Rakoth Kandan Sambandam
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Rakoth Kandan Sambandam has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Divya Vetriveeran
AIML & Data Science, Christ University
6 shared publications - J. Jenefa
Christ University
4 shared publications - C. Manjunatha Swamy
Christ University
2 shared publications - D. Vinodha
Christ University
2 shared publications - D. Vinodha
CHRIST (Deemed to be University)
2 shared publications - Arul Venkatachalam
Department of Computer Science and Engineering Bharath Institute of Higher Education and Research Chennai India
1 shared publication
Latest Publications
- Open Access
- 21 September 2026
- Published by PIAP - Industrial Research Institute for Automation and Measurements in Journal of Automation, Mobile Robotics and Intelligent Systems
- p. 170-182
Automated Detection and Severity Grading of Knee Osteoarthritis from X-ray Images Using Machine Learning with Detailed Literature Review
A groundbreaking approach to the detection and severity grading of knee osteoarthritis utilizing X-ray imaging and a Convolutional Neural Network (CNN) architecture, specifically EfficientNet B5. Osteoarthritis, a prevalent degenerative joint disease, requires accurate and timely diagnosis for effective management. The proposed methodology focuses on enhancing diagnostic accuracy through advanced deep-learning techniques. The study leverages a diverse dataset of knee X-ray images sourced from various clinical settings, encompassing a spectrum of OA severity levels. The EfficientNet B5 architecture, known for its superior performance and efficiency, is employed for feature extraction and classification. The CNN model is trained and validated on a definite dataset, achieving an impressive accuracy of 95.84% in differentiating knee osteoarthritis from normal conditions to multiple grades like Healthy, Minimal, Doubtful, Moderate, and Severe. Further enhances clinical utility and incorporates a severity grading mechanism, providing clinicians with a detailed assessment of the disease progression. The severity grading is accomplished through fine-tuning the CNN model on a sub-dataset with annotated severity levels. The integration of severity grading demonstrates the model’s ability to detect knee osteoarthritis accurately and quantify its severity with high precision. The findings of this research highlight the potential of Efficient Net B5- based CNNs as a robust and efficient tool for knee osteoarthritis diagnosis and severity grading. The reported accuracy of 95.84% positions the proposed model as a promising asset in clinical settings, offering a reliable and automated solution for early detection and precise evaluation of knee osteoarthritis. This research, with its potential to significantly improve the diagnostic capabilities of musculoskeletal disorders, ultimately enhances patient care and outcomes by providing a more accurate and timely diagnosis.
- 21 September 2026
- Published by PIAP - Industrial Research Institute for Automation and Measurements in Journal of Automation, Mobile Robotics and Intelligent Systems
- p. 170-182
Open AccessAutomated Detection and Severity Grading of Knee Osteoarthritis from X-ray Images Using Machine Learning with Detailed Literature Review
A groundbreaking approach to the detection and severity grading of knee osteoarthritis utilizing X-ray imaging and a Convolutional Neural Network (CNN) architecture, specifically EfficientNet B5. Osteoarthritis, a prevalent degenerative joint disease, requires accurate and timely diagnosis for effective management. The proposed methodology focuses on enhancing diagnostic accuracy through advanced deep-learning techniques. The study leverages a diverse dataset of knee X-ray images sourced from various clinical settings, encompassing a spectrum of OA severity levels. The EfficientNet B5 architecture, known for its superior performance and efficiency, is employed for feature extraction and classification. The CNN model is trained and validated on a definite dataset, achieving an impressive accuracy of 95.84% in differentiating knee osteoarthritis from normal conditions to multiple grades like Healthy, Minimal, Doubtful, Moderate, and Severe. Further enhances clinical utility and incorporates a severity grading mechanism, providing clinicians with a detailed assessment of the disease progression. The severity grading is accomplished through fine-tuning the CNN model on a sub-dataset with annotated severity levels. The integration of severity grading demonstrates the model’s ability to detect knee osteoarthritis accurately and quantify its severity with high precision. The findings of this research highlight the potential of Efficient Net B5- based CNNs as a robust and efficient tool for knee osteoarthritis diagnosis and severity grading. The reported accuracy of 95.84% positions the proposed model as a promising asset in clinical settings, offering a reliable and automated solution for early detection and precise evaluation of knee osteoarthritis. This research, with its potential to significantly improve the diagnostic capabilities of musculoskeletal disorders, ultimately enhances patient care and outcomes by providing a more accurate and timely diagnosis.
