Revolutionizing Education: AI and Emerging Cloud Technologies for On-Demand Learning Advanced Explorations in Machine Learning Computer Vision and Iot, 2026
Quantum-Enhanced Zero-Knowledge Compression Used for Cloud IoT Healthcare: A Scalable, Privacy-Preserving QZ-HCN Framework Rajasekaran P., Duraipandian M., Johny Renoald Albert, R. Jamuna, Usha Moorthy International Journal of Intelligent Systems, 2026 The Internet of Medical Things (IoMT) in the IoT with Cloud Healthcare (CHI) creates a high volume of real‐time medical data, but traditional compression methods suffer high computation costs, privacy leaks and quantum attacks, while advanced cryptographic algorithms such as homomorphic encryption are costly and have poor scalability for the real‐time system application. In this work, we propose a quantum‐enhanced zero‐knowledge healthcare compression network (QZ‐HCN) that associates zero‐knowledge proofs (ZKPs) with quantum‐inspired deep learning (QIDL) by introducing an innovative adaptive quantum‐supported ZKP verification mechanism (AQ‐ZKV) and a quantum fusion autoconventional neural network (QF‐AutoCNN) technique to achieve efficient, privacy‐preserving compression. For healthcare IoT datasets, QZ‐HCN can reach 98.16% in accuracy, 97.09% in F‐measure, 96.32% in precision and 97.45% in recall, with a throughput of 449.57 bits/s; processing time is reduced to 0.85 s, and memory cost is minimised to be only 192 kbits, which outperforms CNN‐Encryption (90.23% accuracy), proxy re‐encryption and homomorphic encryption by at most 13 percentage points in accuracy and 75 percentage points in memory efficiency. The secure and scalable management for CHI data is achieved by QZ‐HCN, which solves the problems of privacy threats and space costs of real‐time medical applications.
Brain Tumour Detection System Using Deep CNN and GenAI Poovizhi P, Rajasekaran P, Vijaya Kumar T, Sadhana S A, Sanjana N, Aakash S 2026 6th International Conference on Advances in Electrical Computing Communications and Sustainable Technologies Icaect 2026, 2026 The identifying of brain tumor is one of the most important elements of the whole process of medical diagnosis nowadays and the result must be very accurate and explainable so that the clinical decision-making process be enabled. In general, the new method can be utilized for a simplified version of a deep convolutional neural network (CNN) which is adjusted to emphasis on the extraction and the identification of the predominant features of the medical imaging data. When all the redundant or unnecessary data are rejected, the process becomes very precise for diagnostics and at the same time, it has a very small computational load. The core of this process is a model that has been pre-trained and optimized to identify tumor patterns by feature prioritization. Light, a few features, and an effective model design were what mainly contributed to the successful carrying out of the categorization. Hence, for clinical application which makes the framework computationally and scaled, this is highly suitable. One of the things that separate this hybrid system from the others is that this is a GenAI (Generative Artificial Intelligence) module that has been implemented in order to solve the interpretability problem of conventional deep learning models. The reason for classification is the combination of the most important tumor features gained from the study together with the descriptive, human-like explanations that are the result of the generative power of this component, thus, the CNN is persuaded to make its output further. The feedback loop, in addition to the explanation, points out and reports any errors or inaccuracies in the classification) thus, there is a way for the system to be improved by the use of the reiterative approach. Moreover, it was very user-friendly and the doctors could have an intuitive interface to use the system for interpretability and categorization. The solution is a revelation of the potential of future AI-based healthcare technology development and also more brain tumor detection is the best way to the improved results in the form of diagnostic results that are more accurate and medical doctors that are empowered.
