AI for Hospital Administration, Staff Scheduling, and Operational Efficiency: Transforming Healthcare Operations Through Intelligent Automation K. Saravanan, M. Sadhasivam, G. Sumathy, A. Maheshwari, M. G. Dinesh Breakthroughs in Smart Nursing with Generative AI, 2026 Artificial Intelligence (AI) is reshaping hospital administration, workforce management, and operational efficiency by enabling intelligent automation, predictive insights, and data-driven decision-making. This chapter explores the integration of AI technologies including machine learning, deep learning, natural language processing, and reinforcement learning within key hospital operational domains such as administrative workflows, staff scheduling, resource allocation, and performance optimization. It highlights how AI enhances patient flow, reduces delays, improves asset management, and supports real-time operational intelligence. The chapter also discusses challenges related to data privacy, bias, system interoperability, governance, and workforce acceptance. Finally, emerging trends and future research directions, including digital twins, federated learning, and explainable AI, are explored to guide the development of resilient and efficient healthcare operations.
Virtual Electrode - Driven Graph and Contrastive Learning Framework for EEG Based Stress Detection Deeparani S, Sumathy G Proceedings 2025 International Conference on Recent Innovation in Science Engineering and Technology Icriset 2025, 2025 Optimal electrode selection is a critical challenge in EEG-based mental stress detection. This work proposes a novel framework combining virtual electrode generation, Graph Convolutional Networks (GCNs), contrastive learning, and channel-wise attention to enhance spatial resolution, robustness, and adaptability across subjects. Virtual electrodes improve spatial detail through interpolation, while GCNs capture spatial correlations between real and synthetic electrodes. Contrastive learning enhances the discriminability of the features by making similar or closely related stress-level embeddings close and the dissimilar or far apart and data augmentation enhances the generalizing ability of the model when subjected to different conditions. The channel attention mechanism is used to dynamically emphasis or de-emphasize the electrode importance in order to minimize the noise effect. Through the thorough testing in DEAP and SWELL-KW datasets, the accuracy is excellent (95.21 percent), the generalization is good since the testing is done on subject-independent mode, and the system is robust to noisy and skewed data. Our method has been proposed as ready to be deployed in real-time, which is an excellent solution to robust and interpretable EEG-based stress detection.
Optimizing CNN-BPNN Architectures for High-Accuracy Diagnosis of Multi-Class Thyroid Disorders Sumathy G, Sreelekha Nedunuri, Subashree Guruchandar Proceedings 2025 International Conference on Recent Innovation in Science Engineering and Technology Icriset 2025, 2025 Thyroid disease is prevalent throughout the world, and accurate diagnosis is crucial for treatment. In this study, we discuss a new approach to study thyroid diseases utilizing Convolutional Neural Networks (CNNs) for image classification and Backpropagation Neural Networks (BPNNs) for classifying datasets. Our dataset of thyroid images includes images collected via ultrasound, MRI, and CT scans. Each image is associated with the relevant thyroid disease category including hyperthyroidism, hypothyroidism, thyroid nodules, etc. In addition, the data was comprised of non-image features that included patient demographics, symptoms, and medical history. For classifying the images, we reviewed an architecture based on CNNs, which can automatically learn important attributes of the thyroid images. The CNN model has been trained using a labeled image dataset taking advantage of the convolutional and pooling layers for feature extraction, followed by fully connected layers for classification. A BPNN architecture has been created to classify the dataset and assess the subject in relation to non-image attributes associated with a thyroid disease. The BPNN is trained using back propagation, and by doing so, it learns the underlying patterns and relationships between a wide range of features, and the associated disease categories. The predictions have been drawn together and presented as a single system to assess the thyroid disease. The overall predictions were compiled using ensemble learning methods from the CNN and BPNN models, and the subsequent predictions improved the overall classification accuracy. The ability of the described system was assessed using non-overlapping test datasets, using measures of accuracy, precision, recall, f1-score and area under the ROC curve (AUC). The system was also validated against the domain experts, to ensure clinical usefulness and reliability. The outcome demonstrated the potential of integrating CNN's and BPNN's for a comprehensive evaluation of thyroid disease, providing a promising approach for accurate and reliable diagnosis in clinical practice.