- 24 August 2025
- Published by Springer Nature
- p. 41-52
Enhancing Reliability and Accuracy in Wind Energy Forecast Using CNN-LSTM Hybrid Model
- 24 August 2025
- Published by Springer Nature
- p. 41-52
Enhancing Reliability and Accuracy in Wind Energy Forecast Using CNN-LSTM Hybrid Model
- 28 June 2025
- Published by World Scientific Pub Co Pte Ltd in International Journal of Modeling, Simulation, and Scientific Computing
- Vol. 17 (03)
Improvized machine learning model for extracting building footprints from collapsed images using high-resolution remote sensing images
We propose the development of a robust Enhanced U-Net framework for detecting building objects in images compromised by collapse. Traditional approaches often struggle to identify smaller buildings obstructed by taller structures, trees, or cloud coverage. However, recent advancements in machine learning algorithms present promising opportunities to address these challenges and improve the accuracy of building object detection and damage assessment. The proposed method employs the Siamese U-Net framework, enhanced with novel machine learning algorithms to overcome limitations in existing methodologies and increase the accuracy and reliability of damage assessment, even in complex scenarios. By using augmented satellite images during testing and lowering the building threshold value, our model can accurately predict damaged buildings and retrieve the footprints of smaller structures. The results of this research will advance image analysis techniques, especially in scenarios where collapsed structures pose significant identification and damage assessment challenges. This will be invaluable for government disaster management agencies, insurance companies, and other related organizations.
- 28 June 2025
- Published by World Scientific Pub Co Pte Ltd in International Journal of Modeling, Simulation, and Scientific Computing
- Vol. 17 (03)
Improvized machine learning model for extracting building footprints from collapsed images using high-resolution remote sensing images
We propose the development of a robust Enhanced U-Net framework for detecting building objects in images compromised by collapse. Traditional approaches often struggle to identify smaller buildings obstructed by taller structures, trees, or cloud coverage. However, recent advancements in machine learning algorithms present promising opportunities to address these challenges and improve the accuracy of building object detection and damage assessment. The proposed method employs the Siamese U-Net framework, enhanced with novel machine learning algorithms to overcome limitations in existing methodologies and increase the accuracy and reliability of damage assessment, even in complex scenarios. By using augmented satellite images during testing and lowering the building threshold value, our model can accurately predict damaged buildings and retrieve the footprints of smaller structures. The results of this research will advance image analysis techniques, especially in scenarios where collapsed structures pose significant identification and damage assessment challenges. This will be invaluable for government disaster management agencies, insurance companies, and other related organizations.
