M SUCHITHRA

@srmist.edu.in

Associate Professor , Department of Computing Technologies , School of Computing
SRM Institute of Science and Technology , Kattankulathur

M SUCHITHRA


                  Download Compact PDF  
https://researchid.co/suchithram

EDUCATION

BTech, ME, PhD

RESEARCH INTERESTS

Computer Science Engineering - AI, IoT, Vision
27

Scopus Publications

Scopus Publications

  • Personalized Drug Recommendation Using ClinicalBERT-Augmented Graph Neural Networks
    Anushka Pandey, Suyash Kumar, M. Suchithra
    Lecture Notes in Networks and Systems, 2026
  • An Implementation of Adaptive Multi-CNN Feature Fusion Model With Attention Mechanism With Improved Heuristic Algorithm for Kidney Stone Detection
    Gunasekaran Kulandaivelu, M Suchitra, R Pugalenthi, Ruchika Lalit
    Computational Intelligence, 2025
    Nowadays, most people have been admitted to emergencies with severe pain caused by kidney stones worldwide. In this case, diverse imaging approaches are aided in the detection process of stones in the kidney. Moreover, the specialist acquires better diagnosis and interpretation of this image. Here, computer‐aided techniques are considered the practical techniques, which it is used as the auxiliary tool for the process of diagnosis. Most urologists have failed to train the type of kidney stone identification effectively and it is operator‐dependent. Concerning the surgical operation, there is a requirement for accurate as well as adequate detection of stone position in the kidney. Thus, it has made the detection process even more difficult. To overcome the challenging issues, an effective detection model for kidney stones using classifiers. Initially, the input images are collected from the standard resources. Further, the input images are subjected to the adaptive multi‐convolutional neural network with attention mechanism (AMC‐AM) feature fusion model, in which, the pertinent features are extracted from the three networks: Visual Geometry Group16 (VGG16), Residual Network (ResNet), and Inception net. Thus, the three distinct features are obtained for the feature fusion procedure. Finally, the resultant features are subjected as input to the final layer of CNN. In the proposed network, the model is integrated with the attention mechanism and also the parameter tuning is done by proposing the modified social distance of coronavirus mask protection algorithm (MSD‐CMPA). Therefore, the performance is examined using different metrics and compared with other baseline models. Hence, the proposed model overwhelms the outstanding results in detecting the kidney stones that aid the individual in getting rid of kidney disorders.
  • A Comprehensive review of Brain Disorder Prediction: Current Techniques and Insights with Future Research Directions
    Oormila L, M. Suchithra
    6th International Conference on Innovative Trends in Information Technology Secure Trustworthy and Socially Responsible AI Icitiit 2025, 2025
    The brain is the body's central nervous system. Recently, more deep learning techniques are developed for the prediction of brain disorders. Current technologies for diagnosis or detection of brain disorders are getting more difficult due to the variety of brain illnesses. Numerous investigations effectively employed medical imaging methods to identify, diagnose, and treat human brain problems in their early stages. An essential component of medical image analysis is machine learning. Nonetheless, human specialists continue to do the feature selection stage based on their domain-specific expertise. This makes it difficult for non-experts to apply machine learning methods to medical image analysis. To understand the significance of brain disorder prediction, the past literature works are surveyed. Further, the datasets used in the brain disorder prediction process are analyzed and also the experimental tools involved in the conventional brain disorder prediction models are listed for obtaining detailed insights in this domain. Moreover, the techniques and algorithms used for predicting the brain disorder are classified. Additionally, the performance metrics employed for validating the brain disorder prediction techniques are provided for observing the efficiency of these models. Finally, the research gaps and the future directions of the brain disorder prediction models are explained to improve the advanced prediction techniques.
  • An Advanced EEG based Human Emotion Detection with Giant Armadillo Optimization Algorithm
    G. Kalyana Chakravarthy, M Suchithra
    7th International Conference on Energy Power and Environment Icepe 2025, 2025
