SWATHI MIRTHIKA G.L

@srmist.edu.in

Assistant Professor, Engineering & Technology
SRM Institute of Science and Technology, Kattankulathur

SWATHI MIRTHIKA G.L


                        Download Compact PDF  
https://researchid.co/dr.swathimirthikagl
Research primarily focuses on medical recommender systems using Knowledge Graphs and Large Language Models (LLMs), along with related areas such as Graph Theory, Artificial Intelligence, and Deep Learning. She has presented her work in various national and international conferences, and has published research papers in reputed journals and book chapters. She also contributes to the scientific community by serving as a reviewer for multiple journals.

EDUCATION

Dr. Swathi Mirthika G. L is an Assistant Professor in the Department of Computing Technologies, School of Computing, Kattankulathur Campus, SRM Institute of Science and Technology. She completed her Ph.D. in 2025 and holds over 7 years of academic experience. She earned her Master’s degree in Information Technology from Vel Tech Multi Tech Dr. RR Dr. SR Engineering College (2013) and her B.Tech in Information Technology from Sasurie College of Engineering (2010).

RESEARCH, TEACHING, or OTHER INTERESTS

Artificial Intelligence, Drug Discovery, Computer Engineering, Health Information Management
11

Scopus Publications

13

Scholar Citations

2

Scholar h-index

Scopus Publications

  • Data-driven drug treatment: enhancing clinical decision-making with SalpPSO-optimized GraphSAGE
    Swathi Mirthika G.L., Sivakumar B, S. Hemalatha
    Computer Methods in Biomechanics and Biomedical Engineering, 2026
    Safe drug recommendation systems play a crucial role in minimizing adverse drug reactions and enhancing patient safety. In this research, we propose an innovative approach to develop a safety drug recommendation system by integrating the Salp Swarm Optimization-based Particle Swarm Optimization (SalpPSO) with the GraphSAGE algorithm. The goal is to optimize the hyper parameters of GraphSAGE, enabling more accurate drug-drug interaction prediction and personalized drug recommendations. The research begins with data collection from real-world datasets, including MIMIC-III, Drug Bank, and ICD-9 ontology. The databases provide comprehensive and diverse clinical data related to patients, diseases, and drugs, forming the foundation of a knowledge graph. It represents drug-related entities and their relationships, such as drugs, indications, adverse effects, and drug-drug interactions. The knowledge graph's integration of patient data, disease ontology, and drug information enhances the system's accuracy to predict drug-drug interactions as well as identifying potential detrimental drug reactions. The GraphSAGE algorithm is employed as the base model for learning node embeddings in the knowledge graph. To enhance its performance, we propose the SalpPSO algorithm for hyper parameter optimization. SalpPSO combines features from Salp Swarm Optimization and Particle Swarm Optimization, offering a robust and effective optimization process. The optimized hyper parameters lead to more reliable and accurate drug recommendation system. For evaluation, the dataset is split into training and validation sets and compared the performance of the modified GraphSAGE model with SalpPSO-optimized hyper parameters to the standard models. The experimental analysis conducted in terms of various measures proves the efficiency of the proposed safe recommendation system, offering valuable for healthcare experts in making more informed and personalized drug treatment decisions for patients.
  • Brain Tumor Progression Forecasting via Multi- Modal Deep Learning
    Soshya Joshi, G. L. Swathi Mirthika, T. Krithika, Sriraam Balaji
    Revolutionizing Drug Research and Personalized Medicine Through AI and Machine Learning, 2026
    Accurate forecasting of brain tumor progression is essential for effective diagnosis and treatment planning. This chapter proposes a multi-modal deep learning framework that integrates spatial and temporal analysis of MRI and PET/CT images to improve tumor progression prediction. The model combines ResNet-50 for extracting detailed spatial features with a Long Short-Term Memory network. By fusing anatomical and functional imaging information, the framework provides a comprehensive understanding of tumor behavior and growth patterns. The proposed approach was evaluated on both clinical and publicly available datasets comprising diverse tumor types. Experimental results demonstrate high predictive performance, achieving accuracies of up to 99.40% on clinical data and 98.31% on online datasets, outperforming conventional CNN-based models. This work highlights the potential of hybrid deep learning architectures as reliable decision-support tools for radiologists, enabling early detection, improved monitoring, and personalized treatment planning in neuro-oncology.
  • Benchmarking Traditional ML Approaches in Phishing URL Detection
    T. S. Sangeetha, Keerthi Jayan, Sreya John, D. Vetriselvi, G. L. Swathi Mirthika, Nisha Thorakkattu Thorakattil Madathil, K. S. Jishnu
    Navigating Public Security in the Age of Post Truth Challenges and Implications, 2026
    Phishing attacks continue to pose a major cybersecurity challenge by exploiting deceptive URLs to obtain sensitive information. Although deep learning approaches such as CNNs, RNNs, and Transformers have demonstrated state of the art detection performance, traditional machine learning classifiers remain widely utilized because of their efficiency,interpretability, and relatively low resource requirements. In this study, we implement and evaluate ten machine learning models including Logistic Regression, Gradient Boosting, CatBoost, XGBoost, and Multi-Layer Perceptron on a publicly available phishing URL dataset from Kaggle. Experimental results show that ensemble based models,particularly XGBoost and Random Forest, achieve the highest accuracy, while Logistic Regression offers competitive performance with the advantages of simplicity, interpretability, and low computational overhead. The findings highlight the tradeoffs between accuracy, interpretability, and computational cost,providing practical guidance for selecting appropriate models in real world phishing detection systems.
  • HC-DMAformer: hybrid convolutional-dynamic multi-level attention transformer for efficient and accurate EEG-based autism detection
    V. Kavitha, R. Vidhya, G. L. Swathi Mirthika, K. Suresh, S Hemavathi
    Evolving Systems, 2025
  • Bridging Data Complexity with GATNet for Learning in Interconnected Electronic Medical Records Graphs
    G.L. Swathi Mirthika, B Sivakumar
    Ingenierie Des Systemes D Information, 2024
