Shiny Duela Johnson
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Shiny Duela Johnson has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Sridevi Srinivasan
Department of Computer Science and Engineering, SRM Institute of Science and Technology, Chennai, India
2 shared publications - Anjelin Genifer Edward Thomas
Department of Computer Science and Engineering, SRM Institute of Science and Technology, Ramapuram Campus, Chennai, India
1 shared publication Dioline Sara
Department of Information Technology Sri Sairam Institute of Technology India Article history
1 shared publication- Dioline Sara
Department of Electronics and Communication Engineering, Gitam School of Technology, Bengaluru, India
1 shared publication - Jasper John Sahayam
Ponjesly College of Engineering, Nagercoil, India
1 shared publication - K. Sivaraman
Department of Computer Science Engineering, Bharath University, Chennai - 600073, Tamil Nadu
1 shared publication
Latest Publications
- Open Access
- 22 September 2026
- Published by Springer Nature in Egyptian Journal of Radiology and Nuclear Medicine
- Vol. 57 (1) , 201
Toward interpretable multiclass semantic intelligence: explainable brain tumor segmentation and automated grade stratification
Purpose: Accurate brain tumor grade classification from magnetic resonance imaging is essential for diagnosis, treatment planning, and prognosis assessment. Although deep learning models have demonstrated strong performance, many lack clinical validation, robustness across datasets, and interpretability for decision support. This study proposes an explainable hybrid deep learning framework that integrates tumor segmentation, feature optimization, grade classification, and model attribution for high-grade and low-grade gliomas, with evaluation across multiple MRI datasets under controlled experimental conditions. Methods: The proposed framework, Towards Interpretable Multiclass Semantic Intelligence: Explainable Brain Tumor Segmentation and Automated Grade Stratifi cation, integrates brain tumor segmentation, optimized feature selection, and discriminative classifi cation to support automated tumor analysis and grade stratifi cation. Tumor regions are segmented using a Path Aggregation Network (PANet) integrated with Graph Convolutional Neural Networks (GCNNs) and ResNet-50 to preserve relevant structural characteristics. From the segmented tumor regions, shape, texture, and Histogram of Oriented Gradients (HOG) features are extracted to characterize morphological and visual patterns. A modified Reptile Search Algorithm (RSA) is then employed to select discriminative features, aiming to reduce feature redundancy and mitigate over fitting. The selected features are subsequently classified using DenseNet-201 within a Siamese architecture to enhance robustness to inter-patient variability. Model interpretability is incorporated through Gradient-weighted Class Activation Mapping (Grad-CAM), which provides visual explanations of the regions contributing to model predictions. The framework is evaluated using the BraTS 2019, BraTS 2020, and BraTS 2021 datasets, together with additional MRI datasets. Results: The proposed model achieved classification accuracies of 99.12%, 99.23%, and 98.86% on the benchmark datasets. The average precision, sensitivity, and specificity exceeded 98% across datasets, demonstrating strong discrimination between tumor grades and consistent generalization across imaging protocols. Visualization analysis highlighted tumor regions contributing to classification decisions. Conclusions: The proposed model achieved classification accuracies of 99.12%, 99.23%, and 98.86% on the benchmark datasets. The average precision, sensitivity, and specificity exceeded 98% across datasets, demonstrating strong discrimination between tumor grades and consistent generalization across imaging protocols. Visualization analysis highlighted tumor regions contributing to classification decisions. Conclusions: The proposed framework demonstrates promising performance and provides interpretable model-attribution information under the evaluated experimental conditions. It may serve as a potential computer-aided decision-support approach, but prospective multicenter validation, calibration, clinician-in-the-loop assessment, and workflow integration are required before routine clinical deployment.