Unveiling fetal heart health: harnessing auto-metric graph neural networks and Hazelnut tree search for ECG-based arrhythmia detection M. Suganthy, B. Sarala, G. Sumathy, W. T. Chembian Computer Methods in Biomechanics and Biomedical Engineering, 2025 Fetal electrocardiogram (ECG) provides a non-invasive means to assess fetal heart health, but isolating the fetal signal from the dominant maternal ECG remains challenging. This study introduces the FHH-AMGNN-HTSOA-ECG-AD method for enhanced fetal arrhythmia detection. It employs Dual Tree Complex Wavelet Transform for denoising and utilizes an Auto-Metric Graph Neural Network (AMGNN) optimized by the Hazelnut Tree Search Algorithm (HTSOA). This integration enables accurate classification of normal and abnormal fetal heart signals. Experimental results demonstrate that the proposed approach significantly outperforms existing methods in terms of accuracy, precision, and specificity.
Enhanced Waste Segregation Using Vision Transformers and YOLO Sumathy G, Maria Emilia Camargo, Walter Priesnitz Filho, Mithileysh Sathiyanarayanan Proceedings 1st International Conference on Frontier Technologies and Solutions Icfts 2025, 2025 Effective waste management is important due to rapid urbanization, meaning cities generate more waste. Managing it properly is important for the environment. The main challenge in the waste management process is separating waste into biodegradable (organic) and non biodegradable (inorganic) categories. Traditional methods of waste segregation are time-consuming, so there is a need for advanced technology. Our proposed state-of-the-art method for real-time object detection, YOLO, simplified the detection process, reduced the computational resources needed and made it faster and more efficient. To make Vision Transformers (ViTs) perform better, we used methods that enhance how they extract features and fit different datasets during training. That includes TACO and TrashNet. We perform hyperparameter tuning to optimize detection accuracy. The process yields better results for materials like clear glass (90%) and PET plastic (85%), while e-waste detection is only challenging for 45% of materials. The proposed method is revolutionizing waste segregation, minimizing labor and contributing to smart city initiatives. Future work will concentrate on enhancing classification accuracy for visually similar materials and e-waste materials.
Machine Learning for Software Development: Real-Time Communication System for Predicting Climate Condition Anurag Vijay Agrawal, G. Sumathy, A. Maheshwari, K. Saravanan, K. Revathi, Sampath Boopathi AI Frameworks and Tools for Software Development, 2025 This chapter elaborates on how machine learning is changing climatic condition prediction and analysis. Conventional techniques for climatic modeling simply cannot handle the extraordinary complexity and non-linearity inherent in climate systems quite often. As such, with advanced machine learning techniques, such as deep learning, reinforcement learning, and ensemble methods, masked patterns can be discovered, the accuracy of predictions enhanced, and the uncertainties associated with climate data can be handled. Applications such as temperature forecasting, extreme weather prediction, and long-term climate trend analysis are discussed. It also discusses the integration of satellite data, IoT-enabled sensors, and high-performance computing to enhance real-time monitoring and forecasting capabilities. This chapter explores the potential of machine learning in enhancing climate science by enabling proactive decision-making, addressing data scarcity, interpretability, and ethical considerations.