    Heterogeneous graphs are a data format for graphs that could define complicated and diverse real-world interactions by accommodating distinct sorts of nodes and edge types.Heterogeneous graphs organize varied medical data to help patients, therapies, drugs, and healthcare practitioners make informed decisions.Medical recommendation systems use them to represent and analyze complicated connections between healthcare data items.Heterogeneous graphs can potentially be constructed and analyzed using the Graph Attention Network (GAT).The purpose of this research is to tackle the issue of implementing a complicated and extremely diverse dataset, which consists of: Using the GATNet (Graph Attention Network) method, we will show how to perform two things: (1) Construct a model with several attributes and relationships using EMR (electronic medical record), and (2) Use that model in a disease prognostic prediction challenge.The initial graph database utilizes a graphical depiction of a patient's progression, showcasing a query of a predictive network that produces analytical findings of AUROC-0.75 and AUPRC-0.17 which is 0.03% & 0.02% higher compared to the existing models.
  • Healthcare innovation: Embracing technology to promote long-term global wellness
    Swathi Mirthika G. L, B. Sivakumar
    Technologies for Sustainable Healthcare Development, 2024
    The practical use of technology in healthcare is a nascent discipline focused on promoting sustainability within the healthcare sector. Nevertheless, the research in this field has experienced significant and swift growth. Although this expansion has been beneficial for the discipline, it has also rendered it more challenging to comprehend its full scope. Thus, it has become impossible to answer questions about the biggest technological shifts in sustainable healthcare research, the most important revolution articles, their influence, and the most prolific and prominent scholars. Understanding the logical framework of technology for sustainable long-term healthcare information is difficult. An in-depth assessment of studies on revolutionary healthcare technology that promotes long-term health addressed these issues. It did so to address some of the issues. Over the past four years, the COVID-19 pandemic, artificial intelligence, machine learning, and the internet of things have propelled sustainable healthcare technology research.
  • Graph Representation Learning for Predicting Diverse Sources of Drug Interactions
    International Journal of Intelligent Engineering and Systems, 2024
    Drug treatment strategies to reduce dose-related hazards is a tried-and-true method for preventing drug resistance and enhancing the efficiency of the monotherapy.Except when certain drugs pile up.Most adverse medication effects are induced by antagonistic drug-drug interactions.New medications and monitoring patients' use of more effective medication combination therapies require precise Drug-Drug Interaction (DDI) prediction.Several machine learning-based DDI prediction methods exist.This wide range of strategies uses drug-related and substancerelated traits covertly.Graph embeddings and deep learning are applied to benchmark datasets to overcome this.The Simplified Molecular Input Line Entry System (SMILE) method is introduced for preprocessing, and the GCNet is applied for DDI prediction.Moreover, the graph is also constructed based on that the similarity is identified using link prediction.The proposed method provides an accuracy range of 0.934, Mean Squared Error (MSE) of 0.082, and Root Mean Squared Error (RMSE) of 0.352, which assists in more effectively reducing adverse drug reactions.
  • Enhancing Hybrid Filtering for Evolving Recommender Systems: Dealing with Data Sparsity and Cold Start Problems
    Swathi Mirthika G. L., B. Sivakumar
    Proceedings 2024 2nd International Conference on Inventive Computing and Informatics Icici 2024, 2024
    The recommendation system has become common due to the widespread use of the Internet and the availability of a wide range of products. It serves as a useful tool in assisting decision-making on online purchases. Challenges in recommender system such as difficulties in starting a system from scratch, unavailability of important data, excessive focus on specific areas, lack of up-to-date information, scarcity of data, and incorrect metadata. The traditional recommendation system relies on the collaborative filtering algorithm. The most effective suggestion techniques are Collaborative Filtering and Content Based Filtering. Despite their popularity, these filtering systems still suffer from several limitations, including the Cold Start Problem, Sparsity, and Scalability, all of which result in subpar suggestions and to provide a recommendation-anticipation system based on a hybrid approach in this article. The suggested solution combines content-based filtering with collaborative filtering.This research established a model for an online shopping recommendation system that takes into account both people and products by analyzing two types of conventional algorithms.
  • Integrating Ed Tech and cybersecurity into the curriculum
    Deepa V., Sivakumar B., Swathi Mirthika G. L., Abinaya K.
    Handbook of Research on Current Trends in Cybersecurity and Educational Technology, 2023
    The chapter aims to discuss how people interested in incorporating different aspects of cyber security into the curriculum in higher education, corporate learning, and non-traditional learning environments can do so. Due to the COVID-19 outbreak, more businesses are switching to online platforms, which has resulted in a dramatic spike in the need for cybersecurity controls. The objective is to share how the immense impact of educational technology (ED Tech) in the age of COVID-19 lockdown is digitally focused on educational content and the basic need to be flexible and adaptable for students. Overall, this chapter proposes that the need for cyber security is becoming a more crucial facet of life. So, the report by intelligence system can examine threats and the platforms/applications for processing data can be examined. Cyber ED Tech protects against a variety of attacks, including cloud security breaches, denial of service (DoS), malware, zoombombing, password policy violations, and so on.
  • Recommendation System Based on Clustering Techniques Using Collaborative Filtering Method
    G. L. Swathi Mirthika, B. Sivakumar
    Lecture Notes in Networks and Systems, 2023
  • Drug Interaction Prediction using Various Methods to Reduce Adverse Effects
    Swathi Mirthika G. L, B. Sivakumar
    2022 6th International Conference on Trends in Electronics and Informatics Icoei 2022 Proceedings, 2022

