DEEPVEINDETECT: An AI-Powered Predictive Analytics Framework for Early Diagnosis and Progression Monitoring of Varicose Vein Disorders S Pavithra, Pavithran N, S. Jancy, A. Viji Amutha Mary, Mercy Paul Selvan Proceedings of the 2026 6th International Conference on Image Processing and Capsule Networks Icipcn 2026, 2026 One of the typical vascular diseases, which are represented by the swollen, twisted veins in the limbs that get huge and, therefore, resulted in the development of the inability to transport blood and the malfunction of the valves of the veins. In their absence, complications, such as occurrence of venous ulcers, thrombophlebitis, bleeding and chronic venous insufficiency, can develop to a significant impact on the life and health. DeepVeinDetect is an automated MATLAB system that is concerned with the early detection of varicose vein using the latest image processing and machine learning solutions. This system consists of four stages, namely, image capture, preprocessing, feature extraction, and classification. The patterns of the veins are taken using high-resolution near-infrared or visible light images which are then subjected to preprocessing algorithms such as noise filtering, conversion to grayscales, histogram equalization and contrast enhancement which enhance the vein visibility. Edge detection (Sobel, Canny), morphological operation, skeletonization and geometric analysis are used to extract features used to quantify the width, curvature, bifurcations and intensity variations of veins. Machine learning models are applied to perform classification i.e. SVM and KNN. A MATLAB GUI is an available gadget with real-time analysis, visualization, as well as reporting, which is an easy-to-use non-invasive and cost-effective and accurate early varicose vein screening tool.
PREDICTIVE ANALYTICS FOR PHOTO EXHIBITION PLANNING Shakti Prakash Jena, Dr. Mercy Paul Selvan, Manpreet Singh, Sweta Kumari Barnwal, Om Prakash, Pooja Abhijeet Alone Shodhkosh Journal of Visual and Performing Arts, 2025 Predictive analytics has become an innovative feature of cultural institutions and other creative industries that will allow them to plan and optimize the experience of the exhibition based on the data. This paper introduces a predictive modeling system specifically designed to be used in the planning of the photo exhibition, which incorporates previous attendance trends, demographic factors, is an audience response metric, social media interactions, and ticketing transactions into a single analytical chain. The suggested system conducts the severe data preprocessing, including noise reduction, outlier correction, and normalization as well as the multiple level feature engineering to create the robust predictors of visitor attendance, the engagement level, the distribution of the dwell time, and revenue streams. Several algorithmic paradigms are considered, such as the Linear Regression, the Random Forest, the XGBoost, the LSTM, and the ARIMA forecasting each optimized by systematic hyperparameter tuning strategy. Incorporating both the temporal and behavioral and environmental features, the framework improves the predictive accuracy and understandability of forecasts that are necessary in the exhibition schedules, content management and staff management needs, marketing segmentation, and spatial arrangement planning. Experimental evidence shows that tree-based ensemble algorithms, as well as hybrid architecture of deep learning, significantly outperform classical baselines, especially in the context of a nonlinear visitor behavior and dynamically changing preferences of the audience.
TransBreastNet a CNN transformer hybrid deep learning framework for breast cancer subtype classification and temporal lesion progression analysis Aluri Brahmareddy, Mercy Paul Selvan Scientific Reports, 2025 Breast cancer continues to be a global public health challenge. An early and precise diagnosis is crucial for improving prognosis and efficacy. While deep learning (DL) methods have shown promising advances in breast cancer classification from mammogram images, most existing DL models remain static, single-view image-based, and overlook the longitudinal progression of lesions and patient-specific clinical context. Moreover, the majority of models also limited their clinical usability by designing tests for subtype classification in isolation (i.e., not predicting disease stages simultaneously). This paper introduces BreastXploreAI, a simple yet powerful multimodal, multitask deep learning framework for breast cancer diagnosis to fill these gaps. TransBreastNet, a hybrid architecture that combines convolutional neural networks (CNNs) for spatial encoding of lesions, a Transformer-based modular approach for temporal encoding of lesions, and dense metadata encoders for fusion of patient-specific clinical information, forms the backbone of our system. The breast cancer subtype and disease stage are predicted simultaneously from a dual-head classifier. They are then used to construct temporal lesion sequences, either by employing genuine longitudinal data or by adding sequence augmentation to sample sequences, thereby strengthening the model's ability to learn Progression Patterns. We conduct extensive experiments on a public mammogram dataset and demonstrate that our model outperforms several state-of-the-art baselines in both subtype classification, achieving a macro accuracy of 95.2%, and stage Prediction, with a macro accuracy of 93.8%. We also provide ablation studies, which confirm how every module contributes to the framework. Unlike prior static single-view models, our framework jointly models spatial, temporal, and clinical features using a CNN-Transformer hybrid design. It simultaneously predicts breast cancer subtypes and lesion progression stages, while generating synthetic temporal lesion sequences where longitudinal data is scarce. Built-in explainability modules enhance interpretability and clinical trust. BreastXploreAI offers a robust, scalable, and clinically relevant approach to diagnosing breast cancer from full-field digital mammogram (FFDM) images. ZH is computationally capable of analyzing spatial, temporal, and clinical features simultaneously, which enables a more informed diagnosis and lays the foundation for improved clinical decision support systems in oncology.
