Graph Based Routing Optimisation in 5G/6G Wireless Mesh Networks with Mixed Mobility Nodes Aarthi D, Kodeeswari. K, Harishchander Anandaram, M.Sri Soundharyaa, S. Sowmiya, F.V. Jayasudha Proceedings of 5th International Conference on Communication Computing and Electronics Systems Iccces 2026, 2026 The heterogeneity of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{5 G} / \mathbf{6 G}$</tex> wireless mesh networks, including UAVs, vehicles, and edge devices, leads to frequent topology changes and intermittent links, both challenging the conventional routing protocols. This work proposes the GROF, Graph-based Routing Optimization Framework, a time-varying graph approach, fusing lightweight graph neural networks (GNN) with reinforcement routing for latencyminimizing, high-reliability path prediction in mixed-mobility meshes. GROF builds dynamic graphs from local link sensing, computes node embeddings by localized message passing, and selects the next hop using a distributed RL policy. Simulation results with 50-200 nodes, static to mobile ratio of 70:30, running on realistic mobility and traffic mixes, show improvement of PDR by <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\sim 5-12 \%$</tex> and reduction of end-to-end delay by <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$20-40 \%$</tex>, as compared to AODV, OLSR, HWMP, and a DRL baseline, while reducing routing overhead and energy per delivered packet. GROF's distributed graph learning is designed for deployment at the edge and promises good scalability to denser topologies.
Gradient Aware Attention Networks for Interpretable Chest X Ray Pathology Identification K.Kodeeswari, Aarthi D, Bhupesh Goyal, K.Arulmozhiarasu, Hirald Dwaraka Praveena, Harishchander Anandaram Proceedings of 6th International Conference on Expert Clouds and Applications Icoeca 2026, 2026 Chest X ray imaging is extensively utilized for the diagnosis of pulmonary diseases, but the majority of deep learning based diagnosis systems are black boxes, making them less trustworthy for clinical applications. To overcome the limitation that deep learning based models face, in this work, a novel Gradient Aware Attention Network (GAAN), able to provide interpretability for the identification of chest X ray pathologies, is proposed. Unlike other possibility based explanation systems, the proposed approach inherently includes the gradient values within the attention learning approach. Based on the approach, the model is able to map the discriminative regions with attention. In addition, the model is optimized through joint learning. Comprehensive experiments conducted on the NIH ChestX ray14, CheXpert and MIMIC CXR databases show that the proposed model performs better compared to the state of the art CNN models, attention models and explanation models. In addition, the approach exhibits accuracy of 88.1%, AUC value of 0.921 and F1 value of 0.867. Moreover, the approach improves the localization map significantly with IoU value 0.517. Based on the experimental validation, the approach can produce attention maps that possess clear, pristine and relevant attributes with values focused on the regions such as the consolidation region and pleura region.
Hybrid deep learning model for autism spectrum disorder diagnosis D. Aarthi, S. Kannimuthu Scientific Reports, 2025 Autism spectrum disorder (ASD) is a neurodevelopmental condition pertaining to the communication, social connectivity and conduct of individuals. ASD individuals develop symptoms such as recurrent actions, atypical facial expressions and challenges in social engagement. ASD prediction depends on various measures such as functional Magnetic Resonance Imaging (fMRI) data, game-based assessments, kinematic traits, questionnaires, head activity analysis, motor activities and eye-tracking. Traditional diagnostic approaches are subjective. These approaches are clinician-dependent and time-consuming. This has resulted in various challenges for the early detection of the condition. This work evaluated the performance of five hybrid approaches such as MobileNetV2+BiLSTM, ResNet50+LSTM, EfficientNetB4, InceptionV3 and MobileNetV2+GRU. Each model was meticulously refined to achieve optimal performance on the facial image dataset obtained from the Kaggle repository. The hybrid MobileNetV2+GRU model showed high performance with 95.5% test accuracy, 95.94% precision, and 95.45% F1-score. When the suggested hybrid model was compared with the remaining models, the latter outperformed with a ROC value of 98%. The findings highlight the optimal performance and generalizability of the proposed MobileNetV2+GRU model in ASD diagnosis in children.
