S. Sujatha

S. Sujatha

Auto-generated profile, based on publicly available information

S. Sujatha

Professor, Department of EEE, Sri Sairam College of Engineering

Research Focus & Metrics

Academic Career Length

17Years (2010-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 April 2010 in Asian Journal of Agricultural Research
Verified
Journal Article
Closed access
SSS. Sujatha
S. Sujatha
BJBaby Joseph
Baby Joseph
PSP.S. Sumi
P.S. Sumi

1 January 2025 in Informing Science: The International Journal of an Emerging Transdiscipline
Verified
Journal Article
Open access
ASAnnapantula Sudhakar
Annapantula Sudhakar
SSSujatha S
Sujatha S
SMSathiya M.
Sathiya M.
SASIVARAMAKRISHNAN ARULSWAMY
SIVARAMAKRISHNAN ARULSWAMY
BSBalambigai Subramanian
Balambigai Subramanian
VKVenkata Ramana K
Venkata Ramana K
Aim/Purpose: Traditional hyperparameter tuning methods’ inefficiency and low accuracy in diabetes prediction. Application and evaluation of Bayesian optimization aim to increase the accuracy of diabetes prediction models. Background: Bayesian optimization (BO) has emerged as a powerful technique for hyperparameter tuning in machine learning models with its methodical approach for optimizing complex functions with expensive evaluations. Selecting the optimal hyperparameters for models can significantly increase prediction accuracy in the healthcare sector, particularly for diabetes prediction, hence improving patient outcomes and resource management. Methodology: Among the well-known diabetes databases used in the study is the Pima Indian Diabetes Database. Among the machine learning models developed and whose hyperparameters are modified via Bayesian optimization are Support Vector Machines (SVM), Random Forest, and Gradient Boosting Machines (GBM). The optimization process is compared with traditional tuning methods to assess improvements in model performance. Contribution: Problems include the increasing prevalence of diabetes worldwide and the role vital function prediction models play in early diagnosis and therapy. Especially when dealing with enormous datasets common in healthcare, grid search, and random search are sometimes computationally taxing and inefficient. However, Bayesian optimization promises a more practical and economical approach by selecting hyperparameters iteratively based on earlier evaluations. Findings: Bayesian optimization not only yields faster calculation times but also outperforms traditional methods. More precisely, models adapted by Bayesian optimization show greater sensitivity and specificity, which are crucial for accurate and prompt diabetes diagnosis. Recommendation for Researchers: By highlighting the specific elements and network characteristics that are behind the anomalies it discovers, on a real-time or large scale dataset. Future Research: This work can be improvised using several recent artificial intelligence algorithms with the integration of IoT on real-time datasets.

21 June 2024
Verified
Book Chapter
Closed access
DMD. Manimegalai
D. Manimegalai
RGR. Gunasekari
R. Gunasekari
SSS. Sujatha
S. Sujatha
KMKARTHIKEYAN Madhu
KARTHIKEYAN Madhu
AUA. Umasankar
A. Umasankar
This chapter explores the integration of artificial intelligence of things (AIoT) with tele-health monitoring, presenting an intelligent remote patient activity tracking system. Using interconnected sensors and machine learning models, the system monitors various patient activities and vital signs, providing real-time insights. The proposed AIoT-enabled health monitoring device tracks activities such as running, sleeping, walking, and exercising, while observing vital signs like heart rate and breathing patterns. The chapter delves into the system's architecture, functionalities, and technical intricacies, emphasizing its potential to transform healthcare delivery. It highlights the role of AIoT-driven solutions in enhancing patient care and fostering personalized treatment approaches, offering a glimpse into the promising future of intelligent healthcare delivery.