RECENT SCHOLAR PUBLICATIONS

  • Benchmarking Traditional ML Approaches in Phishing URL Detection
    TS Sangeetha, K Jayan, S John, D Vetriselvi, GLS Mirthika, NTT Madathil, ...
    Navigating Public Security in the Age of Post-Truth: Challenges and … , 2026
    2026
  • Brain Tumor Progression Forecasting via Multi-Modal Deep Learning
    S Joshi, GLS Mirthika, T Krithika, S Balaji
    Revolutionizing Drug Research and Personalized Medicine Through AI and … , 2026
    2026
  • HC-DMAformer: hybrid convolutional-dynamic multi-level attention transformer for efficient and accurate EEG-based autism detection
    V Kavitha, R Vidhya, GL Swathi Mirthika, K Suresh, S Hemavathi
    Evolving Systems 16 (3), 99 , 2025
    2025
  • Data-driven drug treatment: enhancing clinical decision-making with SalpPSO-optimized GraphSAGE
    SM GL, Sivakumar B, Hemalatha
    Computer Methods in Biomechanics and Biomedical Engineering, 1-23 , 2024
    2024
    Citations: 5
  • Bridging Data Complexity with GATNet for Learning in Interconnected Electronic Medical Records Graphs
    SM GL, B Sivakumar
    Ingenierie des Systemes d'Information 29 (4), 1541 , 2024
    2024
  • Healthcare Innovation: Embracing Technology to Promote Long-Term Global Wellness
    BS Swathi Mirthika G. L
    Technologies for Sustainable Healthcare Development 1, 37 - 56 , 2024
    2024
  • Graph Representation Learning for Predicting Diverse Sources of Drug Interactions
    SMGL Sivakumar B
    The International Journal of Intelligent Engineering and Systems 17 (4), 666 … , 2024
    2024
  • Enhancing Hybrid Filtering for Evolving Recommender Systems: Dealing with Data Sparsity and Cold Start Problems
    SM GL, B Sivakumar
    2024 Second International Conference on Inventive Computing and Informatics … , 2024
    2024
    Citations: 2
  • Drug-Drug Interaction Prognosis: A Comprehensive Analysis of Methods to Minimize Adverse Drug Effects
    SMGL Sivakumar B
    Advances in Health and Disease 73, 187 - 207 , 2023
    2023
  • Recommendation System Based on Clustering Techniques Using
    GLS Mirthika, B Sivakumar
    Intelligent Sustainable Systems: Proceedings of ICISS 2023, 29 , 2023
    2023
  • Recommendation System Based on Clustering Techniques Using Collaborative Filtering Method
    GL Swathi Mirthika, B Sivakumar
    International Conference on Intelligent Sustainable Systems, 29-36 , 2023
    2023
    Citations: 1
  • A Systematic Review on Drug Interaction Prediction Using Various Methods to Reduce Adverse Effects
    GLS Mirthika, B Sivakumar
    International Journal of Information Systems and Social Change (IJISSC) 14 … , 2023
    2023
    Citations: 1
  • Integrating Ed Tech and Cybersecurity Into the Curriculum
    V Deepa, B Sivakumar, GL Swathi Mirthika, K Abinaya
    Handbook of Research on Current Trends in Cybersecurity and Educational … , 2023
    2023
  • Drug interaction prediction using various methods to reduce adverse effects
    SB Swathi Mirthika G.L
    2022 6th International Conference on Trends in Electronics and Informatics … , 2022
    2022
    Citations: 4
  • Supervised Approach to Extract Sentiments from Unstructured Text
    MGLS Mirthika, MS Yamini
    2013

