproducible approach for building and evaluating a news-text classification application.

DOI: https://doi.org/10.5281/zenodo.23394198

Published by: vikaspatanker

Authors: Madhusudan Rao Karri

Abstract: The rapid growth of healthcare information generated from electronic health records, clinical applications, laboratory systems, medical devices, insurance platforms, and external healthcare services has created significant challenges in managing heterogeneous and distributed data environments. Differences in data structures, formats, semantics, metadata definitions, and governance policies can reduce data interoperability, quality, consistency, and analytical usability. This research proposes a Unified Metadata-Driven Framework for Healthcare Data Modeling, Integration, and Governance to provide a systematic approach for managing enterprise healthcare data throughout its lifecycle. The proposed framework uses metadata as a central mechanism for describing data structures, business definitions, relationships, lineage, quality rules, integration mappings, and governance policies. It integrates metadata-driven data modeling with standardized data integration processes to support consistent transformation and exchange of healthcare information across heterogeneous systems. The framework also incorporates governance capabilities for data quality, security, privacy, access control, lineage, ownership, and regulatory compliance. A layered architecture comprising metadata management, data modeling, integration, governance, quality management, and monitoring components is proposed to improve scalability and operational consistency. The framework can further support automated schema mapping, metadata validation, lineage tracking, and policy enforcement, thereby reducing manual intervention and improving data management efficiency. Evaluation of the framework can be performed using measures such as integration efficiency, data quality, metadata completeness, transformation accuracy, processing time, lineage coverage, and governance compliance. The proposed approach provides a unified foundation for building scalable, interoperable, and governed healthcare data environments capable of supporting analytics, reporting, clinical decision-making, and enterprise data management.

DOI: https://doi.org/10.5281/zenodo.23391272

Published by: vikaspatanker

Authors: Dhakchitha C, Kaviyashree S, Rithika K, Subashini R

Abstract: Maintaining integrity in invigilated examinations is labor-intensive, and human proctors cannot continuously observe every candidate. This paper presents Exam Malpractice Detector, a lightweight, real-time monitoring system built with OpenCV and Python that flags suspicious behavior from a single camera feed. The system combines four inexpensive visual cues: multiple-face detection using a Haar cascade, frame-differencing-based motion analysis, face-position-based "looking away" estimation, and Canny-edge contour analysis for unexpected objects. Detections are mapped to a three-tier status (Normal, Doubt/Review, Confirmed Malpractice) with confidence values. Each alert triggers cooldown-limited, hash-deduplicated, JPEG-compressed evidence capture. Alerts are also projected onto a 5×7 seat grid, where weighted scores produce a color-coded hall heatmap. At session end, a structured PDF report containing session metadata, violation summary, heatmap, event table, and screenshots is generated automatically for staff review. The design keeps a human in the loop: the system flags and documents, and staff decide. Because it needs no GPU and no training data, it can run on commodity hardware.

DOI: https://doi.org/10.5281/zenodo.23389223

Published by: vikaspatanker

Authors: Ahmed M. Sakr, Ibrahim Abd Ellatif, Ahmed S. Hashad, Mohamed A. Khalaf, Ismail Amer

Abstract: This study experimentally investigates the cyclic dynamic performance of Basalt Textile Reinforced Mortar (BTRM)-strengthened and unstrengthened masonry hydraulic piers under dry and wet conditions. Six specimens, comprising three control and three BTRM-strengthened piers, were tested using a shaking-table setup under progressively increasing harmonic excitation. The primary objective was to compare the number of loading cycles sustained until failure and examine the associated failure mechanisms. The results demonstrated that BTRM strengthening substantially increased the cyclic endurance of the masonry piers compared with the unstrengthened controls. The control specimens experienced severe damage after relatively few loading cycles, whereas the strengthened specimens sustained considerably more cycles before failure. The observed failure of the strengthened specimens involved basalt textile rupture and internal separation of masonry units. The dry and wet specimens were also compared to examine the influence of water exposure on cyclic performance. The findings demonstrate the potential of BTRM strengthening to improve the resistance of masonry hydraulic piers to repeated dynamic loading

DOI: https://doi.org/10.5281/zenodo.23364924

Published by: vikaspatanker

Authors: Ms. Shashiprabha V, Assistant Professor Mr. Joel Joseph

Abstract: The present study focuses on the forensic evaluation of ink removal using vinegar and ethanol on contract bond paper under varying temperature conditions. The study aims to investigate the effects of these chemical agents on ink visibility, paper characteristics, and the detectability of potential alterations in written documents. Vinegar, containing acetic acid, and ethanol are examined as treatment agents to determine their effects on ink markings and the physical integrity of contract bond paper at different temperatures. The evaluation considers observable changes such as ink fading, residual ink traces, paper discolouration, surface texture, and fibre damage. Untreated paper samples serve as controls for comparison. Forensic examination methods, including visual inspection and microscopic analysis, may be used to identify characteristics associated with chemical treatment. The study also considers temperature as a factor that may influence the interaction between the treatment agents, ink, and paper substrate. The findings are expected to contribute to the understanding of document alteration and the limitations of detecting chemically treated ink entries in questioned documents. This research may provide useful insights for forensic document examiners in assessing the authenticity and integrity of contract documents. The conclusions will be based on the experimental results obtained.

DOI:

Published by: vikaspatanker