1 January 2025 in Informing Science: The International Journal of an Emerging Transdiscipline
Verified
Journal Article
Open access
PDF file
SCSushama C
Sushama C
SRShaik Mohammad Rafee
Shaik Mohammad Rafee
JJJaimala Jha
Jaimala Jha
SSSujatha S
Sujatha S
JSJagadeesan S
Jagadeesan S
MKMohana Krishna
Mohana Krishna
Aim/Purpose: This work attempts to investigate the application of Long Short Term Memory (LSTM)-based predictive analytics in the medical field. The scope comprises the development and application of LSTM models to forecast outcomes, including patient diagnosis, treatment responses, healthcare resource consumption, and other relevant variables. Background: Predictive analytics has become popular in many other industries, including healthcare, since it can analyze enormous amounts of data and project future patterns. For LSTM, a variant of encoder-decoder LSTM-based recurrent neural network (RNN), time-series prediction activities have demonstrated superior performance. Accurate projections in the medical industry can lead to better patient outcomes, optimal use of resources, and cost cuts. Methodology: The method calls for compiling and preparing medical data from numerous sources. Using prior data, LSTM models will learn temporal patterns and relationships between target outcomes and input variables. Several techniques, including feature engineering, hyperparameter tuning, and model evaluation, will be employed to maximize LSTM-based predictive analytics. Contribution: This work supports the growth of predictive analytics in the healthcare industry by demonstrating the accuracy of LSTM models in forecasting significant clinical outcomes. This research could change healthcare decision-making processes, improving general operational effectiveness, patient treatment, and resource use. Findings: Among other healthcare outputs, the results suggest that LSTM-based predictive analytics can faithfully project patient diagnosis, illness progression, and therapy responses. The models outperform traditional forecasting methods over multiple datasets in terms of reliability and accuracy. Recommendation for Researchers: Future studies can focus on bettering model designs, leveraging various data modalities, and including predictive analytics in clinical decision support systems. Cooperation among data scientists, medical professionals, and legislators will help to realize the possibilities of LSTM-based predictive analytics in healthcare. Future Research: This work can be enhanced in future research using several deep learning algorithms with a real-time industrial dataset.

1 March 2026 in Journal of Renewable and Sustainable Energy
Verified
Journal Article
Closed access
SSS. Siva Subramanian
S. Siva Subramanian
SSS. Sujatha
S. Sujatha
VTV. Tamilselvan
V. Tamilselvan
BAB. Ashok
B. Ashok
The energy industry is increasingly emphasizing renewable energy sources for the environmental benefits, energy security, and cost-effectiveness. Microgrids (MG) are important to combine these sources and guarantee the quality of power systems, their reliability, and efficiency. Fuel cells (FCs), especially developed in MG, are simple, efficient, and modular, but have issues such as high temperature. In order to address this, researchers propose to combine energy storage technologies such as batteries and supercapacitors (SCs). The integration of batteries and SC enhances system efficiency, mitigates FC limitations, and optimizes performance in hybrid energy systems, fostering sustainability and resilience in energy landscapes. This research presents a cutting-edge approach to MG energy management and power-quality enhancement, combining SC, batteries, and FCs. The proposed approach involves the application of differential evolution mutation, leopard seal optimization, and multi-strategy enhanced location to do the robust parameter tuning, which significantly enhances the performance of the system. This fractional controller, applied to the inverter, ensures efficient operation with minimized stress in the hybrid energy storage system, improved battery state of charge, reduced harmonics, as well as enhanced overall system efficacy. The proposed technique is evaluated using both experimental and MATLAB platforms and compared with other approaches. The proposed approach achieves a maximum voltage regulation of 11%, while the external energy maximization system achieves 3.7%. Also, fractional controller with advanced optimization, achieving superior performance in DC-link voltage (±0.15 V), SC and battery current regulation (±0.1 and ±0.05 A), and minimal frequency deviation (–1.4 Hz).

28 June 2024
Verified
Book Chapter
Closed access
SMSheerin Banu M.
Sheerin Banu M.
DSD Sugumar
D Sugumar
SSSujatha S.
Sujatha S.
VTVeeramakali T.
Veeramakali T.
Patient-centered AI solutions for dealing with Alzheimer's ailment is a research-based totally article that explores the ability of synthetic intelligence (AI) in aiding the management and remedy of Alzheimer's disease (AD). Alzheimer's disorder is a revolutionary neurodegenerative sickness that impacts thousands and thousands of people worldwide. And not using a recognized treatment, managing advert calls for a comprehensive and personalized technique. This text highlights the key demanding situations in cutting-edge advert control and introduces the function of AI in addressing those demanding situations. It additionally discusses the potential benefits of patient-targeted AI answers in improving early detection, analysis, and remedy of advert. Through an assessment of existing studies, this article provides promising AI applications which include advanced imaging, biomarker evaluation, and predictive modeling for identifying individuals prone to ad and monitoring disorder development.