Comparative Analysis on Deep Learning Models for Cryptocurrency Prediction Tanishtha Gulati, Hashwanth Y, C. Sherin Shibi, Babitha Lincy R, Jency Rubia J, G. Sumathy Proceedings of 6th International Conference on Intelligent Communication Technologies and Virtual Mobile Networks Icicv 2025, 2025 Cryptocurrency price prediction carries important social and economic implications as it directly influences financial markets and personal investments. Statistically, it is essential but difficult to make accurate predictions because of high volatility and sophisticated market behavior. Precision and responsiveness are difficult to achieve, and this has been demonstrated by conventional predictive models like ARIMA and basic recurrent neural networks. It is difficult for traditional machine learning models to identify sophisticated market patterns. This paper helps to overcome these challenges by investigating advanced deep learning methods, namely Temporal Fusion Transformer (TFT), Graph Neural Networks and ARIMA Hybrid Models, integrated with specialized data transformation methods. These modern approaches help in improving the price prediction accuracy, which is higher than that of conventional models, with Graph Neural Networks (GNN) demonstrating accuracy over 90% for all leading cryptocurrencies. Initial results show that in the process of forecasting price direction for cryptocurrencies such as Bitcoin, Ethereum, Litecoin, and Binance Coin, GNN has consistently outperformed traditional methods while TFT has also performed well as a runner up model.
SmartCart Engine: Prompt-Powered Text Analysis and Product Validation Yash Goel, Vibhu Jain, Sumathy G Proceedings of 9th International Conference on Science Technology Engineering and Mathematics the Role of Emerging Technologies in Digital Transformation Iconstem 2024, 2024
Construction of Delaunay Triangles for Face Recognition in Images 14th International Conference on Advances in Computing Control and Telecommunication Technologies Act 2023, 2023
Eye point center localization through distance vector fields and improved grid based measure of eye field rotation Journal of Advanced Research in Dynamical and Control Systems, 2019
AI for Hospital Administration, Staff Scheduling, and Operational Efficiency: Transforming Healthcare Operations Through Intelligent Automation K Saravanan, M Sadhasivam, G Sumathy, A Maheshwari, MG Dinesh Breakthroughs in Smart Nursing With Generative AI, 213-240 , 2026 2026
Optimizing CNN-BPNN Architectures for High-Accuracy Diagnosis of Multi-Class Thyroid Disorders G Sumathy, S Nedunuri, S Guruchandar 2025 International Conference on Recent Innovation in Science Engineering … , 2025 2025
Virtual Electrode-Driven Graph and Contrastive Learning Framework for EEG Based Stress Detection S Deeparani, G Sumathy 2025 International Conference on Recent Innovation in Science Engineering … , 2025 2025
Unveiling fetal heart health: harnessing auto-metric graph neural networks and Hazelnut tree search for ECG-based arrhythmia detection M Suganthy, B Sarala, G Sumathy, WT Chembian Computer Methods in Biomechanics and Biomedical Engineering 28 (10), 1671-1684 , 2025 2025
Comparative Analysis on Deep Learning Models for Cryptocurrency Prediction T Gulati, Y Hashwanth, CS Shibi, G Sumathy 2025 6th International Conference on Intelligent Communication Technologies … , 2025 2025
Illegal Boat Detection Using Satellite Imagery with Deep Learning S Anvi, P Aarav, G Sumathy 2024 International Conference on Innovative Computing, Intelligent … , 2025 2025
Gaussian weighting—based random walk segmentation and DCNN method for brain tumor detection and classification KV Rani, G Sumathy, LK Shoba, P Sivalakshmi Multimedia Tools and Applications 84 (8), 4675-4702 , 2025 2025 Citations: 3