- 22 September 2026
- Published by Springer Nature in Egyptian Journal of Radiology and Nuclear Medicine
- Vol. 57 (1) , 201
Open AccessToward interpretable multiclass semantic intelligence: explainable brain tumor segmentation and automated grade stratification
Purpose: Accurate brain tumor grade classification from magnetic resonance imaging is essential for diagnosis, treatment planning, and prognosis assessment. Although deep learning models have demonstrated strong performance, many lack clinical validation, robustness across datasets, and interpretability for decision support. This study proposes an explainable hybrid deep learning framework that integrates tumor segmentation, feature optimization, grade classification, and model attribution for high-grade and low-grade gliomas, with evaluation across multiple MRI datasets under controlled experimental conditions. Methods: The proposed framework, Towards Interpretable Multiclass Semantic Intelligence: Explainable Brain Tumor Segmentation and Automated Grade Stratifi cation, integrates brain tumor segmentation, optimized feature selection, and discriminative classifi cation to support automated tumor analysis and grade stratifi cation. Tumor regions are segmented using a Path Aggregation Network (PANet) integrated with Graph Convolutional Neural Networks (GCNNs) and ResNet-50 to preserve relevant structural characteristics. From the segmented tumor regions, shape, texture, and Histogram of Oriented Gradients (HOG) features are extracted to characterize morphological and visual patterns. A modified Reptile Search Algorithm (RSA) is then employed to select discriminative features, aiming to reduce feature redundancy and mitigate over fitting. The selected features are subsequently classified using DenseNet-201 within a Siamese architecture to enhance robustness to inter-patient variability. Model interpretability is incorporated through Gradient-weighted Class Activation Mapping (Grad-CAM), which provides visual explanations of the regions contributing to model predictions. The framework is evaluated using the BraTS 2019, BraTS 2020, and BraTS 2021 datasets, together with additional MRI datasets. Results: The proposed model achieved classification accuracies of 99.12%, 99.23%, and 98.86% on the benchmark datasets. The average precision, sensitivity, and specificity exceeded 98% across datasets, demonstrating strong discrimination between tumor grades and consistent generalization across imaging protocols. Visualization analysis highlighted tumor regions contributing to classification decisions. Conclusions: The proposed model achieved classification accuracies of 99.12%, 99.23%, and 98.86% on the benchmark datasets. The average precision, sensitivity, and specificity exceeded 98% across datasets, demonstrating strong discrimination between tumor grades and consistent generalization across imaging protocols. Visualization analysis highlighted tumor regions contributing to classification decisions. Conclusions: The proposed framework demonstrates promising performance and provides interpretable model-attribution information under the evaluated experimental conditions. It may serve as a potential computer-aided decision-support approach, but prospective multicenter validation, calibration, clinician-in-the-loop assessment, and workflow integration are required before routine clinical deployment.
- 22 June 2026
- Published by Springer Nature in The European Physical Journal Plus
- Vol. 141 (6) , 721
Shape-aware loss ROSE-based segmentation and shuffle attention quantum convolutional network for nasal mucosa cell classification
- 22 June 2026
- Published by Springer Nature in The European Physical Journal Plus
- Vol. 141 (6) , 721
Shape-aware loss ROSE-based segmentation and shuffle attention quantum convolutional network for nasal mucosa cell classification
- Open Access
- 4 October 2024
- Published by Springer Nature in Cognitive Neurodynamics
- Vol. 18 (6) , 4103
Correction to: Optimizing feature subset for schizophrenia detection using multichannel EEG signals and rough set theory
[This corrects the article DOI: 10.1007/s11571-023-10011-x.].
- 4 October 2024
- Published by Springer Nature in Cognitive Neurodynamics
- Vol. 18 (6) , 4103
Open AccessCorrection to: Optimizing feature subset for schizophrenia detection using multichannel EEG signals and rough set theory
[This corrects the article DOI: 10.1007/s11571-023-10011-x.].
