G. Vijayakumari

G. Vijayakumari

Auto-generated profile, based on publicly available information

G. Vijayakumari

Department of Electronics and Communication Engineering, Kangeyam Institute of Technology, Nathakadaiyur, India

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

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

Representative Publications

View All Publications
24 April 2025
Verified
Conference Paper
Closed access
VGVijayakumari G
Vijayakumari G
OMO Cyril Mathew
O Cyril Mathew
The main difficulty in developing accurate and robust models of machine learning for diagnosing skin cancer has always been that of class imbalance. This paper hereby presents an Enhanced Attention-Driven Generative Adversarial Networks (EA-GANs) framework especially designed to address class imbalance problems related to datasets used in skin cancer. Unlike the traditional Generative Adversarial Networks (GANs), in the proposed EA-GAN, adaptive attention mechanism along with a progressive feature refinement module allows for the creation of class-specific high-quality synthetic dermoscopic images. In this proposed method, the use of the adaptive attention mechanism enhances the focus synthesis of fine-grained features of the lesions and quality of the produced images. It progressively learns to improve upon underrepresented characteristics of each class. Data generation tasks are evaluated on the ISIC 2020 skin lesion dataset. The computations of Fréchet Inception Distance and Structural Similarity Index have also been used for the evaluation of generated images to determine the realism and diversity of outputs. EfficientNet-B3 classifiers is employed to classify the skin lesion based on the expansion of data, which reached a 94.2% accuracy rate on 23% advancement of baselines for the detection methodology of the minority class. This approach, besides improving the impacts of class imbalance, improves the generalization abilities of deep learning classifiers and therefore is a scalable solution, appropriate for practical clinical use. The proposed EA-GAN model sets a new benchmark for data augmentation by GANs in the context of medical imaging: ensuring reliable, high-quality synthetic samples while addressing the constraints associated with existing datasets.

11 May 2026 in Journal of the Electrochemical Society
Verified
Journal Article
Closed access
VGVijayakumari G
Vijayakumari G
COCyril Mathew O
Cyril Mathew O
Mortality of skin cancer which is one of the most prevalent cancers in the world is greatly minimized by its early diagnosis. Recent developments in artificial intelligence (AI), specifically deep learning, have displayed a lot of potential in the process of automating the process of identifying and classifying skin lesions on the basis of medical images. In this proposal, a Skin Lesion Identification using Deep learning (SLI-DL) model, which operates on the basis of electrochemical Techniques, is proposed to assist dermatologists in making quality and timely diagnostic decisions. Electrochemical techniques are analytical methods used to study chemical properties through electrical signals. The framework applies CNNs for skin lesion analysis through image preprocessing and classification into benign, malignant, or suspicious types. Trained on labeleddermoscopy datasets, pre-trained models such as VGGNet and ResNet improve accuracy. Moreover, it has real-time feedback, which allows analysing lesions instantly and making a quick clinical decision. This is an overall method with lesion segmentation, localization and classification extending the diagnostic range of a variety of dermatological disorders.

14 August 2026 in Discover Applied Sciences
Verified
Journal Article
Open access
PDF file
KDK. Deepa
K. Deepa
SVS. K. Vedhashree
S. K. Vedhashree
VGV. Gowrishankar
V. Gowrishankar
GVG. Vijayakumari
G. Vijayakumari
TST. Sathya
T. Sathya
PRP. Ramakrishnan
P. Ramakrishnan
The rapid increase in screen time across all age groups worldwide has led to simultaneous surge in retinal diseases too. This article aims to shed light on the different kinds of retinal disease classification such as Age-related Macular Degeneration (AMD), glaucoma, cataract, hypertensive retinopathy, pathological myopia etc. The proposed work avails 8000 retinal fundus images from Ocular Disease Intelligent Recognition dataset applying unsharp mask filter for preprocessing and SegNet algorithm for segmentation. YOLOv11 technique is used for extracting features and Kernel Extreme Learning Machine for classification. The results are finally optimized using Improved Squirrel Search Optimization algorithm for betterment. Comparative analysis is done against existing classifiers of ResNet50, DenseNet201 and EfficientNet. Results show that the proposed system has achieved a hold-out test accuracy of 98.3%, recall score of 97.5%, precision value of 96.9%, F1-score of 97.1% and MCC value of 96.4%. A 5-fold cross-validation further confirmed generalizability, yielding an average accuracy of 97.63%, which is considered the main estimate of model generalization.

30 January 2026
Verified
Book Chapter
Closed access
RSR. Saranya
R. Saranya
GVG. Vijayakumari
G. Vijayakumari
SSSundaresan Srividhya
Sundaresan Srividhya
VKV. Karthi
V. Karthi
MAMahmoud Ahmad Al-Khasawneh
Mahmoud Ahmad Al-Khasawneh

30 January 2026
Verified
Book Chapter
Closed access
GVG. Vijayakumari
G. Vijayakumari
NGN. Gomathi
N. Gomathi
PKP. Kokila
P. Kokila
JAJ. Amudha
J. Amudha
MSM. Chithirai Pon Selvan
M. Chithirai Pon Selvan

30 January 2026
Verified
Book Chapter
Closed access
CAC. Alakesan
C. Alakesan
GTG. Thirunavukkarasu
G. Thirunavukkarasu
DAD. Anandakumar
D. Anandakumar
GVG. Vijayakumari
G. Vijayakumari
BBBalamurugan Balusamy
Balamurugan Balusamy