S Berlin Shaheema

@stellamaryscoe.edu.in

Assistant Professor, Artificial Intelligence and Data Science
Stella Marys College of Engineering

S Berlin Shaheema


               Download Compact PDF  
https://researchid.co/bshaheema

RESEARCH, TEACHING, or OTHER INTERESTS

Computer Vision and Pattern Recognition, Computer Science, Artificial Intelligence
19

Scopus Publications

89

Scholar Citations

5

Scholar h-index

3

Scholar i10-index

Scopus Publications

  • Explainable AI for Diabetic Retinopathy Screening: Enhancing Clinician Trust in Deep Learning Predictions
    S Berlin Shaheema, Sujitha N., Arulraj N K, Suryaraj C K, S Satish Kumar, S Berlin Shiny
    Conference Proceedngs Wccst 2026 World Conference on Computational Science and Technology, 2026
    Diabetic Retinopathy (DR) is the causes of vision impairment in diabetes people worldwide. Early diagnosis is vital to avoid the progression of disease to severe stages like Proliferative DR. Prolonged levels of glucose cause damage to the retina, can result in irreversible blindness if not identified in time. A novel framework based on the Double Deep Q-Network (DDQN) is proposed for the effective categorization of the severity levels of DR. Proposed system, the DDQN approach is based on the reinforced learning method to optimize the decision-making process. In the DDQN approach, the overestimation bias is avoided by the decoupling of the action selection and evaluation processes. In the proposed system, the images are first p reprocessed b efore passing the images through the convolutional layer for feature extraction. The proposed system improves the stability of the learning process and the efficiency of convergence. The experimental results have shown the superiority of the DDQN-based model by providing an F1-score of 96.34%, precision of 98.67%, sensitivity of 96.54%, and accuracy of 98.95% for the healthy retina and three stages of DR. The proposed methodology is superior to existing models and has shown strong potential for practical applications.
  • IoT based Smart Waste Bins for Waste Collection, Waste segregation And disposal
    S Siddharth, T Vickneshwari, S. Berlin Shaheema, N Sujitha, G. Gifta Jerith, A. Adlin Arul
    2026 IEEE International Students Conference on Electrical Electronics and Computer Science Sceecs 2026, 2026
    The rapid urbanization of regions like Chennai, Tamil Nadu, has led to significant challenges in waste management, with approximately 100 tons of waste generated daily. Alarmingly, only 30% of this waste is properly segregated, and a mere 15% is recycled, resulting in landfill overflow, environmental pollution, and inefficient waste disposal practices. The S-Bin, an innovative IoT-based Smart Waste Bin system, has been developed to address these critical issues by introducing automation and intelligence into the waste management process. A novel waste management algorithm optimizes collection routes by analyzing factors such as waste type, bin capacity, and link stability. The S-Bin features a real-time notification system that alerts municipal authorities, coupled with a smart locking mechanism to prevent unauthorized access and ensure secure disposal. The SBin promotes higher recycling rates and provides datadriven insights for strategic waste management planning. In addition, the system integrates predictive analytics to forecast waste generation trends, enabling proactive scheduling of collection vehicles. The S-Bin’s cloud-based dashboard facilitates real-time monitoring and analytics for decision-makers. Overall, this intelligent framework supports sustainable urban development by improving operational efficiency and reducing the ecological footprint of waste disposal.
  • A Federated Learning Based Automatic Brain Tumor Segmentation and Feature Extraction in MRI Images
    S Berlin Shaheema, Jebasingh Kirubakaran S J, S. Selvi, M. Supriya, R. Isaac Sajan, A. Adlin Arul
    Conference Proceedngs Wccst 2026 World Conference on Computational Science and Technology, 2026
    The automated segmentation and feature extraction of brain tumor from the MRI brain images is extremely significant to improve the precision level in medical diagnosis and treatment. An efficient brain tumor segmentation technique based on federated learning is proposed to effectively trade off segmentation accuracy and patient data privacy. The proposed system uses efficient preprocessing techniques, segmentation algorithms, and feature extraction algorithms to determine the exact size, shape, and location of brain tumors. The proposed brain tumor segmentation technique is based on federated learning, which enables the proposed system to train multiple medical institutions simultaneously without compromising patient data privacy. The proposed brain tumor segmentation technique is tested with the BraTS datasets and shown significant improvements in segmentation accuracy, sensitivity, and specificity while maintaining patient data privacy. The proposed brain tumor segmentation technique is efficient and feasible to produce an efficient outcome in real-world scenarios. Moreover, the proposed approach improves the interoperability of healthcare systems and eliminates data.
  • Accurate Leaf-Based Identification of Plant Diseases and Classification Using Shape Attentive U-Net
    Vickneshwari T, Anjana S, K R N Aswini, Sujitha N, S. Selvi, S Berlin Shaheema