- 25 May 2024
- Published by Springer Nature in New Generation Computing
- Vol. 42 (4) , 845-858
A Neoteric Segmentation Approach for Lung Ultrasound Images
- 25 May 2024
- Published by Springer Nature in New Generation Computing
- Vol. 42 (4) , 845-858
A Neoteric Segmentation Approach for Lung Ultrasound Images
- 8 January 2024
- Published by Springer Nature in Cognitive Neurodynamics
- Vol. 18 (2) , 431-446
Optimizing feature subset for schizophrenia detection using multichannel EEG signals and rough set theory
Schizophrenia (SZ) is a mental disorder that causes lifelong disorders based on delusions, cognitive deficits, and hallucinations. By visual assessment, SZ diagnosis is time-consuming and complicated, because brain states are more effectively revealed by electroencephalogram (EEG) signals, which are effectively used in SZ diagnosis. The application of existing deep learning methods in SZ detection is effective in the classification of 2-dimensional images, and these methods require more computational resources. Therefore, dimensionality reduction is necessary for SZ diagnosis using EEG signals. To reduce the dimensionality of the data, an improved CAO (ICAO) dimensionality reduction method is proposed, which integrates horizontal and vertical crossover approaches with AOA. The optimal feature subset is achieved by satisfying the ICAO conditions, and a fitness function is evaluated based on rough sets for improved accuracy in feature selection. Therefore a Crossover-boosted Archimedes optimization algorithm (AOA) with rough sets for Schizophrenia detection (CAORS-SD) was proposed using multichannel EEG signals from both SZ and normal patients. The signals are decomposed using multivariate empirical mode decomposition into multivariate intrinsic mode functions (MIMFs). Entropy metrics such as spectral entropy, permutation entropy, approximate entropy, sample entropy, and SVD entropy are evaluated on the MIMF domain to detect SZ. The processing time of the kernel support vector machine classifier is minimized with fewer features, reducing the risk Fof overfitting. Accuracy, sensitivity, specificity, precision, and F1-score of the CAORS-SD model should be conducted to diagnose SZ. Therefore, the proposed CAORS-SD method achieves the higher performance of accuracy, sensitivity, specificity, precision, and F1-score values of 96.34, 98.95, 96.86, 98.52, and 96.74% respectively. Also, the CAORS-SD method minimizes the error rate and significantly reduces the execution time.
- 8 January 2024
- Published by Springer Nature in Cognitive Neurodynamics
- Vol. 18 (2) , 431-446
Optimizing feature subset for schizophrenia detection using multichannel EEG signals and rough set theory
Schizophrenia (SZ) is a mental disorder that causes lifelong disorders based on delusions, cognitive deficits, and hallucinations. By visual assessment, SZ diagnosis is time-consuming and complicated, because brain states are more effectively revealed by electroencephalogram (EEG) signals, which are effectively used in SZ diagnosis. The application of existing deep learning methods in SZ detection is effective in the classification of 2-dimensional images, and these methods require more computational resources. Therefore, dimensionality reduction is necessary for SZ diagnosis using EEG signals. To reduce the dimensionality of the data, an improved CAO (ICAO) dimensionality reduction method is proposed, which integrates horizontal and vertical crossover approaches with AOA. The optimal feature subset is achieved by satisfying the ICAO conditions, and a fitness function is evaluated based on rough sets for improved accuracy in feature selection. Therefore a Crossover-boosted Archimedes optimization algorithm (AOA) with rough sets for Schizophrenia detection (CAORS-SD) was proposed using multichannel EEG signals from both SZ and normal patients. The signals are decomposed using multivariate empirical mode decomposition into multivariate intrinsic mode functions (MIMFs). Entropy metrics such as spectral entropy, permutation entropy, approximate entropy, sample entropy, and SVD entropy are evaluated on the MIMF domain to detect SZ. The processing time of the kernel support vector machine classifier is minimized with fewer features, reducing the risk Fof overfitting. Accuracy, sensitivity, specificity, precision, and F1-score of the CAORS-SD model should be conducted to diagnose SZ. Therefore, the proposed CAORS-SD method achieves the higher performance of accuracy, sensitivity, specificity, precision, and F1-score values of 96.34, 98.95, 96.86, 98.52, and 96.74% respectively. Also, the CAORS-SD method minimizes the error rate and significantly reduces the execution time.
- Open Access
- 1 November 2022
- Published by Science Publications in Journal of Computer Science
- Vol. 18 (11) , 1110-1120
Bin Bay: An Optimized Smart Waste Disposal System for a Sustainable Urban Life using LSTM and Fog Computing
To create green and smart cities, a smart and planned WasteManagement System (WMS) is a vital requirement. With the increase in the urbanpopulation and service delivery difficulties, smart cities are integratingtechnologies such as smart waste bins and Radio Frequency Identification (RFID)tags into their day-to-day operations to make solid waste management andcontrol more effective. Fog computing in waste management is becoming common insmart utilities as it provides benefits like a minimal amount of data sent tothe cloud, reduced bandwidth, and Low data latency. The proposed work focuseson residential customers using waste collection services. Smart Waste DisposalSystem for a sustainable urban life using Fog Computing called Bin Bay isproposed to optimize the collection of waste with the shortest path usingGoogle Map API support from Fog Layer for which four major parameters areconsidered namely: The distance between truck and bin, the status of the bin(what percentage of it is filled), the capacity of the truck and the positionof the truck. The Bin Bay model helps to savetime by determining the most efficient waste pickup route for waste disposal.The RFID Reader and Scanner help in updating the fog nodes with the date, time,and volume of trash pickup which could be further analyzed to keep on-timeservice delivery. Also, predictions on future waste growth shall be determinedby doing in-depth analysis in the cloud using LSTM which will intelligentlylearn and predict the waste patterns based on the summary submitted by theaggregate node of the fog layer and since the optimized routes lead to lessfuel consumption, CO2 emission also shall be greatly reduced.