Real-time electronic circuit control via brain-computer interface with machine learning K. P. Manikandan, D. Aarthi, Sanjit Das, R. Manikandan, Ravi Kumar Saidala, G. Manikandan Concepts and Applications of Brain Computer Interfaces, 2025 The purpose of this investigation is to investigate the possibility of achieving with machine instruction algorithms. By utilizing neural impulses, brain-computer interface (BCI) technologies make it possible for the human brain to communicate directly with the outside world. Developing a strong framework that can analyze brain exertion patterns through the application of modern machine literacy methods is the primary emphasis of this project. One of the most important goals is to optimize signal processing techniques to improve the signal-to-noise ratio. Another important objective is to enforce adaptive algorithms to provide effective real-time control. By providing implicit operations in assistive technology and neuroprosthetics, the approach that has been developed intends to facilitate smooth and seamless commerce between electronic systems and pharmaceutical corporations. The experimental confirmation process entails putting the system through a series of tests to determine its dependability, sensitivity, and responsiveness across a variety of circuit configurations.
AI-Powered Traffic Violation Detection System using Machine Learning Aarthi D, B. Keerthika, R. Kanchana, Jose P, Saurabh Chandra, S. P. Santhoshkumar Proceedings of the 6th International Conference on Smart Electronics and Communication Icosec 2025, 2025 This paper explores the development and implementation of an AI-powered violation detection system leveraging machine learning techniques. The system aims to automate the identification of rule infractions across various domains, such as traffic management, security surveillance, and regulatory compliance. By training machine learning models on labeled datasets of violation and non-violation instances, the system can learn to recognize complex patterns and anomalies indicative of prohibited activities. The investigation focuses on the application of machine learning architectures specifically convolutional neural networks (CNNs) for processing visual data or transformers for handling sequential data. Bookmarked messages copy message performance evaluation demonstrates the system’s ability to achieve high accuracy and efficiency in detecting violations, thereby reducing the reliance on manual monitoring and enhancing overall enforcement capabilities.
Early Detection of Robust Fetal Health Prediction Leveraging Using Machine Learning S. P. Santhoshkumar, Aarthi D, D. Jayasutha, Sunil Kumar Sahu, S.Vidhya, Kanakaprabha. S 2025 Global Conference on Information Technology and Communication Networks Gitcon 2025, 2025 Early identification of the better maternal and newborn outcomes, this paper examines how glowing unusual machine learning models identify fetal health issues using cardiotocography (CTG) data. The XGBoost model proved to be the most dependable model for classifying fetal health following extensive training and testing, by an accurateness of 96.01% and a Phi Coefficient of 0.8899. Despite somewhat lagging behind XGBoost, ensemble approaches like stacking and blending also produced high accuracies of 95.54% with MCC values of 0.8768. The Random Forest model was situated as a good but less dependable extra with an impressive accuracy of 94.60% and MCC of 0.8474. The model employed Logistic Regression worst by an accuracy of 87.79% and an MCC of 0.6655, while K-Nearest Neighbors and Support Vector Machine models presented practical efficacy. Due to its wonderful accuracy and balanced performance, the XGBoost model is optional for clinical applications in fetal health classification based on these findings. The model’s possible to enable early diagnosis and decrease the maternal and newborn mortality to timely intervention.
Machine Learning-Based Land Encroachment Detection Using Satellite Imagery S. Malathi, Aarthi D, B. R. Supreeth, K. Bhargavi, Muthumeena S, Mohit Tiwari Proceedings 2025 International Conference on Recent Innovation in Science Engineering and Technology Icriset 2025, 2025 The combination of urban development with low regulatory oversight and forest clearing produces land encroachment as a major issue. Interruption detection techniques based on manual surveys and basic remote sensing face accuracy and cost limitations as well as delayed detection capabilities. A land encroachment detection method that uses CNNs and multi-temporal satellite imagery to address the existing problems associated with land encroachment detection. The classification process uses spatial-temporal analysis and spectral indices including NDVI, NDBI, and NDWI, enhancing the detection accuracy. The developed system delivered 94.7% accuracy which exceeded both RF at 86.4% and SVM at 88.2% while attaining precision of 93.5% along with recall of 95.2%. The results show that the detection system achieves dependable results on various soil and land types. Real-time monitoring of the system relies on Google Earth Engine and GIS visualization tools that have been deployed for implementation. The proposed approach delivers land management solutions along with automated prevention systems for encroachment that help both urban planners and policymakers to work more effectively.