Privacy-enhanced data compression using quantum zk-SNARKs and variational auto-encoders in cloud-IoT based healthcare sensor data for medical applications Rajasekaran P, Duraipandian M, Johny Renoald Albert Aip Advances, 2025 The increasing rate of growth of the Internet of Things (IoT) in cloud-hospitality health has brought in data storage, transmission, and security challenges with the advent of quantum-enabled threats. Traditional compression methods struggle with computational inefficiency and the threat of invasion of privacy. This paper proposes a Quantum-Enhanced Zero-Knowledge Healthcare Compression Network for solving these challenges by combining Zero-Knowledge Proofs and Quantum-Inspired Deep Learning. The main goal is to provide privacy-preserving, efficient data compression along with optimizing computation costs and safeguarding sensitive healthcare records. Drawbacks in present cryptographic techniques, e.g., high computational costs in homomorphic encryption and scalability limitations in blockchain, require a novelty Adaptive Quantum-Assisted Zero-Knowledge Verification and Quantum Fusion-AutoCNN Encoder (QF-AutoCNN) to overcome this research. This work’s originality lies in combining Quantum zk-SNARKs, Hybrid Quantum Feature Encoding, and Reinforcement Learning-Based Challenge Optimization to provide better security, compression ratio, and verification efficiency. Experimental results show better accuracy (0.9816), improved F-measure (0.9709), and less computational overhead, better than other current methods such as convolutional neural networks-encryption and proxy re-encryption. This research greatly adds to safe cloud healthcare IoT by lessening privacy threats, maximizing storage space, and minimizing processing time, guaranteeing real-time handling of medical information.
Vision-cart: IoT - Enabled Smart Glasses for an Enhanced Shopping Experience Poovizhi P, Vijaya Kumar T, Rajasekaran P, Priya Dharshini S, Rithiha M, Dhanushkumaran A Proceedings of 7th International Conference on Inventive Material Science and Applications Icima 2025, 2025 Vision-Cart is a cutting-edge assistive system on smart glasses to improve the shopping experience, with the aim of assisting visually impaired individuals by utilizing advanced AI and computer vision technologies. The goal of this project is to create an AI-based object detection system with which visually impaired individuals can easily move around and identify objects in their surroundings. With a high-resolution camera, the system translates visual data and YOLOv5 enables real-time object detection, identifying individual objects, shelves, and other useful items. These objects are conveyed to the user in real-time audio signals via earphones, making it simpler to interact with the shopping environment. This wearable technology, owing to its simplicity, enables shopping as an independent and accessible process, offering visually impaired consumers greater accessibility, confidence, and independence in daily activities. Compared to existing models such as SSD MobileNet (accuracy of 90%) and Faster R-CNN (accuracy of 96%), the proposed Vision-Cart system utilizing YOLOv5n achieves an overall detection accuracy of 98%, ensuring both reliability and real-time performance. Blind consumers tend to face challenges in identifying products on their own while shopping because of the lack of efficient assistive technologies. This paper resolves the problem through the creation of Vision-Cart, an IoT-based smart glasses platform consisting of a high-resolution camera, YOLOv5n object detection, OCR, and a text-to-speech (TTS) module for real-time product recognition. The platform takes product images, undergoes processing through computer vision algorithms, and provides audio feedback through earphones. The result shows increased accuracy in object recognition and shopping independence, which allows visually impaired users to make their own choices without any outside help.
Optimized Deep Learning for Breast Cancer Detection using APO Algorithm Vijaya Kumar T, Rajasekaran P, Kaviya D S, Kavin Kumar V, Prithivirajan C M, Ragul B Proceedings of 3rd International Conference on Intelligent Cyber Physical Systems and Internet of Things Icoici 2025, 2025 Deep learning has revolutionized Medical Image Classification, significantly improving the precision and reliability of disease diagnosis. This study presents an optimized Deep Learning framework for Breast Cancer Detection using the Arctic Puffin Optimization (APO) algorithm. APO was employed to fine-tune critical hyperparameters of deep convolutional neural networks (CNNs), effectively balancing exploration and exploitation, accelerating convergence, and reducing training time. The proposed APO-CNN model was evaluated on a high-resolution mammogram dataset, achieving a classification accuracy of 99.64%, outperforming conventional optimization approaches. Experimental findings demonstrate that APO not only enhances model precision but also reduces overfitting and improves generalization, making it highly suitable for real-world clinical deployment. By integrating APO into Deep Learning frameworks, this work provides a scalable and adaptable solution that supports radiologists in early and accurate diagnosis, contributing to improved clinical decision-making. Furthermore, APO’s optimization strategy strengthens the interpretability of CNN models, fostering greater trust in AI-driven healthcare solutions. Future research will explore APO’s applicability across broader medical imaging domains, extending its impact beyond breast cancer classification.
