Longitudinal Risk Prediction of Hospital Readmission in Diabetes Management Using a Temporal Attention-Gated BiLSTM with Focal Loss M.Mallikarjuna Rao, C. Manjunath, D. Venkatesh, S. Vasundhara Proceedings of the 5th International Conference on Sentiment Analysis and Deep Learning Icsadl 2026, 2026 Thirty-day hospital readmission continues to be a critical quality-of-care indicator in chronic diseases, particularly Diabetes Mellitus. Readmissions significantly increase healthcare expenditure and patient morbidity. Although Electronic Health Records (EHRs) contain rich longitudinal patient information, many existing prediction models rely on snapshot-based features extracted from a single hospital encounter, thereby failing to capture disease progression over time. To address this limitation, this paper proposes a Temporal Attention-Gated Bidirectional Long Short-Term Memory (TA-BiLSTM) model that leverages sequential patient visit histories and incorporates a focal loss objective to handle severe class imbalance inherent in medical datasets. Experiments conducted on the UCI Diabetes 130-US Hospitals dataset (over <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{1 0 0, 0 0 0}$</tex> encounters) demonstrate that the proposed model achieves an AUROC of 0.696, outperforming Logistic Regression (0.582) and standard LSTM (0.621) models with statistical significance <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(\mathbf{p}<0.05)$</tex>. Furthermore, the integrated temporal attention mechanism enhances clinical interpretability by identifying influential visits contributing to readmission risk. The findings indicate that longitudinal, explainable deep learning models can support accurate and clinically meaningful readmission risk stratification.
Digital innovation in higher educational system: Application of information and communication technology Dilip I. Sangotra, S. Vasundhara, Rippandeep Kaur, Akanksha Srivastava, M. Sambathkumar AI Powered Educational Games and Simulations, 2025 Information and communication technology (ICT) connected to digital global education is crucial for sustainable education management and has a significant impact on all aspects of human life. Digital transformation has improved students' experiences and pedagogy changed institutional models and applied digital abilities, maturity and initiatives. This leads to progressive and sustainable institutional approaches in higher education. In common sense, it is positively associated with digital contests, maturity, and digital initiatives taken in higher educational institutions, which is also a discovery that promotes students adopting digital transformation. The path model promotes the adoption of the digital efforts of educational institutions, using maturity and initiative by ICT architecture, virtual labs, participant involvement, digital culture, and strategic foundation.
Pentapartitioned Neutrosophic Pythagorean Connectedness R. Radha, Annapoorna M S, S. Vasundhara, Parul Arora, Ankita Tiwari, et al. Neutrosophic Sets and Systems, 2025 The aim of this paper is to introduce the concept of Pentapartitioned Neutrosophic Pythagorean [PNP] connectedness is introduced and its properties are also studied. Also we investigate some interrelations between these types of Pentapartitioned Neutrosophic Pythagorean connectedness. We show that the continuous image of Pentapartitioned Neutrosophic Pythagorean connected space is PNP connected.
A study on the association between select cryptocurrencies and the Indian FOREX values Hema Neelam, Lingam Sampath, Phanidra Kumar Katkam, S. Vasundhara Digital Transformation and Sustainability of Business, 2025 This article examines the association between four cryptocurrencies such as Bitcoin, Ethereum, Ripple and Cardano out of top ten cryptocurrencies in the India according to the Forbes report Jan 2024 and Indian FOREX value. Such as US ($), GBP (i), Euro (€) and Japanese Yen (V) by using daily data from 01-01-2019 to 31-12-2023 i.e., 5 years. To assess the association between select cryptocurrencies and the Indian FOREX value, statistical techniques like correlation, variance inflation factor, Durbin-Watson stat, and multiple regression analysis are used. Except ADA with Japanese Yen (V) remaining cryptocurrencies such as BTC, ETH and XRP had positive correlation with foreign exchange rates. The authors found that there is a significant association between the cryptocurrencies and Indian foreign currencies exchange values of US ($), GBP (i), EURO (€) and Japanese Yen (V). It is also concluded from the study that, the Indian FOREX market has negative impact on cryptocurrency market.