- 25 May 2025
- Published by Springer Nature
- p. 513-525
An Ensemble Approach Using ResNet and DenseNet for Cataract Detection
- 25 May 2025
- Published by Springer Nature
- p. 513-525
An Ensemble Approach Using ResNet and DenseNet for Cataract Detection
- 17 May 2025
- Published by Springer Nature
- p. 341-351
Sign Language Recognition Using Hand Gestures
- 29 January 2025
- Published by Wiley in Optimal Control Applications and Methods
- Vol. 46 (3) , 1193-1207
Optimized Multi‐Scale Attention Convolutional Neural Network for Micro‐Grid Energy Management System Employing in Internet of Things
The combination of micro‐grid energy management systems (EMSs) with the Internet of Things (IoT) offers a promising way to improve energy use and distribution. However, challenges such as device compatibility and the difficulty of managing energy efficiently make it hard to implement these systems effectively. This study offers a significant advancement in energy management by using IoT for microgrid systems. An Optimized Multi‐scale Attention Convolutional Neural Network for microgrid EMS employing IoT (OMACNN‐MGEMS‐IoT) is proposed in this study, which enables efficient monitoring and control of energy resources. The proposed model's input data are gathered from the MQTT dataset. This research employs a Regularized Bias‐aware Ensemble Kalman Filter (RBAEKF) for pre‐processing input data, ensuring the removal of outliers and updating missing values. The MACNN is then used for effective fault detection within the microgrid. To enhance its performance, the Sheep Flock Optimization Algorithm (SFOA) is introduced to optimize the MACNN parameters, ensuring accurate fault detection. Implemented on the MATLAB platform, the performance of the OMACNN‐MGEMS‐IoT method is assessed through various performance metrics, demonstrating significant improvements. Notably, the proposed method achieves higher cost reductions of 25%, 22%, and 26% compared to existing approaches such as the IoT platform for energy management in multi‐micro grid systems (IoT‐PEM‐MMS), a micro‐grid system infrastructure implementing IoT for efficient energy management in buildings (MSII‐IoT‐EEM) and a hybrid deep learning‐based online energy management scheme for industrial microgrids (HDL‐OEM‐IM). The findings highlight the impact of the proposed OMACNN‐MGEMS‐IoT method in enhancing energy efficiency and cost‐effectiveness in microgrid systems.
- 29 January 2025
- Published by Wiley in Optimal Control Applications and Methods
- Vol. 46 (3) , 1193-1207
Optimized Multi‐Scale Attention Convolutional Neural Network for Micro‐Grid Energy Management System Employing in Internet of Things
The combination of micro‐grid energy management systems (EMSs) with the Internet of Things (IoT) offers a promising way to improve energy use and distribution. However, challenges such as device compatibility and the difficulty of managing energy efficiently make it hard to implement these systems effectively. This study offers a significant advancement in energy management by using IoT for microgrid systems. An Optimized Multi‐scale Attention Convolutional Neural Network for microgrid EMS employing IoT (OMACNN‐MGEMS‐IoT) is proposed in this study, which enables efficient monitoring and control of energy resources. The proposed model's input data are gathered from the MQTT dataset. This research employs a Regularized Bias‐aware Ensemble Kalman Filter (RBAEKF) for pre‐processing input data, ensuring the removal of outliers and updating missing values. The MACNN is then used for effective fault detection within the microgrid. To enhance its performance, the Sheep Flock Optimization Algorithm (SFOA) is introduced to optimize the MACNN parameters, ensuring accurate fault detection. Implemented on the MATLAB platform, the performance of the OMACNN‐MGEMS‐IoT method is assessed through various performance metrics, demonstrating significant improvements. Notably, the proposed method achieves higher cost reductions of 25%, 22%, and 26% compared to existing approaches such as the IoT platform for energy management in multi‐micro grid systems (IoT‐PEM‐MMS), a micro‐grid system infrastructure implementing IoT for efficient energy management in buildings (MSII‐IoT‐EEM) and a hybrid deep learning‐based online energy management scheme for industrial microgrids (HDL‐OEM‐IM). The findings highlight the impact of the proposed OMACNN‐MGEMS‐IoT method in enhancing energy efficiency and cost‐effectiveness in microgrid systems.
- 5 December 2024
- Published by Springer Nature
- p. 429-438
A Comprehensive Review of Small Building Detection in Collapsed Images: Advancements and Applications of Machine Learning Algorithms
- 5 December 2024
- Published by Springer Nature
- p. 429-438
A Comprehensive Review of Small Building Detection in Collapsed Images: Advancements and Applications of Machine Learning Algorithms
- 17 February 2024
- Published by Springer Nature
- p. 129-140
Secure Identity Based Authentication for Emergency Communications
- 17 February 2024
- Published by Springer Nature
- p. 129-140
Secure Identity Based Authentication for Emergency Communications