Sound based Railway Track Fault Detection System using Deep CNN Architecture Rajasekaran P, Poovizhi P, S.Sangamithra, K.Tania Paulson, K.Dharane, C.Vikram Proceedings of 7th International Conference on Inventive Material Science and Applications Icima 2025, 2025 Railway track faults pose significant safety risks and operational challenges, often resulting in derailments, accidents, and service delays. Traditional inspection techniques such as manual observation and ultrasonic testing are labor-intensive, time-consuming, and may fail to detect subtle anomalies. This research addresses these limitations by developing a sound-based railway track fault detection system that captures audio signals during train movement and converts them into spectrogram images. A Convolutional Neural Network (CNN) model is trained to classify these spectrograms to detect various types of faults accurately. The proposed system achieves high accuracy in detecting subtle track faults, is cost-effective, scalable, and capable of real-time non-invasive prediction. Experimental results demonstrate the superiority of the proposed method over conventional approaches in terms of precision and early fault identification.
RETRACTION:Secure cloud storage for IoT based distributed healthcare environment using blockchain orchestrated and deep learning model P. Rajasekaran, M. Duraipandian Journal of Intelligent and Fuzzy Systems, 2024 Internet of Things (IoT), a distributed healthcare system has integrated different medical resources with sensors and actuators. In this research paper proposes a secure healthcare monitoring system for IoT based distributed healthcare systems in the cloud using blockchain and deep learning (DL) mechanisms. The proposed system involved three phases: secure data transmission, data storage, and disease classification system. Initially, the patients are authenticated via blockchain mechanism and their data is encrypted via Effective Key-based Rivest Shamir Adelman (EKRSA), in which the keys are generated using Circle chaotic map and Linear inertia weight-based Honey Badger Optimization (CLHBO) algorithm. Next, in the data storage phase, these encrypted IoT data are securely stored in the cloud using blockchain technology in a distributed manner. Finally, in the disease classification, the data are gathered from the publicly available dataset, and these collected datasets are preprocessed to handle missing values and data normalization. After that, the proposed system applies a radial basis kernel-based linear discriminant analysis (RBKLDA) model to reduce the dimensionality of the dataset. At last, the disease classification is done by optimal parameter-centered bidirectional long short-term memory (OPCBLSTM). The proposed EKRSA system archives maximum throughput of 99.05% and reliability of 99.66, which is superior to the existing approaches. The OPCBLSTM is investigated for its disease classification process, the proposed one achieves 99.64% accuracy with less processing time of 6 ms, which is superior to the existing classifiers. The experimental analysis proves that the system attained better security and classification metrics results than the existing methods.
Analysis on Early Prediction of Cotton Plant Leaf Diseases Using CatBoost Algorithm M Harish, T Vijaya Kumar, P Rajasekaran, P Poovizhi, G Pondinesh, R Vigneshwaran 2nd International Conference on Sustainable Computing and Smart Systems Icscss 2024 Proceedings, 2024 As the global population steadily increases, the imperative for improved crop production becomes increasingly critical to ensure food security for everyone. Addressing agricultural diseases is a primary step, requiring the identification and widespread monitoring of these ailments. To solve this problem, the proposed project introduces a robust Cotton Leaf Diseases Detection System employing advanced machine learning techniques. Leveraging the VGG16 model, the system extracts essential features from cotton leaf images and stores them in a CSV file. The Cat Boost algorithm is then applied to the datasets for training, resulting in the creation of a highly accurate model file. This model showcases excellent classification capabilities, allowing for the detection of various cotton leaves diseases. The project optimizes the accuracy and efficiency of disease detection by integrating deep learning and gradient boosting. The integration of image feature extraction and Cat Boost algorithm not only enhances the system's performance but also offers a scalable and adaptable solution for real-world applications. This pioneering approach represents a notable progression in precision agriculture, contributing significantly to the early and accurate diagnosis of cotton leaf diseases. This, in turn, contributes to more effective crop management strategies.
