KAVITHA DURAIPANDIAN

Scopus Publications

Diagnosing Dementia Disorder Detection Using an Improved Eliminate Particle Swarm Optimization (IEPSO) Based on Convolutional Neural Networks
Kavitha Duraipandian and Murugan Ambigapathy
Springer Science and Business Media LLC

Designing a dynamic framework for people counting using YOLO-PC
Kavitha Duraipandian, Balaji Radhakrishnan Padmanabhan, and Vishesh Ranka
AIP Publishing

Impact of blockchain in healthcare system
Shiela David, Kavitha Duraipandian, Deepanjali Chandrasekaran, Digvijay Pandey, Nidhi Sindhwani, and Binay Kumar Pandey
Elsevier

The Intercorrelation between Zero trust, Dark web, and its Implications in the field of Digital Forensics
Gautham Gopan, Aditya Subramanian, Bhavani Sai K B, Kavitha Duraipandian, and Mithileysh Sathiyanarayananan
IEEE
The integral question, regarding how the data connected to an individual or an organization can be securely stored, transferred and updated has been under the hammer for quite a while. With respect to the advancements of technology and the varying trends in the usage of data, there has been attempts to make data as secure as possible, but the undeniable fact that there has been an acute rise in the number of data breaches that's happens world-wide, clearly indicates the need for a much deeper analysis that connects the origin, the methodology, the demand for breached data, the possible solutions and various other micro-factors. This is where developments like zero trust can help in deriving solutions that might help in the reduction of such breaches. The Analytics of Dark web, will give a broader picture of what happens to the breached data and how digital forensics can be more effective in this analysis.

Detection of Corona Virus from Chest X-Rays Using CNN Algorithm
Kavitha Duraipandian and Saravanakumar Somasundaram
IEEE
COVID19's global epidemic has wreaked havoc on our lives in every aspect. Healthcare systems, to be more specific, was pushed beyond their limits. Artificial intelligence developments have paved the way for the creation of complicated applications that can meet a wide range of requirements. Precision in clinical practice is necessary. In this study, machine learning-based deep learning models that were customized and pretrained were used. Convolutional Neural Networks that's utilized from detected COVID-19 respiratory pneumonia complications. Then more number of COVID-19 patients' radiographs pictures were collected locally. In Data was also used from three publicly available datasets. There are four options for evaluating performance. The public dataset was utilized first for training and testing. Second, data from both the local and national levels]. A variety of public sources were used to train and test the models. Because all diagnostic procedures have little retrieved data at the moment, medical conciliation should examine the likelihood of incorporating X-rays into illness diagnosis based on the data, while all research-based X-ray is carried out. It is possible to approach the problem from various angle.

Vaccine Awareness and Enhancing Networked Healthcare System Across the Globe
Deboleena Bhattacharyya, Kavitha Duraipandian, Sharanya Rajan, and Mithileysh Sathiyanarayanan
IEEE
Immunisation is the procedure of providing a vaccine to a person in order to protect him or her against disease. This process has been widely recognized and adopted as one of the world's most successful and cost-effective health interventions. Vaccines have been saving millions of lives worldwide from deadly infectious diseases and viruses, such as hepatitis, measles and polio. However, the COVID outbreak in the late 2019 has witnessed huge devastation on the global health front. For now, vaccine is the only cost-effective health intervention to control the spread of virus and completely eradicating it. Technological breakthroughs are making a significant contribution to the improvement of healthcare. Blockchain technology is one example of such a disruptive technology. Blockchains have the potential to improve the healthcare system in a variety of ways. In this paper, we have thoroughly discussed how we can create vaccine awareness across the globe by focusing on parents, healthcare workers, frontline workers, policymakers, media, and ultimately how everyone must work together to ensure that every individual in every country gets the vaccine. We also discussed how blockchain technology may be applied to many sectors of healthcare and the benefits it can provide in terms of enhancing global network healthcare systems.

Reveal: Online Fake Job Advert Detection Application using Machine Learning
C Prashanth, Deepanjali Chandrasekaran, Bhuvanashree Pandian, Kavitha Duraipandian, Thomas Chen, and Mithileysh Sathiyanarayanan
IEEE
New technologies are rapidly emerging to fight increasing Job scams. Online Job scams are onerous to detect, thus giving the perpetrators plenty of time to flee the area in which the crime was committed, because of this fact the criminals can be in another country far away from the scene of the crime by the time it is detected. In today's digital world, we see many such instances where a particular person is targeted. The introduction of the internet and the quick access of social networking sites (including Twitter and Instagram) prepared the door for unprecedented levels of knowledge distribution in human history. Humans can be vulnerable and easily deceived making technological advances inadequate for Online Job scams. Fake recruitments are advertised to entice people to apply, so fraudsters can gain personal information such as residential address, email address, contact number, date of birth, previous job history, bank details and steal complete identify. In this paper, we developed Reveal, a machine learning-based web application, to identify fake online job advertisements such that the applicants are cautious in applying for jobs that are authentic and reliable.