Heart Disease Prediction through Hybrid LSTM and HREF Models Sireesha Moturi, Srilakshmi Mutyam, Durga Vyshnavi Grandhisila, Rabia Basri Shaik Kagaji, Geeta Padole, S.Vasundhara 2025 IEEE 6th Global Conference for Advancement in Technology Gcat 2025, 2025 Cardiovascular disease (CVD) continues to be a leading global cause of mortality, indicating a need for new methods to maintain a high precision for CVD diagnosis and early CVD detection. We introduce here a novel hybridized ensemble model by improving a a Bidirectional Long Short-Term Memory (BiLSTM) based neural network through a Hybrid Refined Ensemble Framework (HREF) for improving classification. The hybrid model was applied to the non-standardized Cleveland dataset, which contains clinical-related cardiology parameters relevant to cardiovascular health. The dataset was thoroughly pre-processed prior to training, involving outlier removal, missing value transformation, feature normalization, and subsequent utilization of SMOTEENN for the purpose of class balancing. The BiLSTM part is designed, to some extent, to capture intricate relationships behind the features, and the ensemble part through HREF boosts the prediction reliability through the combined output of numerous ensembles. From experiments carried out on the hybrid model, accuracy was 94.7 with ROC-AUC 0.9474 and good precision and recall scores. The findings of this research indicate that deep learning combined with ensemble refinement is a reliable and effective approach to support the early detection of heart disease in real clinical settings.
IoT-based Water Quality Monitoring using ESP32 Sampathirao Govinda Rao, Durga Sri Harshitha B, Tirukutchu Modhita, V. Vijaya Rama Raju, Srungaram Vasundhara, P. Vijaya Lakshmi Proceedings of 8th International Conference on Computing Methodologies and Communication Iccmc 2025, 2025 Water quality monitoring is crucial for public health, agriculture, and industry, as it helps detect contaminants like heavy metals and harmful chemicals that cause waterborne diseases. Safe water ensures a healthy lifestyle, efficient agriculture, and smooth industrial operations. Early detection of pollutants reduces their impact, making monitoring essential. With advancements in sensor technologies and IoT, accuracy and efficiency have significantly improved. This research utilizes an ESP32 microcontroller with sensors to measure temperature, turbidity, and total dissolved solids (TDS) in water samples. The low-cost, energy-efficient ESP32 enhances accessibility for smart water management applications. By integrating ESP32 with IoT, real-time monitoring becomes more efficient, ensuring safer water resources. This study builds a hand-held Internet of Things (IoT) node that tracks three key indicators in real time: temperature, turbidity, and total dissolved solids. An ESP32 polls sensors every ninety seconds, calibrates the readings, and posts them through Wi-Fi to a PHP server. A web dashboard and Android view display live charts and year-long trends. Field trials on tap, pond, and saline water resulted Mean Absolute Errors (MAE) of 0.7°C, 0.5 NTU, and 17 ppm when compared with commercial meters, and furthermore, delay stays below three seconds. The board draws 0.4 W, so a 2,000 mAh power bank runs it for two days. These results show that a low-cost ESP32 system can deliver continuous, accurate water-quality data to households, farms, and small plants without skilled staff or bulky kits.
