Aarav Kannan Jayakumar
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Aarav Kannan Jayakumar has a total of 44 co-authors. Here are the top co-authors by number of shared publications:
R. Rajalakshmi
12 shared publicationsS. Jancy
Sathyabama Institute of Science And Technology,Department of Computer Science and Engineering,Chennai,India
3 shared publicationsG. Kavitha
2 shared publications- Nalini Manogaran
Department of CSE (Cyber Security), S.A. Engineering College (Autonomous), Chennai, India
2 shared publications R. S. Amshavalli
Sathyabama Institute of Science and Technology,Chennai,India
2 shared publications- Rajalakshmi Raja
Department of Computer Science and Engineering, Sathyabama Institute of Science and Technology, Chennai, India
2 shared publications
Latest Publications
- Open Access
- 25 August 2026
- Published by Springer Nature in Discover Computing
- Vol. 29 (1) , 571
Development of blockchain-based secured routing mechanism in software defined networks using deep learning over IoT sector
- 25 August 2026
- Published by Springer Nature in Discover Computing
- Vol. 29 (1) , 571
Open AccessDevelopment of blockchain-based secured routing mechanism in software defined networks using deep learning over IoT sector
- 7 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Clinical Decision Support System for Heart Failure Prognosis Using Gradient Boosting and Django-based Web Frameworks
Heart failure is a common, potentially fatal medical illness for which prompt and precise prognostication is essential to improve patient outcomes. In the current technology era, effective decision-making in healthcare requires the integration of sophisticated computational tools. A thorough Heart Failure Prognostication System built using the Django web development platform is presented in this project. Using a variety of patient characteristics and medical records, the system uses a data-driven approach to prognostic analysis. The web application makes use of Django’s scalability and flexibility to offer healthcare professionals a user-friendly interface for securely entering and managing patient data. Strong machine learning algorithms based on large datasets are integrated into the backend to forecast the course and severity of heart failure. Real-time risk assessment, individualized treatment suggestions, and prognostic insight visualization are just a few of this system’s key features. An engaging and responsive user experience is guaranteed by the smooth communication between the front-end and back-end made possible by the Django framework. The solution also complies with industry standards for privacy and data security. With the help of this initiative, healthcare professionals will be able to identify heart failure shortly and will lead to better patient outcomes and care. The web application built using Django not only simplifies the prognostication process but also lays the groundwork for future developments in cardiovascular health informatics.
- 7 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Clinical Decision Support System for Heart Failure Prognosis Using Gradient Boosting and Django-based Web Frameworks
Heart failure is a common, potentially fatal medical illness for which prompt and precise prognostication is essential to improve patient outcomes. In the current technology era, effective decision-making in healthcare requires the integration of sophisticated computational tools. A thorough Heart Failure Prognostication System built using the Django web development platform is presented in this project. Using a variety of patient characteristics and medical records, the system uses a data-driven approach to prognostic analysis. The web application makes use of Django’s scalability and flexibility to offer healthcare professionals a user-friendly interface for securely entering and managing patient data. Strong machine learning algorithms based on large datasets are integrated into the backend to forecast the course and severity of heart failure. Real-time risk assessment, individualized treatment suggestions, and prognostic insight visualization are just a few of this system’s key features. An engaging and responsive user experience is guaranteed by the smooth communication between the front-end and back-end made possible by the Django framework. The solution also complies with industry standards for privacy and data security. With the help of this initiative, healthcare professionals will be able to identify heart failure shortly and will lead to better patient outcomes and care. The web application built using Django not only simplifies the prognostication process but also lays the groundwork for future developments in cardiovascular health informatics.
- 29 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Intelligent Web-Enabled System for COVID-19 Classification from CT Scans Using VGG16 and Convolutional Neural Networks
Disease spreads from viruses, like COVID-19, stem from agents like SARS-CoV-2. Common symptoms associated with this virus include fever, cough, indigestion, muscle pain, and fatigue. Across many nations, the RT-PCR test stands out as the predominant molecular test employed for tracking virus transmission. There is, however, a long processing time, and the ingredients are in short supply. This work proposes to use chest CT scan images as input to identify patients with COVID-19 by utilising a deep neural network architecture. The stages like feature extraction, where the features of the picture are extracted using a pre-trained model called VGG16. The second phase in which a multilayer neural network classifies the image based on its COVID classification and NO COVID classification. Implementing a Web platform that makes our architecture easy for interested people to understand, access, and use. Using Python libraries for neural network design, the deep learning algorithm was implemented.