    Emotion detection is the technique of determining the emotions for identifying the human mood or feelings. .In this Manuscript, Augmented Physics-Informed Neural Networks Optimized with Giant Armadillo Optimization Algorithm for Human Emotion Detection (APINN-GOA-HED) is proposed. First, data regarding human emotions is retrieved from the EEG dataset. Data-Adaptive Gaussian Average Filtering is used as a pre-processing technique to remove noise from collected data.Statistical characteristics such as RMS and variance are extracted using the High-Order synchro extracting transform (HSET) after pre-processed data has been exposed to feature extraction. The Augmented Physics-Informed Neural Network (APINN) is then fed the extracted features in order to efficiently categorize human emotions including happy, relaxed, disgusted, sad, and neutral. In general, the classifier of Augmented Physics-Informed Neural Networks does not exhibit the capability to modify optimization strategies in order to identify the ideal parameters for a system that accurately detects human emotions. Giant Armadillo Optimization is therefore suggested as a way to improve Augmented Physics-Informed Neural Networks, which correctly classify human emotions.The ROC analysis of the proposed APINN-GOA-HED approach is 0.87%, 0.88%, and 0.89% higher than that of the current methods, respectively.
  • Fingerspelling for Indian Sign Language using Swin Transformer
    M Suchithra, Ayushi Gupta, Abhilasha Kasaraneni
    Proceedings of 6th International Conference on Intelligent Communication Technologies and Virtual Mobile Networks Icicv 2025, 2025
    In this study, the Swin Transformer is used to predict Indian Sign Language (ISL) alphabets (A-Z) and is incorporated into an interactive Streamlit-based interface. The study intends to improve the accuracy and usability of ISL recognition systems. After being refined on a unique ISL dataset, the Swin Transformer achieved an 88% test accuracy. Using Google Translator, pyspellchecker and gTTS libraries, the system can correct formed words, translate words and synthesize voice in regional Indian languages like Hindi, Tamil, Gujarati and in international languages like Japanese, French and Spanish. The user-friendly interface makes it easy for users to fingerspell words and obtain corrected text and translated audio outputs along with signed videos of the words. Future improvements include the ability to process data in real-time.
  • NeXt-Ray: Next Gen X-Ray diagnosis Using Deep Learning and GenAI
    Suchithra M, Shivansh Guleria, Sanskar Patil
    6th International Conference on Innovative Trends in Information Technology Secure Trustworthy and Socially Responsible AI Icitiit 2025, 2025
    The incorporation of deep learning and generative artificial intelligence (GenAl) into chest radiography presents a transformative advancement in medical diagnostics, improving the interpretability and automation. Chest radiographs play a critical role in the detection of conditions such as Infiltration, Pneumothorax, Pneumonia, Hernia and Edema; however, their interpretation has historically necessitated considerable expertise and time. This study leverages deep learning to expedite and improve the accuracy of chest X-ray analysis, while utilizing GenAl, specifically through a GPT-based model, to automate the generation of structured, clinically relevant diagnostic reports. Grad-CAM (Gradient-weighted Class Activation Mapping) visualizations are incorporated within these reports to provide interpretability, highlighting regions of interest associated with identified pathologies. These insights are presented in a structured PDF, combining visual and textual data to support rapid diagnostic decisions. This integrated system demonstrates a scalable and robust solution for chest radiography analysis, poised to advance clinical workflows and support improved patient outcomes. The synergy between deep learning, GenAl, and interpretability tools holds immense potential to revolutionize diagnostic processes and enhance healthcare delivery.
  • Intelligent Pipeline Inspection: A Deep Learning Approach for Multi-Type Defect and Methane Leak Classification
    M. Suchithra, Vishal Khumar P.D., Arnav Singh, Niloy Nath
    2025 International Conference on Computing Technologies Icoct 2025, 2025
    Pipeline infrastructure integrity is crucial for industrial safety and environmental protection. This study presents a comprehensive machine learning framework using YOLOv8 for detecting critical pipeline defects and methane leak classification. By leveraging advanced deep learning techniques, we develop a robust system capable of identifying six primary pipeline defects—Deformation, Obstacle, Rupture, Disconnect, Misalignment, and Deposition—while simultaneously classifying methane leakage events with high precision. The proposed approach integrates state-of-the-art object detection methodologies with specialized machine learning algorithms to enable real-time, scalable defect detection and environmental monitoring. Through CUDA-accelerated training on high-resolution imagery (416×416 pixels) and advanced optimization strategies, our model demonstrates exceptional performance in early defect identification and methane leak classification.