A Genetic Algorithm Based Synthetic Data Augmentation Framework for Enhanced Breast Cancer Classification S Vasundhara IEEE Conference, 1-6 , 2026 2026
Digital Innovation in Higher Educational System: Application of Information and Communication Technology DI Sangotra, S Vasundhara, R Kaur, A Srivastava, M Sambathkumar AI-Powered Educational Games and Simulations, 367-390 , 2026 2026 Citations: 1
AI’s Impact on Marketing Success S Lingam Available at SSRN 5879922 , 2025 2025
Supporting of Crop Yield Prediction Using Machine Learning Algorithm Techniques HN S. Vasundhara,Madhavi lata Mangipudi,Supriya Vaddi Proceedings of the 7th International Conference on Communications and Cyber … , 2025 2025
Proceedings of the 7th International Conference on Communications and Cyber Physical Engineering: ICCCE 2024, 28–29 Febuary, Hyderabad, India A Kumar, S Mozar Springer Nature , 2025 2025 Citations: 5
Heart Disease Prediction through Hybrid LSTM and HREF Models S Moturi, S Mutyam, DV Grandhisila, RBS Kagaji, G Padole, ... 2025 IEEE 6th Global Conference for Advancement in Technology (GCAT), 1-6 , 2025 2025
BSE SENSEX Price Prediction Using the ARIMA Model—A Study of Econometrics H Neelam, N Soumya Sony, S Vasundhara, C Bhargavi The Digital Edge: Transforming Business Systems for Strategic Success … , 2025 2025
An Efficient Method for Heart and Lung Disease Classification Using Siasme-BiLSTM Model P Saravanabhava, N Aparna, JP Patra, N Nishant 2025 3rd International Conference on Data Science and Network Security … , 2025 2025
A Novel Approach to IoT Based Plant Health Monitoring System Based on MLR-RBF Approach AK Koshariya, V Tomar, UR Patil, S Sakthivel, N Nishant 2025 3rd International Conference on Data Science and Network Security … , 2025 2025
IoT-based Water Quality Monitoring using ESP32 SG Rao, T Modhita, VVR Raju, S Vasundhara, PV Lakshmi 2025 8th International Conference on Computing Methodologies and … , 2025 2025
Genetic Algorithm-Optimized Machine Learning Models for Soil Health Assessment in Dryland Wheat Cultivation S Vasundhara, A Sridhar, B Ramesh, S Deshpande, YM Manu 2025 8th International Conference on Computing Methodologies and … , 2025 2025 Citations: 1
A study on the association between select cryptocurrencies and the Indian FOREX values H Neelam, L Sampath, PK Katkam, S Vasundhara Digital Transformation and Sustainability of Business, 770-774 , 2025 2025
Enhancing Distance Education Assessment using Open-Ended Questions and Innovative Measurement Techniques with SVM-WOA MN Geetha, A Thangam, S Ariawan, V Prabakaran, BS Kumar, N Nishant 2025 3rd International Conference on Data Science and Information System … , 2025 2025
The development of a framework for a comprehensive approach to stress management interventions at work based on BiLSTM-RNN model SM Imdadullah, A Pathak, K Soujanya, B Ramakrishna, A Chauhan, ... Hybrid and Advanced Technologies, 346-351 , 2025 2025
Enhancing entrepreneurial skills: A BERT-CNN-LSTM approach to student practical training G Pavani, A Dimari, N Tyagi, P Sampath, N Nishant, D Pulugu Hybrid and Advanced Technologies, 131-136 , 2025 2025
Optimising enterprise human resource management using attention-BiGRU-CapsNet model J Santhosh, D Indoria, Z Pasha, P Gottumukkala, K Soujanya, N Nishant Hybrid and Advanced Technologies, 322-327 , 2025 2025
A novel approach to constructing features and models for intrusion detection systems using SAE-ELM model DP Singh, VW Gangane, P Sukumar, J Kaushal, N Nishant, H Patil Hybrid and Advanced Technologies, 401-406 , 2025 2025 Citations: 1
Pentapartitioned neutrosophic pythagorean connectedness R Radha, S Vasundhara, P Arora, A Tiwari, R Prema Neutrosophic Sets and Systems 83 (1), 54 , 2025 2025 Citations: 2
Monitoring in-Vehicle Air Quality in Real Time Through IoT and Enhanced Deep Learning Models KP Kumar, J Kumar, N Malik, A Jadhav, P Sukumar, N Nishant 2024 First International Conference on Software, Systems and Information … , 2024 2024
Developing an Advanced IoT-Enabled Smart Agriculture Management System to Enhance Crop Growth: A Hybrid A-Residual-UNet Based Approach S Vasundhara IEEE, 6 , 2024 2024