Mental Health Analysis of Employees using Machine Learning Techniques
Sujal BH, Neelima K, Deepanjali C, Bhuvanashree P, Kavitha Duraipandian, Sharanya Rajan, and Mithileysh Sathiyanarayanan
IEEE
In recent years, there has been increasing research and analysis on the importance of mental health in achieving global sustainable development goals. Mental stability can be affected by different situations - illness can affect the individual's emotion, thoughts, attitude and decision making. Mental disorders are becoming more common for the employees due to stress in their workplace. Mental illness may cause Personality Disorder, Anxiety Disorder, Phobias, Psychotic disorders, Depression, mood disorders, eating disorders and a few more. In this paper, we analyzed the cause of mental health disorders among the employees from the Open Sourcing Mental Illness (OSMI) Mental Health in Tech Survey dataset. Here we analyzed the severity of mental illness for working employees based on different factors or attributes which includes self-employed, mental health history in employees family, company offering beneficial health effects, whether the employee is in treatment for mental illness and much more using Machine Learning algorithms. The main objective for analyzing such data is to educate the public about mental illness in a working environment, thus it helps in lowering the problems with mental disorders. This paper supports and advises about the cause of severe mental health behaviours, and to prevent any unfortunate happenings due to various factors in a working atmosphere. Hence, this provides an estimation of how employees are affected in both Tech and Non-Tech companies. This analysis brings us to a conclusion of answering questions of whether the location matters, the number of employees in a company matters, whether mental health is taken into consideration or not and how these external factors affect one's mental health and physical well-being in the working sectors.

An Intellectual Methodology for Secure Health Record Mining and Risk Forecasting Using Clustering and Graph-Based Classification
D. Shiny Irene, V. Surya, D. Kavitha, R. Shankar, and S. John Justin Thangaraj
World Scientific Pub Co Pte Lt
The objective of the research work is to analyze and validate health records and securing the personal information of patients is a challenging issue in health records mining. The risk prediction task was formulated with the label Cause of Death (COD) as a multi-class classification issue, which views health-related death as the “biggest risk.” This unlabeled data particularly describes the health conditions of the participants during the health examinations. It can differ tremendously between healthy and highly ill. Besides, the problems of distributed secure data management over privacy-preserving are considered. The proposed health record mining is in the following stages. In the initial stage, effective features such as fisher score, Pearson correlation, and information gain is calculated from the health records of the patient. Then, the average values are calculated for the extracted features. In the second stage, feature selection is performed from the average features by applying the Euclidean distance measure. The chosen features are clustered in the third stage using distance adaptive fuzzy c-means clustering algorithm (DAFCM). In the fourth stage, an entropy-based graph is constructed for the classification of data and it categorizes the patient’s record. At the last stage, for security, privacy preservation is applied to the personal information of the patient. This performance is matched against the existing methods and it gives better performance than the existing ones.

Deep Analysis of Dementia Disorder Using Artificial Intelligence to Improve Healthcare Services
D Kavitha, A Murugan, and Mithileysh Sathiyanarayanan
IEEE
Dementia is a worldwide concern and early discovery of dementia is substantial for the administration of mental illness and healthy living. The data acquired from Magnetic resonance imaging (MRI), Positron Emission Tomography (PET) and Computed Tomography (CT) are used in identifying dementia. Albeit numerous non-AI and AI techniques have been implemented to understand the reasons behind dementia there is very little success to predict at an early stage. Considering the self-determination medicine, a new healthcare model that could be delivered by 5G, there is a pressing need to diagnose dementia patients in early stages using deep learning techniques and that can constantly improve assessment and diagnostic tools for distinguishing people with normal brain aging from those who will develop mild cognitive impairment. We will conduct state-of-the-art research to identify various deep learning algorithms and approaches that can be used for the diagnosis of dementia and aid in predicting at an early stage. We will develop and evaluate an intelligent algorithm that will predict and distinguish people with normal brain aging from those with mild cognitive impairment on a constant basis, which will be a good fit to the future healthcare model that will be delivered by 5G.

VIRAL: The Effectiveness of Virtual Lab Teaching in Developing Higher Order Thinking Skills
K. Raja, J. Shiny Duela, Kavitha Duraipandian, Sushila Rajan, and Mithileysh Sathiyanarayanan
IEEE
Virtual Innovation & Research Acceleration Lab (VIRAL) is an ongoing teaching pedagogy project between SRM Institute of Science & Technology, Ramapuram Campus, Chennai and MIT Square, London. In the higher education sector, the year 2020 posed new obstacles. All three waves of the COVID-19 pandemic necessitated a shift in the mentality of educators, teachers, and students from all over the world. The concerns and obstacles associated with higher educational institutions’ rapid and abrupt transition from face-to-face learning to technology-assisted virtual learning. This study examines the effectiveness of Virtual Innovation & Research Acceleration Labs (VIRAL) teaching in developing Higher Order Thinking Skills (HOTS) among engineering students. The researchers employed a Quasi-Experimental method. The sample consisted of 98 Engineering students who have regular laboratory classes as part of their curriculum. A group consisting of 49 students was selected as a control group and another group was selected as an experimental group randomly. The result of the study indicates that the experimental group has a greater mean of higher-order thinking skills than the control group. Thus it is concluded that the virtual lab teaching method for research and innovation has a significant positive effect in enhancing higher thinking skills in engineering. The study also has implications for facilitation in the higher order of inquiry processes and improvement of slow learners.

Analysis and diagnosis of dementia disorder using deep learning methods: A comprehensive review
Kavitha D.
Institute of Advanced Scientific Research

KAVITHA DURAIPANDIAN

EDUCATION

RESEARCH INTERESTS

Scopus Publications

Publications

RESEARCH OUTPUTS (PATENTS, SOFTWARE, PUBLICATIONS, PRODUCTS)

Industry, Institute, or Organisation Collaboration