AI Based Intrusion Detection System for IoT Enabled Smart Industries Sathyaseelan R, Rajasekaran P, Vijaya Kumar T, Poovizhi P, Sasitharan T, Vignesh M 2nd International Conference on Emerging Research in Computational Science Icercs 2024, 2024 In the modern era of extensive data utilization, network security encounters significant threats necessitating innovative countermeasures. This research introduces a Hybrid Network Intrusion Detection System (HNIDS) that integrates Convolutional Neural Networks (CNNs) with k-Nearest Neighbors (KNN) to address the challenges of detecting intrusions in data-intensive, imbalanced environments. By leveraging CNNs for feature extraction and employing SMOTE oversampling combined with Tomek-links under sampling to balance datasets, the system achieves accurate detection of intrusion patterns. Evaluation results highlight enhanced detection accuracy compared to standalone models, showcasing the system's capability to manage complex network security scenarios effectively.
Data Preservation in Chatbot with Cloud Deployment Vijaya Kumar T, Rajasekaran P, Jeevika L, Lavan S, Tharshan R 7th International Conference on Trends in Electronics and Informatics Icoei 2023 Proceedings, 2023
Enhancement of V2X Technology with Artificial Intelligence R. Binisha, P. Rajasekaran, K. Kalpana, T. ChithrakumarThangaraj, U. Ramani, S. IrfanBasha Proceedings International Conference on Augmented Intelligence and Sustainable Systems Icaiss 2022, 2022
RECENT SCHOLAR PUBLICATIONS
Revolutionizing Education: AI and Emerging Cloud Technologies for On-Demand Learning P Poovizhi, V Kumar, P Rajasekaran Advanced Explorations in Machine Learning, Computer Vision, and IoT, 250-264 , 2026 2026
Brain Tumour Detection System Using Deep CNN and GenAI P Poovizhi, P Rajasekaran, T Vijaya Kumar, SA Sadhana, N Sanjana, ... 2026 Sixth International Conference on Advances in Electrical, Computing … , 2026 2026
Quantum‐Enhanced Zero‐Knowledge Compression Used for Cloud IoT Healthcare: A Scalable, Privacy‐Preserving QZ‐HCN Framework P Rajasekaran, M Duraipandian International Journal of Intelligent Systems 2026 (1) , 2026 2026
Privacy-enhanced data compression using quantum zk-SNARKs and variational auto-encoders in cloud-IoT based healthcare sensor data for medical applications JR Albert AIP Advances 15 (10) , 2025 2025 Citations: 3
Optimized Deep Learning for Breast Cancer Detection using APO Algorithm V Kumar, P Rajasekaran, DS Kaviya, CM Prithivirajan, B Ragul 2025 3rd International Conference on Intelligent Cyber Physical Systems and … , 2025 2025
Sound based Railway Track Fault Detection System using Deep CNN Architecture P Rajasekaran, P Poovizhi 2025 7th International Conference on Inventive Material Science and … , 2025 2025
Vision-cart: IoT-Enabled Smart Glasses for an Enhanced Shopping Experience P Poovizhi, V Kumar, P Rajasekaran, M Rithiha, A Dhanushkumaran 2025 7th International Conference on Inventive Material Science and … , 2025 2025
Wearable sensors in human activity recognition: A survey P Rajasekaran, AK Hemanth, JD Jayaseeli, P Robert, SS Suryah P. and Suryah, SS, Wearable Sensors in Human Activity Recognition: A Survey … , 2025 2025 Citations: 2
Network intrusion detection system using Optuna N Parekh, A Sen, P Rajasekaran, JDD Jayaseeli, P Robert 2024 International Conference on IoT Based Control Networks and Intelligent … , 2024 2024 Citations: 8
A Multi Model to Enhance The Detection and Classification of Chronic Kidney Disease Using Machine Learning RB Bharathiraj, P Rajasekaran, T Vijaya Kumar, P Poovizhi, ... 2024 International Conference on Emerging Research in Computational Science … , 2024 2024