Interactive Virtual Reality Skill Enhancer for Girls with Autism Spectrum Disorder and Intellectual Disabilities: A Mixed Methods Study G Sumathy, A Singh, A Prasad Journal of Science 15, 100122 , 2025 2025
Machine Learning for Software Development: Real-Time Communication System for Predicting Climate Condition AV Agrawal, G Sumathy, A Maheshwari, K Saravanan, K Revathi, ... AI Frameworks and Tools for Software Development, 287-306 , 2025 2025
Self-powered triboelectric sensors for biomedical applications RB Lincy, JJ Rubia, CS Shibi, N Kanimozhi, CS Sheeba, G Sumathy Self-Powered Sensors, 139-157 , 2025 2025 Citations: 2
StudentsConnect: A Web Application to Track Library Permissions and Leave Requests in Hostels Using MERN G Sumathy, A Maheshwari, A AR, C Sherin Shibi, N Kanimozhi 2024 International Conference on Innovative Computing, Intelligent … , 2024 2024
SIRT: A distinctive and smart invasion recognition tool (SIRT) for defending IoT integrated ICS from cyber-attacks MS Kavitha, G Sumathy, B Sarala, JJ Hephzipah, R Dhanalakshmi, ... International Journal of Critical Infrastructure Protection 47, 100720 , 2024 2024 Citations: 4
Real-time masked face recognition using deep learning-based double generator network G Sumathy, M Usha, S Rajakumar, P Jayapriya Signal, Image and Video Processing 18 (Suppl 1), 325-334 , 2024 2024 Citations: 5
Intelligent Transportation Systems: Exploring Digital Twin Technologies in Smart Grid, Transportation Systems and Smart Cities N Kanimozhi, G Sumathy, A Maheshwari, AR Arunarani, C Sherin Shibi 2024 International Conference on Advances in Data Engineering and … , 2024 2024 Citations: 9
Elevating Financial Literacy through AI-Enhanced Real-Time Simulation based Learning G Sumathy, SS Reddy, P Muzamil 2024 Ninth International Conference on Science Technology Engineering and … , 2024 2024
SmartCart Engine: Prompt-Powered Text Analysis and Product Validation Y Goel, V Jain 2024 Ninth International Conference on Science Technology Engineering and … , 2024 2024
Efficient Net Based Brain Encephalopathy using Data Augmentation Techniques for Alzheimer and Peripheral Pathological Outcomes G Sumathy, BSVS Sri, G Chanakya 2024 Ninth International Conference on Science Technology Engineering and … , 2024 2024
A Study on AI and Blockchain-Powered Smart Parking Models for Urban Mobility K Sundaramoorthy, A Singh, G Sumathy, A Maheshwari, AR Arunarani, ... Handbook of Research on AI and ML for Intelligent Machines and Systems, 223-250 , 2024 2024 Citations: 97
Machine Learning in E-Health and Digital Healthcare: Practical Strategies for Transformation TK Sethuramalingam, RG Nadakinamani, G Sumathy, S Myilsamy Handbook of Research on AI and ML for Intelligent Machines and Systems, 276-304 , 2024 2024 Citations: 4
Hybrid K-Means Clustering for Grouping the Special Children Using Computational Techniques G Sumathy, A Maheshwari, AR Arunarani, C SherinShibi, N Kanimozhi, ... 2023 International Conference on Innovative Computing, Intelligent … , 2023 2023 Citations: 1
MOST CITED SCHOLAR PUBLICATIONS
A Study on AI and Blockchain-Powered Smart Parking Models for Urban Mobility K Sundaramoorthy, A Singh, G Sumathy, A Maheshwari, AR Arunarani, ... Handbook of Research on AI and ML for Intelligent Machines and Systems, 223-250 , 2024 2024 Citations: 97
Glioma brain tumor detection using dual convolutional neural networks and histogram density segmentation algorithm B Sarala, G Sumathy, AV Kalpana, JJ Hephzipah Biomedical Signal Processing and Control 85, 104859 , 2023 2023 Citations: 41
Radon transform-based improved single seeded region growing segmentation for lung cancer detection using AMPWSVM classification approach KV Rani, G Sumathy, LK Shoba, PJ Shermila, ME Prince Signal, Image and Video Processing 17 (8), 4571-4580 , 2023 2023 Citations: 22