Energy-Grid-Optimizer Model Utilizing EfficientNet-Inception Framework for Smart Energy Systems and Grid Optimization Aarthi D, P V Narendra Kumar, Pallavikumari Keshavbhai Gamit, Santosh Shivlal Dhamone, P.Nagasekhara Reddy, Nageswara Rao Lavuri Proceedings 1st International Conference on Frontier Technologies and Solutions Icfts 2025, 2025 Indeed, deeply advanced learning for data scientists has been applied to previously energy optimized smart grids through Deep learning and Deep Reinforcement learning techniques. In particular, emerging and gaining ground for smart grids is the use of deep neural network based reinforcement learning approaches. For discrete and continuous action space we use the deep reinforcement learning algorithms. We were able to construct these algorithms into a robust real physical smart grid optimizer with the use of the MATLAB environment. It was observed that the agent could correctly identify the features in the training data related to energy supply and demand and ‘learn’ to make actions that will get it the most rewards. In the scope of our multilevel, systems based method of smart grid modelling, EGO-ENIF model and traditional methods are deployed in tandem for the sake of improvement at different levels. Besides improving the predictive accuracy of the energy consumption patterns, this approach enables energy distribution decision making to be made in real time. With the help of advanced machine learning concept and the established model, we can achieve a more stable and efficient smart grid based on resource optimization and conditions adjustment.
MACAFNet transformer-based multi-atlas fusion framework for autism spectrum disorder classification using functional connectivity D Aarthi, S Kannimuthu Scientific Reports , 2026 2026
Gradient Aware Attention Networks for Interpretable Chest X Ray Pathology Identification K Kodeeswari, D Aarthi, B Goyal, K Arulmozhiarasu, HD Praveena, ... 2026 6th International Conference on Expert Clouds and Applications (ICOECA … , 2026 2026
Graph Based Routing Optimisation in 5G/6G Wireless Mesh Networks with Mixed Mobility Nodes D Aarthi, H Anandaram, MS Soundharyaa, S Sowmiya, FV Jayasudha 2026 5th International Conference on Communication, Computing and … , 2026 2026
Classification and Detection of Brain Tumor Using Deep Learning Techniques D Aarthi, SS Deepika, MJ Devi, P Sowmya, S Subathra International Conference on Edge Computing and Applications, 167-177 , 2025 2025
Machine Learning-Based Land Encroachment Detection Using Satellite Imagery S Malathi, D Aarthi, BR Supreeth, K Bhargavi, S Muthumeena, M Tiwari 2025 International Conference on Recent Innovation in Science Engineering … , 2025 2025
A Recent Research and Evolving Patterns in Computer Vision Employing Deep Learning M Mutharasu, D Aarthi, P Ramya, G Manikandan, R Rawat, N Kalaivani 2025 6th International Conference for Emerging Technology (INCET), 1-6 , 2025 2025 Citations: 1
Vehiguard Transfer Learning-Driven Quantum PSO-LSTM for Automotive Cyber Security D Aarthi, D Gokila, P Loganathan, R Mutha, SV Saveeithaa 2025 International Conference on Data Science, Agents & Artificial … , 2025 2025 Citations: 1
Energy-Grid-Optimizer Model Utilizing EfficientNet-Inception Framework for Smart Energy Systems and Grid Optimization D Aarthi, PVN Kumar, PK Gamit, SS Dhamone, PN Reddy, NR Lavuri 2025 International Conference on Frontier Technologies and Solutions (ICFTS … , 2025 2025
Real-Time Electronic Circuit Control via Brain-Computer Interface With Machine Learning KP Manikandan, D Aarthi, S Das, R Manikandan, RK Saidala, ... Concepts and Applications of Brain-Computer Interfaces, 181-196 , 2025 2025
Certain Investigations on Machine Learning Models for Material Processing D Aarthi, S Kannimuthu International Conference on Recent Advancements in Materials Science and … , 2024 2024
A Novel Machine Learning Model For Financial Volatility Detection In Corporate Companies D Aarthi, R Madala, K Susithra, S Biswal, SS Nanda, K Joshi 2023 6th International Conference on Contemporary Computing and Informatics … , 2023 2023 Citations: 5
A comprehensive analysis of autism spectrum disorder using machine learning algorithms: Survey D Aarthi, S Kannimuthu International Conference on Power Engineering and Intelligent Systems (PEIS … , 2023 2023 Citations: 7