Ai based intrusion detection system for iot enabled smart industries R Sathyaseelan, P Rajasekaran, T Vijaya Kumar, P Poovizhi, ... 2024 International Conference on Emerging Research in Computational Science … , 2024 2024 Citations: 1
analysis on early prediction of cotton plant leaf diseases Using CatBoost algorithm M Harish, TV Kumar, P Rajasekaran, P Poovizhi, G Pondinesh, ... 2024 2nd International Conference on Sustainable Computing and Smart Systems … , 2024 2024 Citations: 4
Classification of Early Skin Cancer Prediction using Nesterov-Accelerated Adaptive Moment Estimation (NADAM) Optimizer Algorithm V Harish, V Kumar, P Rajasekaran, P Poovizhi, R Sridhar 2024 International Conference on Cognitive Robotics and Intelligent Systems … , 2024 2024 Citations: 4
RETRACTED: Secure cloud storage for IoT based distributed healthcare environment using blockchain orchestrated and deep learning model P Rajasekaran, M Duraipandian Journal of Intelligent & Fuzzy Systems 46 (1), 1069-1084 , 2024 2024 Citations: 14
A Deep Learning based System for Detecting Stress Level and Recommending Movie or Music TV Kumar, P Rajasekaran, S Prabhu, V Pratheeks, S Mageshpoopathi, ... 2023 International Conference on Sustainable Computing and Smart Systems … , 2023 2023 Citations: 2
Content restricting age predictor system using artificial intelligence P Rajasekaran, V Kumar, NT Anushruthi, S Goutham, J Gurusaran 2023 8th International Conference on Communication and Electronics Systems … , 2023 2023 Citations: 3
Data Preservation in Chatbot with Cloud Deployment V Kumar, P Rajasekaran, L Jeevika, S Lavan, R Tharshan 2023 7th International Conference on Trends in Electronics and Informatics … , 2023 2023 Citations: 4
Enhancement of v2x technology with artificial intelligence R Binisha, P Rajasekaran, K Kalpana, T ChithrakumarThangaraj, ... 2022 International Conference on Augmented Intelligence and Sustainable … , 2022 2022 Citations: 7
Enhancement of embedded controller firebase database labview interactive IoT based production natural parameters monitoring U Ramani, R Binisha, P Rajasekaran, M Vignesh, P Malini 2022 International Conference on Augmented Intelligence and Sustainable … , 2022 2022 Citations: 7
Resources Provisioning Cost Optimization in a Decentralized Cloud Firewall Framework P Rajasekaran, M Pradeeshwar, S Pratheevi Kumar, RC Puneeth Kumar 2022 1st International Conference on Computational Science and Technology … , 2022 2022
MOST CITED SCHOLAR PUBLICATIONS
Evaluation of antioxidant potential in selected green leafy vegetables B Subhasree, R Baskar, RL Keerthana, RL Susan, P Rajasekaran Food chemistry 115 (4), 1213-1220 , 2009 2009 Citations: 290
Free radical scavenging activity of antitumour polysaccharide fractions isolated from Ganoderma lucidum (Fr.) P. Karst. R Baskar, R Lavanya, S Mayilvizhi, P Rajasekaran 2008 Citations: 42
Enterobacteriaceae group of organisms in sewage-fed fishes P Rajasekaran Adv. Biotech 8, 12-14 , 2008 2008 Citations: 26
RETRACTED: Secure cloud storage for IoT based distributed healthcare environment using blockchain orchestrated and deep learning model P Rajasekaran, M Duraipandian Journal of Intelligent & Fuzzy Systems 46 (1), 1069-1084 , 2024 2024 Citations: 14
Optimization of flavonoids extraction from the leaves of Tabernaemontana heyneana Wall. Using L16 ortho design T Satishkumar, R Baskar, S Shanmugam, P Rajasekaran, S Sadasivam, ... Nat Sci 6 (3), 10-21 , 2008 2008 Citations: 14
