Subhashini R

Subhashini R


Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

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

Representative Publications

View All Publications
1 January 2024 in Computer Systems Science and Engineering
Verified
Journal Article
Open access
PDF file
PNP. Nandakumar
P. Nandakumar
SRSubhashini R
Subhashini R
Heart disease is a major cause of death for many people in the world. Each year the death rate of people affected with heart disease increased a lot. Machine learning models have been widely used for the prediction of heart disease from the different University of California Irvine (UCI) Machine Learning Repositories. But, due to certain data, it predicts less accurately, whereas, for large data, its sub-model deep learning is used. Our literature work has identified that only traditional methods are used for the prediction of heart disease. It will produce less accuracy. To produce more efficacy, Euclidean Distance was used in this work for data pre-processing that will clean the unwanted data and metaheuristics bio-inspired algorithm such as elephant herding optimization (EHO) is utilized for feature selection. Then, this article proposes deep learning models such as convolutional neural network (CNN) and Inception-ResNet-v2 model for the prediction of heart disease from the benchmark dataset such as the UCI Cleveland heart dataset. Finally, the proposed hybrid model utilizes a convolutional neural network with an Inception-ResNet-v2 in the third layer of the architecture that classifies heart disease with the promising result of 98.77%, accuracy for the Cleveland dataset which outperforms all the other state-of-the-art methods. In future work, this model can be used to predict other diseases such as cancer, brain tumor and COVID-19 in available datasets for the betterment of human lives.

2 November 2022 in Biomedicines
Verified
Journal Article
Open access
PDF file
NKNallakaruppan M. K.
Nallakaruppan M. K.
SRSubhashini R
Subhashini R
SSSiva rama krishnan Somayaji
Siva rama krishnan Somayaji
SPSahaya Beni Prathiba
Sahaya Beni Prathiba
The impact analysis of deep learning models for COVID-19-infected X-ray images is an extremely challenging task. Every model has unique capabilities that can provide suitable solutions for some given problem. The prescribed work analyzes various deep learning models that are used for capturing the chest X-ray images. Their performance-defining factors, such as accuracy, f1-score, training and the validation loss, are tested with the support of the training dataset. These deep learning models are multi-layered architectures. These parameters fluctuate based on the behavior of these layers, learning rate, training efficiency, or over-fitting of models. This may in turn introduce sudden changes in the values of training accuracy, testing accuracy, loss or validation loss, f1-score, etc. Some models produce linear responses with respect to the training and testing data, such as Xception, but most of the models provide a variation of these parameters either in the accuracy or the loss functions. The prescribed work performs detailed experimental analysis of deep learning image neural network models and compares them with the above said parameters with detailed analysis of these parameters with their responses regarding accuracy and loss functions. This work also analyses the suitability of these model based on the various parameters, such as the accuracy and loss functions to various applications. This prescribed work also lists out various challenges on the implementation and experimentation of these models. Solutions are provided for enhancing the performance of these deep learning models. The deep learning models that are used in the prescribed work are Resnet, VGG16, Resnet with VGG, Inception V3, Xception with transfer learning, and CNN. The model is trained with more than 1500 images of the chest-X-ray data and tested with around 132 samples of the X-ray image dataset. The prescribed work analyzes the accuracy, f1-score, recall, and precision of these models and analyzes these parameters. It also measures parameters such as training accuracy, testing accuracy, loss, and validation loss. Each epoch of every model is recorded to measure the changes in these parameters during the experimental analysis. The prescribed work provides insight for future research through various challenges and research findings with future directions.

13 January 2022 in International Journal of Grid and High Performance Computing
Verified
Journal Article
Open access
PDF file
SRSubhashini R
Subhashini R
SESathiyamoorthy E.
Sathiyamoorthy E.
Nowadays, medical diseases are one of the primary causes of death, and it is one the major concerns of developed countries. So, the disease identification process needs a lot of attention since if the diseases are idenfied at the early stage, the rate of death can be decreased. Machine learning techniques is one of the popular approaches that is used for identifying the diseases at the early stage. In this paper, two machine learning techniques, namely Naive Bayes classification algorithm and Laplace smoothing technique are used to predict the heart disease. Here, many medical details are used, such as gender, age, fasting blood sugar, blood pressure, cholesterol, etc. to predict the hearth disease of a patient. The proposed decision system supports avoiding unnecessary diagnosis test, which can be highly beneficial to start the treatment quickly. Thus, both time and money can be saved. Both the performance analysis and the experimental results show the efficiency of the proposed scheme over the existing schemes.



6 November 2025
Verified
Conference Paper
Open access
PDF file
VMVidhyashree M
Vidhyashree M
SRSubhashini R
Subhashini R
Technology development has made cyber-security threats more sophisticated, to the point where the detection mechanisms in place are unable to handle the problem. Therefore, the key to solving this issue would be the deployment of a clever and potent intrusion detection system. They monitor computer networks by gathering and evaluating data and information, recognizing user activity and automatically defending against attacks. There is an increasing need for countermeasures for prevention, detection, and protection due to the complexity of computer networks and harmful attacks. The problem of incorrect judgment, incorrect detection, and unsuccessful lack of a consistent response to the attack is one of the essential tests for intrusion detection. The conventional cryptographic security protocols are seriously threatened by the emergence of large-scale quantum computing. They become useless due to quantum assaults, which specifically target the mathematical underpinnings of existing asymmetric cryptography techniques. Longer keys or longer hash function outputs could potentially improve security, making even basic symmetric key cryptography vulnerable, if only slightly. As a result, the cryptographic techniques currently used to protect data will no longer be sufficiently safe and will be exposed to new risks posed by quantum technology.