Machine Learning and Deep Transfer Learning approaches were used to create a Face Mask Identification model for COVID-19 BS Rao, D Aarthi, S Verma, PV Nandankar, P Radhakrishnan 2022 International Conference on Innovative Computing, Intelligent … , 2022 2022 Citations: 7
AUTISM SPECTRUM DISORDER ANALYSIS USING ARTIFICIAL INTELLIGENCE: A SURVEY D Aarthi, M Udhayamoorthi, G Lavanya International Journal of Advanced Research in Engineering and Technology … , 2020 2020 Citations: 2
Utility Mining Algorithms: A Bird’s Eye View D Aarthi, S Lavanya, S Kannimuthu, K Subhashree International Conference on Computer Networks and Inventive Communication … , 2019 2019 Citations: 1
Analysis of human behavior using gaming effects and social factors in video games KB Subhashree, S Lavanya, KS Bhuvaneshwari, D Aarthi Int J Innov Technol Expl Eng 8, 619-25 , 2019 2019 Citations: 3
MOST CITED SCHOLAR PUBLICATIONS
A comprehensive analysis of autism spectrum disorder using machine learning algorithms: Survey D Aarthi, S Kannimuthu International Conference on Power Engineering and Intelligent Systems (PEIS … , 2023 2023 Citations: 7
Machine Learning and Deep Transfer Learning approaches were used to create a Face Mask Identification model for COVID-19 BS Rao, D Aarthi, S Verma, PV Nandankar, P Radhakrishnan 2022 International Conference on Innovative Computing, Intelligent … , 2022 2022 Citations: 7
A Novel Machine Learning Model For Financial Volatility Detection In Corporate Companies D Aarthi, R Madala, K Susithra, S Biswal, SS Nanda, K Joshi 2023 6th International Conference on Contemporary Computing and Informatics … , 2023 2023 Citations: 5
Analysis of human behavior using gaming effects and social factors in video games KB Subhashree, S Lavanya, KS Bhuvaneshwari, D Aarthi Int J Innov Technol Expl Eng 8, 619-25 , 2019 2019 Citations: 3
AUTISM SPECTRUM DISORDER ANALYSIS USING ARTIFICIAL INTELLIGENCE: A SURVEY D Aarthi, M Udhayamoorthi, G Lavanya International Journal of Advanced Research in Engineering and Technology … , 2020 2020 Citations: 2
A Recent Research and Evolving Patterns in Computer Vision Employing Deep Learning M Mutharasu, D Aarthi, P Ramya, G Manikandan, R Rawat, N Kalaivani 2025 6th International Conference for Emerging Technology (INCET), 1-6 , 2025 2025 Citations: 1
Vehiguard Transfer Learning-Driven Quantum PSO-LSTM for Automotive Cyber Security D Aarthi, D Gokila, P Loganathan, R Mutha, SV Saveeithaa 2025 International Conference on Data Science, Agents & Artificial … , 2025 2025 Citations: 1
Utility Mining Algorithms: A Bird’s Eye View D Aarthi, S Lavanya, S Kannimuthu, K Subhashree International Conference on Computer Networks and Inventive Communication … , 2019 2019 Citations: 1
MACAFNet transformer-based multi-atlas fusion framework for autism spectrum disorder classification using functional connectivity D Aarthi, S Kannimuthu Scientific Reports , 2026 2026
Gradient Aware Attention Networks for Interpretable Chest X Ray Pathology Identification K Kodeeswari, D Aarthi, B Goyal, K Arulmozhiarasu, HD Praveena, ... 2026 6th International Conference on Expert Clouds and Applications (ICOECA … , 2026 2026
Graph Based Routing Optimisation in 5G/6G Wireless Mesh Networks with Mixed Mobility Nodes D Aarthi, H Anandaram, MS Soundharyaa, S Sowmiya, FV Jayasudha 2026 5th International Conference on Communication, Computing and … , 2026 2026
Classification and Detection of Brain Tumor Using Deep Learning Techniques D Aarthi, SS Deepika, MJ Devi, P Sowmya, S Subathra International Conference on Edge Computing and Applications, 167-177 , 2025 2025
Machine Learning-Based Land Encroachment Detection Using Satellite Imagery S Malathi, D Aarthi, BR Supreeth, K Bhargavi, S Muthumeena, M Tiwari 2025 International Conference on Recent Innovation in Science Engineering … , 2025 2025
Energy-Grid-Optimizer Model Utilizing EfficientNet-Inception Framework for Smart Energy Systems and Grid Optimization D Aarthi, PVN Kumar, PK Gamit, SS Dhamone, PN Reddy, NR Lavuri 2025 International Conference on Frontier Technologies and Solutions (ICFTS … , 2025 2025
Real-Time Electronic Circuit Control via Brain-Computer Interface With Machine Learning KP Manikandan, D Aarthi, S Das, R Manikandan, RK Saidala, ... Concepts and Applications of Brain-Computer Interfaces, 181-196 , 2025 2025
Certain Investigations on Machine Learning Models for Material Processing D Aarthi, S Kannimuthu International Conference on Recent Advancements in Materials Science and … , 2024 2024