- 29 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Intelligent Web-Enabled System for COVID-19 Classification from CT Scans Using VGG16 and Convolutional Neural Networks
Disease spreads from viruses, like COVID-19, stem from agents like SARS-CoV-2. Common symptoms associated with this virus include fever, cough, indigestion, muscle pain, and fatigue. Across many nations, the RT-PCR test stands out as the predominant molecular test employed for tracking virus transmission. There is, however, a long processing time, and the ingredients are in short supply. This work proposes to use chest CT scan images as input to identify patients with COVID-19 by utilising a deep neural network architecture. The stages like feature extraction, where the features of the picture are extracted using a pre-trained model called VGG16. The second phase in which a multilayer neural network classifies the image based on its COVID classification and NO COVID classification. Implementing a Web platform that makes our architecture easy for interested people to understand, access, and use. Using Python libraries for neural network design, the deep learning algorithm was implemented.
- 21 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Dynamic Lexical Encoding for Neural Intelligent Language Understanding: Advancing Language Inception and Reconstruction via Deep Learning Paradigms
This research optimizes the large language models (LLMs) for language text generation and transformation by integrating natural language processing (NLP) with deep learning (DL) techniques. This is performed to configure the potential of LLMs for demonstrating its performance with feasibility and to deploy the model training for fine-tuning. The proposed approach incorporates advanced LLMs like BERT, GPT-4, T5 to implement text generation, text summarization, text translation and style transfer. The primary objective is to analyze NLP technology for language text application and transformation along with assessments of text quality improvement with contextual diversified inception and complex language processing abilities for the optimization in neuroscience. Our methodology illustrates the sequential techniques of data acquisition and tokenization approaches for pre-processing which is followed by model training. Experimental results are analyzed using METEOR scores and ROUGE scores for free-form language text generation and text summarization respectively. This research represents the potentiality of NLP with transformative abilities regarding language text processing and contextual generation for the perceptual computing.
- 21 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Dynamic Lexical Encoding for Neural Intelligent Language Understanding: Advancing Language Inception and Reconstruction via Deep Learning Paradigms
This research optimizes the large language models (LLMs) for language text generation and transformation by integrating natural language processing (NLP) with deep learning (DL) techniques. This is performed to configure the potential of LLMs for demonstrating its performance with feasibility and to deploy the model training for fine-tuning. The proposed approach incorporates advanced LLMs like BERT, GPT-4, T5 to implement text generation, text summarization, text translation and style transfer. The primary objective is to analyze NLP technology for language text application and transformation along with assessments of text quality improvement with contextual diversified inception and complex language processing abilities for the optimization in neuroscience. Our methodology illustrates the sequential techniques of data acquisition and tokenization approaches for pre-processing which is followed by model training. Experimental results are analyzed using METEOR scores and ROUGE scores for free-form language text generation and text summarization respectively. This research represents the potentiality of NLP with transformative abilities regarding language text processing and contextual generation for the perceptual computing.
- 7 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1722-1727
Improving Diagnosis of Kidney Related Type 2 Diabetes Patients using AI Enhanced Image Processing with Deep Learning and NeuroX Algorithmic Frameworks
The intersection of the Type 2 diabetes and kidney disease imposes significant challenges for accurate diagnosis and timely treatment. This research represents an advanced approach to enhance diagnostic accuracy for patients suffering from kidney-related complications of Type 2 diabetes using AIdriven image processing techniques. By optimizing the use of deep learning (DL), we propose a novel system that integrates convolutional neural networks (CNNs) for analyzing clinical diagnostic data, specifically focussed on renal scans and related diagnostic images. Our approach utilizes a comprehensive dataset of annotated kidney images to train a deep learning model, enabling the detection of subtle pathological features that are indicative of diabetes induced renal damage. The model’s architecture incorporates advanced techniques like transfer learning and data augmentation to improve its generalization capabilities and robustness. By employing AI-enhanced image processing, our system aims to provide more precise and early detection of kidney complications, facilitating better patient management and personalized treatment plans.
- 7 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1722-1727
Improving Diagnosis of Kidney Related Type 2 Diabetes Patients using AI Enhanced Image Processing with Deep Learning and NeuroX Algorithmic Frameworks
The intersection of the Type 2 diabetes and kidney disease imposes significant challenges for accurate diagnosis and timely treatment. This research represents an advanced approach to enhance diagnostic accuracy for patients suffering from kidney-related complications of Type 2 diabetes using AIdriven image processing techniques. By optimizing the use of deep learning (DL), we propose a novel system that integrates convolutional neural networks (CNNs) for analyzing clinical diagnostic data, specifically focussed on renal scans and related diagnostic images. Our approach utilizes a comprehensive dataset of annotated kidney images to train a deep learning model, enabling the detection of subtle pathological features that are indicative of diabetes induced renal damage. The model’s architecture incorporates advanced techniques like transfer learning and data augmentation to improve its generalization capabilities and robustness. By employing AI-enhanced image processing, our system aims to provide more precise and early detection of kidney complications, facilitating better patient management and personalized treatment plans.