Investigation of the Wear Behavior of AA6063/Zirconium Oxide Nanocomposites Using Hybrid Machine Learning Algorithms PMRB Gizachew Assefa Kerga R. Reena Roy,1 Leninisha Shanmugam,1 A. Vinothini ... Journal of Chemistry 2023 , 2023 2023 Citations: 13
An efficient prevention and challenges in wireless sensor networks for energy and security concern G Sumathy, K Geetha, A Maheshwari, AR Arunarani, P Samuel 2023 International Conference on Artificial Intelligence and Knowledge … , 2023 2023 Citations: 13
Traffic light pre-emption control system for emergency vehicles P Priya, A Jose, G Sumathy SSRG International Journal of Electronics and Communication Engineering … , 2015 2015 Citations: 12
A machine learning approach to segment the customers of online sales data for better and efficient marketing purposes T Mathesh, G Sumathy, A Maheshwari 2023 International Conference on Artificial Intelligence and Knowledge … , 2023 2023 Citations: 10
Intelligent Transportation Systems: Exploring Digital Twin Technologies in Smart Grid, Transportation Systems and Smart Cities N Kanimozhi, G Sumathy, A Maheshwari, AR Arunarani, C Sherin Shibi 2024 International Conference on Advances in Data Engineering and … , 2024 2024 Citations: 9
A survey of vision and speech stimulation for cerebral palsy rehabilitation G Sumathy, A Renjith 2014 International Conference on Control, Instrumentation, Communication and … , 2014 2014 Citations: 9
AI powered transformative post generator for LinkedIn using LLM and explicit filter V Jain, Y Goel, M Uma 2023 International Conference on Innovative Computing, Intelligent … , 2023 2023 Citations: 8
Distance-based method used to localize the eyeball effectively for cerebral palsy rehabilitation G Sumathy, J Arokia Renjit Journal of Medical Systems 43 (8), 262 , 2019 2019 Citations: 8
Radon transform-based improved single seeded region growing segmentation for lung cancer detection using AMPWSVM classification approach. SIViP 17: 4571–4580 KV Rani, G Sumathy, LK Shoba 2023 Citations: 7
Real-time masked face recognition using deep learning-based double generator network G Sumathy, M Usha, S Rajakumar, P Jayapriya Signal, Image and Video Processing 18 (Suppl 1), 325-334 , 2024 2024 Citations: 5
SIRT: A distinctive and smart invasion recognition tool (SIRT) for defending IoT integrated ICS from cyber-attacks MS Kavitha, G Sumathy, B Sarala, JJ Hephzipah, R Dhanalakshmi, ... International Journal of Critical Infrastructure Protection 47, 100720 , 2024 2024 Citations: 4
Machine Learning in E-Health and Digital Healthcare: Practical Strategies for Transformation TK Sethuramalingam, RG Nadakinamani, G Sumathy, S Myilsamy Handbook of Research on AI and ML for Intelligent Machines and Systems, 276-304 , 2024 2024 Citations: 4
Gaussian weighting—based random walk segmentation and DCNN method for brain tumor detection and classification KV Rani, G Sumathy, LK Shoba, P Sivalakshmi Multimedia Tools and Applications 84 (8), 4675-4702 , 2025 2025 Citations: 3
Self-powered triboelectric sensors for biomedical applications RB Lincy, JJ Rubia, CS Shibi, N Kanimozhi, CS Sheeba, G Sumathy Self-Powered Sensors, 139-157 , 2025 2025 Citations: 2
Hybrid K-Means Clustering for Grouping the Special Children Using Computational Techniques G Sumathy, A Maheshwari, AR Arunarani, C SherinShibi, N Kanimozhi, ... 2023 International Conference on Innovative Computing, Intelligent … , 2023 2023 Citations: 1
Enhancement of security in cloud computing using optimal risk access control model AR Arunarani, CS Shibi, N Kanimozhi, G Sumathy, A Maheshwari 2023 International Conference on Innovative Computing, Intelligent … , 2023 2023 Citations: 1
Segmentation of Human Eye Pupil with Novel Grid Based Localization Computing ARJ Sumathy. G Indian Journal of Public Health Research & Development 10 (7), 35-41 , 2019 2019 Citations: 1