  • Real-time Deep Learning Speech Output Based Object Recognition
    Sanskar Shrivastava, Jhalak Mahendra, M. Suchithra
    Aip Conference Proceedings, 2024
  • Detection and Classification of Cervical Cancer Using Deep Learning Techniques
    Keshav Rao, M. P. Nanda, M. Suchithra
    Aip Conference Proceedings, 2024
  • Preface
    Metaheuristics Algorithm and Optimization of Engineering and Complex Systems, 2024
  • Metaheuristics algorithm and optimization of engineering and complex systems
    S., Kaliappan, M., Suchithra, R., Thanigaivelan, T., Mothilal
    Metaheuristics Algorithm and Optimization of Engineering and Complex Systems, 2024
  • Metaheuristic and machine learning optimization strategies for complex systems
    R., Thanigaivelan, M., Suchithra, S., Kaliappan, Mothilal, T.
    Metaheuristic and Machine Learning Optimization Strategies for Complex Systems, 2024
  • EEG-based multimodal emotion recognition with optimal trained hybrid classifier
    G. Kalyana Chakravarthy, M. Suchithra, Satish Thatavarti
    Multimedia Tools and Applications, 2024
  • Ensemble Machine Learning-Based Botnet Attack Detection for IoT Applications
    M. Suchithra
    Metaheuristics for Machine Learning Algorithms and Applications, 2024
  • Predicting Hospital Length of Stay Using Light Gradient Boosting Machine Regression
    M. Suchithra, Kumar Shashwat, Mohammed Shoaib Khan
    IFIP Advances in Information and Communication Technology, 2024
  • Retraction Note to: Invariant packet feature with network conditions for efficient low rate attack detection in multimedia networks for improved QoS (Journal of Ambient Intelligence and Humanized Computing, (2021), 12, 5, (5471-5477), 10.1007/s12652-020-02056-1)
    M. Suchithra, M. Baskar, J. Ramkumar, P. Kalyanasundaram, B. Amutha
    Journal of Ambient Intelligence and Humanized Computing, 2023
  • Correction to: Region Centric Minutiae Propagation Measure Orient Forgery Detection with Finger Print Analysis in Health Care Systems (Neural Processing Letters, (2023), 55, 1, (19-31), 10.1007/s11063-020-10407-4)
    M. Baskar, R. Renuka Devi, J. Ramkumar, P. Kalyanasundaram, M. Suchithra, B. Amutha
    Neural Processing Letters, 2023
  • Region Centric Minutiae Propagation Measure Orient Forgery Detection with Finger Print Analysis in Health Care Systems
    M. Baskar, R. Renuka Devi, J. Ramkumar, P. Kalyanasundaram, M. Suchithra, B. Amutha
    Neural Processing Letters, 2023
  • A Novel Deep Learning based Improved Cluster based Region Classifier Algorithm to Recognize and Categorize Emotions using EEG Signals
    G.Kalyana Chakravarthy, M Suchithra
    Proceedings of the 7th International Conference on Intelligent Computing and Control Systems Iciccs 2023, 2023
  • Application of Abstractive Summarization in Multiple Choice Question Generation
    Ayushi Mathur, M Suchithra
    Proceedings of International Conference on Computational Intelligence and Sustainable Engineering Solution Cises 2022, 2022
  • Rice Disease Detection Using Artificial Intelligence and Machine Learning Techniques to Improvise Agro-Business
    Shruti Aggarwal, M. Suchithra, N. Chandramouli, Macha Sarada, Amit Verma, D. Vetrithangam, Bhaskar Pant, Biruk Ambachew Adugna
    Scientific Programming, 2022
  • Invariant packet feature with network conditions for efficient low rate attack detection in multimedia networks for improved QoS
    M. Suchithra, M. Baskar, J. Ramkumar, P. Kalyanasundaram, B. Amutha
    Journal of Ambient Intelligence and Humanized Computing, 2021
  • QoS aware trusted service discovery architecture for optimization in MANETs
    Suchithra M, M. Ramakrishnan
    ACM International Conference Proceeding Series, 2019
  • Discovery of efficient non-Functional QoS requirements based medical web services using model driven architecture
    M. Suchithra, M. Ramakrishnan
    Journal of Medical Imaging and Health Informatics, 2016
  • Non functional QoS criterion based web service ranking
    M. Suchithra, M. Ramakrishnan
    Advances in Intelligent Systems and Computing, 2016
  • A survey on different web service discovery techniques
    M. Suchithra, M. Ramakrishnan
    Indian Journal of Science and Technology, 2015
  • A review on semantic web service discovery algorithms
    International Journal of Soft Computing, 2013