MOST CITED SCHOLAR PUBLICATIONS

  • Data-driven drug treatment: enhancing clinical decision-making with SalpPSO-optimized GraphSAGE
    SM GL, Sivakumar B, Hemalatha
    Computer Methods in Biomechanics and Biomedical Engineering, 1-23 , 2024
    2024
    Citations: 5
  • Drug interaction prediction using various methods to reduce adverse effects
    SB Swathi Mirthika G.L
    2022 6th International Conference on Trends in Electronics and Informatics … , 2022
    2022
    Citations: 4
  • Enhancing Hybrid Filtering for Evolving Recommender Systems: Dealing with Data Sparsity and Cold Start Problems
    SM GL, B Sivakumar
    2024 Second International Conference on Inventive Computing and Informatics … , 2024
    2024
    Citations: 2
  • Recommendation System Based on Clustering Techniques Using Collaborative Filtering Method
    GL Swathi Mirthika, B Sivakumar
    International Conference on Intelligent Sustainable Systems, 29-36 , 2023
    2023
    Citations: 1
  • A Systematic Review on Drug Interaction Prediction Using Various Methods to Reduce Adverse Effects
    GLS Mirthika, B Sivakumar
    International Journal of Information Systems and Social Change (IJISSC) 14 … , 2023
    2023
    Citations: 1
  • Benchmarking Traditional ML Approaches in Phishing URL Detection
    TS Sangeetha, K Jayan, S John, D Vetriselvi, GLS Mirthika, NTT Madathil, ...
    Navigating Public Security in the Age of Post-Truth: Challenges and … , 2026
    2026
  • Brain Tumor Progression Forecasting via Multi-Modal Deep Learning
    S Joshi, GLS Mirthika, T Krithika, S Balaji
    Revolutionizing Drug Research and Personalized Medicine Through AI and … , 2026
    2026
  • HC-DMAformer: hybrid convolutional-dynamic multi-level attention transformer for efficient and accurate EEG-based autism detection
    V Kavitha, R Vidhya, GL Swathi Mirthika, K Suresh, S Hemavathi
    Evolving Systems 16 (3), 99 , 2025
    2025
  • Bridging Data Complexity with GATNet for Learning in Interconnected Electronic Medical Records Graphs
    SM GL, B Sivakumar
    Ingenierie des Systemes d'Information 29 (4), 1541 , 2024
    2024
  • Healthcare Innovation: Embracing Technology to Promote Long-Term Global Wellness
    BS Swathi Mirthika G. L
    Technologies for Sustainable Healthcare Development 1, 37 - 56 , 2024
    2024
  • Graph Representation Learning for Predicting Diverse Sources of Drug Interactions
    SMGL Sivakumar B
    The International Journal of Intelligent Engineering and Systems 17 (4), 666 … , 2024
    2024
  • Drug-Drug Interaction Prognosis: A Comprehensive Analysis of Methods to Minimize Adverse Drug Effects
    SMGL Sivakumar B
    Advances in Health and Disease 73, 187 - 207 , 2023
    2023
  • Recommendation System Based on Clustering Techniques Using
    GLS Mirthika, B Sivakumar
    Intelligent Sustainable Systems: Proceedings of ICISS 2023, 29 , 2023
    2023
  • Integrating Ed Tech and Cybersecurity Into the Curriculum
    V Deepa, B Sivakumar, GL Swathi Mirthika, K Abinaya
    Handbook of Research on Current Trends in Cybersecurity and Educational … , 2023
    2023
  • Supervised Approach to Extract Sentiments from Unstructured Text
    MGLS Mirthika, MS Yamini
    2013