Synthetic dye decolourization, textile dye and paper industrial effluent treatment using white rot fungi Lentines edodes S Shanmugam, P Rajasekaran, JV Thanikal Desalination and Water Treatment 4 (1-3), 143-147 , 2009 2009 Citations: 12
Optimization of thermostable laccase production from Pleurotus eous using rice bran S Shanmugam, P Rajasekaran, TS Kumar Advanced Journal of Biotechnology 6 (7), 12-15 , 2008 2008 Citations: 11
Optimal process for the extraction and identification of flavonoids from the leaves of Polyalthia longifolia using L16 Orthogonal design of experiment TS Kumar, M Sampath, SV Sivachandran, S Shanmugam, P Rajasekaran International Journal of Biological and Chemical Sciences 3 (4) , 2009 2009 Citations: 9
Network intrusion detection system using Optuna N Parekh, A Sen, P Rajasekaran, JDD Jayaseeli, P Robert 2024 International Conference on IoT Based Control Networks and Intelligent … , 2024 2024 Citations: 8
Enhancement of v2x technology with artificial intelligence R Binisha, P Rajasekaran, K Kalpana, T ChithrakumarThangaraj, ... 2022 International Conference on Augmented Intelligence and Sustainable … , 2022 2022 Citations: 7
Enhancement of embedded controller firebase database labview interactive IoT based production natural parameters monitoring U Ramani, R Binisha, P Rajasekaran, M Vignesh, P Malini 2022 International Conference on Augmented Intelligence and Sustainable … , 2022 2022 Citations: 7
Typing of methicillin resistant Staphylococcus aureus using whole cell polypeptide and immunoblotting techniques K Rajaduraipandi, K Panneerselvam, K Ravikumar, P Rajasekaran, ... Adv Biotech 6, 14-17 , 2008 2008 Citations: 5
analysis on early prediction of cotton plant leaf diseases Using CatBoost algorithm M Harish, TV Kumar, P Rajasekaran, P Poovizhi, G Pondinesh, ... 2024 2nd International Conference on Sustainable Computing and Smart Systems … , 2024 2024 Citations: 4
Classification of Early Skin Cancer Prediction using Nesterov-Accelerated Adaptive Moment Estimation (NADAM) Optimizer Algorithm V Harish, V Kumar, P Rajasekaran, P Poovizhi, R Sridhar 2024 International Conference on Cognitive Robotics and Intelligent Systems … , 2024 2024 Citations: 4
Data Preservation in Chatbot with Cloud Deployment V Kumar, P Rajasekaran, L Jeevika, S Lavan, R Tharshan 2023 7th International Conference on Trends in Electronics and Informatics … , 2023 2023 Citations: 4
Privacy-enhanced data compression using quantum zk-SNARKs and variational auto-encoders in cloud-IoT based healthcare sensor data for medical applications JR Albert AIP Advances 15 (10) , 2025 2025 Citations: 3
Content restricting age predictor system using artificial intelligence P Rajasekaran, V Kumar, NT Anushruthi, S Goutham, J Gurusaran 2023 8th International Conference on Communication and Electronics Systems … , 2023 2023 Citations: 3
NOx Reduction of Diesel Engine with Madhuca Indica biodiesel using Selective Catalytic Reduction (SCR) in different flow rates C Solaimuthu, S Chitra, P Rajasekaran, B Abjith, G Jayaprakasan, ... International Organization of Scientific Research, 10 (5), 28-34 , 2014 2014 Citations: 3
Analysis of various water samples for enterobacteriaceae by MPN method JS Kumar, P Rajasekaran, N Saran, PS Kumar, JP Chandran A Journal of Biotechnology 1 , 2013 2013 Citations: 3
Wearable sensors in human activity recognition: A survey P Rajasekaran, AK Hemanth, JD Jayaseeli, P Robert, SS Suryah P. and Suryah, SS, Wearable Sensors in Human Activity Recognition: A Survey … , 2025 2025 Citations: 2