- 1 November 2022
- Published by Science Publications in Journal of Computer Science
- Vol. 18 (11) , 1110-1120
Open AccessBin Bay: An Optimized Smart Waste Disposal System for a Sustainable Urban Life using LSTM and Fog Computing
To create green and smart cities, a smart and planned WasteManagement System (WMS) is a vital requirement. With the increase in the urbanpopulation and service delivery difficulties, smart cities are integratingtechnologies such as smart waste bins and Radio Frequency Identification (RFID)tags into their day-to-day operations to make solid waste management andcontrol more effective. Fog computing in waste management is becoming common insmart utilities as it provides benefits like a minimal amount of data sent tothe cloud, reduced bandwidth, and Low data latency. The proposed work focuseson residential customers using waste collection services. Smart Waste DisposalSystem for a sustainable urban life using Fog Computing called Bin Bay isproposed to optimize the collection of waste with the shortest path usingGoogle Map API support from Fog Layer for which four major parameters areconsidered namely: The distance between truck and bin, the status of the bin(what percentage of it is filled), the capacity of the truck and the positionof the truck. The Bin Bay model helps to savetime by determining the most efficient waste pickup route for waste disposal.The RFID Reader and Scanner help in updating the fog nodes with the date, time,and volume of trash pickup which could be further analyzed to keep on-timeservice delivery. Also, predictions on future waste growth shall be determinedby doing in-depth analysis in the cloud using LSTM which will intelligentlylearn and predict the waste patterns based on the summary submitted by theaggregate node of the fog layer and since the optimized routes lead to lessfuel consumption, CO2 emission also shall be greatly reduced.
- Open Access
- 15 November 2015
- Published by Indian Society for Education and Environment in Indian Journal of Science and Technology
- Vol. 8 (32)
Cloud Services for Efficient Log Management
In the web server, we can find much net traffic, insecurity over the application and also lack of performance measures for the application. Here the time consumption is more for searching the log and, risk of data loss if a server crashes and the fact that data are distributed across multiple servers, storage and processing are outsourced to third party providers and this could result in delays in responding to user requests or other problems that raise concern. So by using software as a service (SaaS) and platform as a service (PaaS) in the cloud, we are enhancing the visualization and performances of application over internet, and then deploying the overall application in the cloud to manage the log details and database storage. In this application, the user can be authenticated by their username and password during sign in. The logs can be managed based on the user activity and can be viewed only by the administrator. The admin activity also can be viewed separately in the admin logs. Thus the cloud itself allows multiple users to visit our site at a time without any net traffic and, if the server crashes, then there is no loss of data which has been stored in the database.
- 15 November 2015
- Published by Indian Society for Education and Environment in Indian Journal of Science and Technology
- Vol. 8 (32)
Open AccessCloud Services for Efficient Log Management
In the web server, we can find much net traffic, insecurity over the application and also lack of performance measures for the application. Here the time consumption is more for searching the log and, risk of data loss if a server crashes and the fact that data are distributed across multiple servers, storage and processing are outsourced to third party providers and this could result in delays in responding to user requests or other problems that raise concern. So by using software as a service (SaaS) and platform as a service (PaaS) in the cloud, we are enhancing the visualization and performances of application over internet, and then deploying the overall application in the cloud to manage the log details and database storage. In this application, the user can be authenticated by their username and password during sign in. The logs can be managed based on the user activity and can be viewed only by the administrator. The admin activity also can be viewed separately in the admin logs. Thus the cloud itself allows multiple users to visit our site at a time without any net traffic and, if the server crashes, then there is no loss of data which has been stored in the database.