ADVANCED CNN-BASED FRAMEWORKS FOR ROBUST AND EXPLAINABLE BREAST CANCER DIAGNOSIS ACROSS MULTI-MODAL IMAGING DATASETS Journal of Theoretical and Applied Information Technology, 2025
An Intelligent Chatbot for the Disabled People S. Jancy, A. Viji Amutha Mary, Mercy Paul Selvan, L. K. Joshila Grace Proceedings of the 7th International Conference on Intelligent Sustainable Systems Iciss 2025, 2025
Autonomous Vehicle: Line Following Robot 16th International Conference on Advances in Computing Control and Telecommunication Technologies Act 2025, 2025
Real Time Facial Mask Detection and Gender Identification 16th International Conference on Advances in Computing Control and Telecommunication Technologies Act 2025, 2025
AI based Proctoring System - A Review Joseph Sam Paul, Omar Farhath, Mercy Paul Selvan 7th International Conference on Inventive Computation Technologies Icict 2024, 2024
Prevention of Water Wastage in Household Pipelines Using IoT A.Viji Amutha Mary, Harisankar J D, J Ramasamy, Mercy Paul Selvan, S Jancy Proceedings of 9th International Conference on Science Technology Engineering and Mathematics the Role of Emerging Technologies in Digital Transformation Iconstem 2024, 2024
Ensuring Safety Drive through a Smart Helmet design using IOT P. Asha, Akshara C V, Aakash S, L. Padma Suresh, D. Usha Nandini, Mercy Paul Selvan Proceedings of International Conference on Circuit Power and Computing Technologies Iccpct 2024, 2024
Enforcement of CNN Model in Drone Detection System A. Viji Amutha Mary, N Anusha, Mercy Paul Selvan, R. Rajalakshmi, S. Jancy, L K Joshila Grace IEEE International Conference on Recent Advances in Science and Engineering Technology Icraset 2024, 2024
A Review of Intelligent Traffic Management Systems R Ashwath Ramanathan, BalaMurugan R, R Sathya Bama Krishna, Mercy Paul Selvan 7th International Conference on Trends in Electronics and Informatics Icoei 2023 Proceedings, 2023
Arduino based Women Safety Tracker Device Barukam Vamshikrishna Yadav, A. Viji Amutha Mary, Mercy Paul Selvan, S. Jancy, L. Suji Helen 7th International Conference on Trends in Electronics and Informatics Icoei 2023 Proceedings, 2023
Healthcare Monitoring System with Fall Detection Mechanism D Bibiana Magdelene, S. Jancy, A. Viji Amutha Mary, L. Suji Helen, Mercy Paul Selvan Proceedings of 8th IEEE International Conference on Science Technology Engineering and Mathematics Iconstem 2023, 2023
Automated Interview through Online Video Interface Joshila Grace L. K, Asha P, Jany Shabu, J. Refonaa, Mercy Paul Selvan Proceedings of the International Conference on Circuit Power and Computing Technologies Iccpct 2023, 2023
Price Prediction and Crop Yield using Machine Learning Algorithm 14th International Conference on Advances in Computing Control and Telecommunication Technologies Act 2023, 2023
Hybrid Multiple Cryptography for Data Encryption M. Anuradha, Sathyapriya Loganathan, G. Suseela, Mercy Paul Selvan, M. Nalini, Chitra Devi D Proceedings of the 8th International Conference on Communication and Electronics Systems Icces 2023, 2023
Menstrual Cycle Tracking Using Deep Learning Shikha Suman, Somdatta Mukherjee, Mercy Paul Selvan, Viji Amtha Mary, S. Jancy, S. Prayla Shyry Proceedings 2023 3rd International Conference on Pervasive Computing and Social Networking Icpcsn 2023, 2023
Covid-19 Future Forecasting Using Supervised Machine Learning Models Bassetti Lakshmi Lavanya, Vemula Katyayani, S. Jancy, S.Pushpa Latha, Viji Amutha Mary.A, L. Sujihelen, Mercy Paul Selvan Proceedings of 8th IEEE International Conference on Science Technology Engineering and Mathematics Iconstem 2023, 2023
Human Emotion Recognition System using Residual Connections L. K Joshila Grace, Mercy Paul Selvan, P. Asha, S. L. Jany Shabu, J. Refonaa Proceedings of the 2022 3rd International Conference on Intelligent Computing Instrumentation and Control Technologies Computational Intelligence for Smart Systems Icicict 2022, 2022
Prediction of Multiple Diseases Using Machine Learning Techniques SaiRam Krishna Parimi, Snehith Reddy, S. Jancy, L. Sujihelen, Mercy Paul Selvan, Viji Amutha A. Mary 2022 International Conference on Communication Computing and Internet of Things Ic3iot 2022 Proceedings, 2022
Crop Recommendation and Yield Production using SVM Algorithm M. Sai Teja, T. Sai Preetham, L. Sujihelen, Christy, S. Jancy, Mercy Paul Selvan Proceedings 2022 6th International Conference on Intelligent Computing and Control Systems Iciccs 2022, 2022
A Prediction of Crop Yield using Machine Learning Algorithm L.Mary Gladence, K. Raghavendra Reddy, M. Prudhvidhar Reddy, Mercy Paul Selvan, Refonaa Proceedings of the 5th International Conference on Trends in Electronics and Informatics Icoei 2021, 2021
Public auditing for secure cloud storage using md5 algorithm Associate Professor, Dept of CSE, Sathyabama Institute of Science, Technology, Chennai, A. Viji Amutha Mary*, Mercy Paul Selvan, Assistant Professor, Dept of CSE, Sathyabama Institute of Science, Technology, Chennai, Christy, Professor, Dept of CSE, Sathyabama Institute of Science, Technology, Chennai International Journal of Recent Technology and Engineering, 2019
Identifying selfish misbehavior nodes in wireless networks with enhanced data transmission rate International Journal of Applied Engineering Research, 2015
Ranking websites by its own features International Journal of Applied Engineering Research, 2014
Enabling social network for chronical patient and prevent information stealing attack International Journal of Applied Engineering Research, 2014