MOST CITED SCHOLAR PUBLICATIONS
Elliptic curve cryptosystems SVS Vasundhara, DKV Durgaprasad Indian J. Appl. Res. 4 (3), 308-311 , 2011 2011.0 Citations: 20
Detection and Classification of Skin Cancer by Using a Parallel Deconvolutional Network Model M Sangeetha, C Karthikeyini, S Vasundhara, D Saravanan, G Arunkumar 2022 International Interdisciplinary Humanitarian Conference for … , 2022 2022.0 Citations: 16
The advantages of elliptic curve cryptography for security S Vasundhara Global Journal of Pure and Applied Mathematics 13 (9), 4995-5011 , 2017 2017.0 Citations: 14
AI-powered marketing revolutionizing customer engagement through innovative strategies S Vasundhara, KS Venkatesh, V Manimegalai, P Sundharesalingam, ... Cases on AI Ethics in Business, 21-46 , 2024 2024.0 Citations: 12
Elliptic curve cryptography and Diffie-hellman key exchange S Vasundhara IOSR Journal of Mathematics 13 (1), 56-61 , 2017 2017.0 Citations: 7
Leveraging drone and GPS technologies for precision agriculture: pest management perspective TK Patel, S Vasundhara, S Rajesha, B Priyalakshmi, S Chinnusamy, ... Revolutionizing pest management for sustainable agriculture, 285-308 , 2024 2024.0 Citations: 6
Data visualization view with Tableau S Vasundhara Stoch Model Appl 25, 178-87 , 2021 2021.0 Citations: 6
Importance of Gender Sensitization S Vasundhara International journal of multidisciplinary and current Educational research … , 2020 2020.0 Citations: 6
Proceedings of the 7th International Conference on Communications and Cyber Physical Engineering: ICCCE 2024, 28–29 Febuary, Hyderabad, India A Kumar, S Mozar Springer Nature , 2025 2025.0 Citations: 5
Developing an advanced IoT-enabled smart agriculture management system to enhance crop growth: A hybrid a-residual-UNet based approach A Magadum, M Goud, S Vasundhara, N Mishra, R Lakshmi, R Mahima 2024 International Conference on Intelligent Algorithms for Computational … , 2024 2024.0 Citations: 3
Machine learning with Monarch butterfly optimization for prediction of emergency patient admission status S Vijayarangam, S Vasundhara, NR Beherac, S Das, S Chandre, ... 2023 Fifth International Conference on Electrical, Computer and … , 2023 2023.0 Citations: 3
Pentapartitioned neutrosophic pythagorean connectedness R Radha, S Vasundhara, P Arora, A Tiwari, R Prema Neutrosophic Sets and Systems 83 (1), 54 , 2025 2025.0 Citations: 2
Machine Learning Algorithms–The Effect of Training and Testing Process S Vasundhara, S Tejaswini, AS Sriyam International Conference on Data Science, Machine Learning and Applications … , 2023 2023.0 Citations: 2
Detection and Classification of Skin Cancer by Using a Parallel Deconvolutional Network Model C Karthikeyini, S Vasundhara, D Saravanan, G Arunkumar 2022 International Interdisciplinary Humanitarian Conference for … , 2022 2022.0 Citations: 2
Elliptic Curves and Cryptography S Vasundhara International Journal of Information Research and Review, January , 2017 2017.0 Citations: 2
Digital Innovation in Higher Educational System: Application of Information and Communication Technology DI Sangotra, S Vasundhara, R Kaur, A Srivastava, M Sambathkumar AI-Powered Educational Games and Simulations, 367-390 , 2026 2026.0 Citations: 1
Genetic Algorithm-Optimized Machine Learning Models for Soil Health Assessment in Dryland Wheat Cultivation S Vasundhara, A Sridhar, B Ramesh, S Deshpande, YM Manu 2025 8th International Conference on Computing Methodologies and … , 2025 2025.0 Citations: 1
A novel approach to constructing features and models for intrusion detection systems using SAE-ELM model DP Singh, VW Gangane, P Sukumar, J Kaushal, N Nishant, H Patil Hybrid and Advanced Technologies, 401-406 , 2025 2025.0 Citations: 1
Impact of Polycystic Ovary Syndrome-Approach with Machine Learning Algorithms S Vasundhara Citations: 1
A Genetic Algorithm Based Synthetic Data Augmentation Framework for Enhanced Breast Cancer Classification S Vasundhara IEEE Conference, 1-6 , 2026 2026.0