    Conference Proceedngs Wccst 2026 World Conference on Computational Science and Technology, 2026
    Accurate detection helps farmers improve crop productivity, safeguard food supplies and adopt sustainable farming methods. Conventional manual inspection and classical machine learning approaches are often time-consuming, subjective, and limited in handling complex disease patterns and high intra-class similarity among leaf images. Although deep learning techniques have achieved promising results, many existing models rely primarily on texture information and lack explicit modeling of structural and shape characteristics of lesions, which restricts their robustness and generalization capability. An automated Plant Disease detection and Classification using a Shape Attentive U-Net (SAUNet). The model integrates a dual-stream architecture consisting of a texture stream based on U-Net and a dedicated shape stream guided by boundary-aware attention. Gated convolutional layers and channel-wise attention are employed to effectively fuse texture and geometric features, enabling precise localization and discrimination of disease regions. Experiments conducted on the Kaggle PlantVillage tomato leaf dataset demonstrate that SAUNet achieves an accuracy of 99.17%, outperforming several compared models. The presented framework is lightweight and accurate, feasible for execution on limited devices such as smartphones, which will enable real-time field-level disease surveillance. This method is applied to a many agricultural applications such as smart farming, early disease surveillance, and precision crop management.
  • Multimodal brain image segmentation: a recent review, challenges and future perspectives
    Berlin Shaheema, Naresh Babu Muppalaneni
    Multimedia Tools and Applications, 2025
  • An explainable deep learning-based panoptic segmentation for brain tumor diagnosis
    Berlin Shaheema, Naresh Babu Muppalaneni, K. Suganya Devi
    Neural Computing and Applications, 2025
  • An explainable Liquid Neural Network combined with path aggregation residual network for an accurate brain tumor diagnosis
    S. Berlin Shaheema, Suganya Devi K., Naresh Babu Muppalaneni
    Computers and Electrical Engineering, 2025
  • A Hybrid Dense-Gated U-Net with an Enhanced Crow Search (ECS)-Based Cyber-Attack Detection and Classification in a Smart Grid
    J. Jasper, B. M. Praveen, S. Berlin Shaheemar, J. Anish Kumar
    Lecture Notes in Electrical Engineering, 2025
  • Explainability based Panoptic brain tumor segmentation using a hybrid PA-NET with GCNN-ResNet50
    S. Berlin Shaheema, Suganya Devi K., Naresh Babu Muppalaneni
    Biomedical Signal Processing and Control, 2024
  • Res2-UNeXt Combined with Federated Learning for Cyber-Attack Detection and Classification in Multi Area Smart Grid Power System
    Jasper J, Praveen B. M, Berlin Shaheema S
    2024 IEEE Silchar Subsection Conference Silcon 2024, 2024
    A smart grid (SG) combines an information network, a communication network, and an electrical grid. With the fast improvement of SG technology, cyber-physical systems have become more complex, making SGs more susceptible to cyber-physical attacks. Protecting energy networks and critical components of communication from external attacks is crucial for maintaining reliable and efficient power distribution. Detecting Intrusions is vital to delivering safe services and notifying system administrators. This research suggests an intrusion classification scheme to detect cyberattacks on contemporary smart power grids that integrate multi-area power systems. It utilizes Hybrid Res2-UNeXt combined with a federated learning-based optimization algorithm to learn complex electrical grid properties. Deep learning with federated learning provides a robust system for detecting and classifying intrusions, enhancing the security of smart grids. The proposed method achieved 96.6 % accuracy when analyzing the original set of features and delivered a maximum accuracy of 99% with the selected data set from the publicly available dataset from Mississippi State University. Therefore, the suggested intrusion categorization method might successfully defend smart power grid systems from online threats.
  • AI-Powered Traffic Surveillance: License Plate Recognition with Non-Helmet Detection Using YOLOv8
    S Berlin Shaheema, Benila. NS, P.Jose, I. Edwin Albert, A. Anorelin, E Infance Tony
    Proceedings 2024 IEEE International Conference on Signal Processing Informatics Communication and Energy Systems Harmonizing Signals Data and Energy Bridging the Digital Future Spices 2024, 2024
  • Diabetic Retinopathy Lesions Severity Identification Using a Hybrid Dense-Gated 3D-UNet
    S Berlin Shaheema, V.P. Kolanchinathan, S Rajalakshmi, Gurumoorthy G, S B Mohan, J. Jasper
    2nd IEEE International Conference on Data Science and Network Security Icdsns 2024, 2024
  • Efficient Diabetes Detection using Hybrid Machine Learning Model
    A S Gowri, P Jose, Karthik S M, S Berlin Shaheema, K M Karuppasway, R. Balamurugan
    International Conference on Distributed Systems Computer Networks and Cybersecurity Icdscnc 2024, 2024
  • Multi-Tier Authentication of User Access in Cloud Storage - A Survey
    S. Shiny, J. Jasper, R. Megiba Jasmine, S. Berlin Shaheema
    Aip Conference Proceedings, 2023
  • Automated Multimodal Brain Tumor Classification Using a YOLOv7 Approach
    Berlin Shaheema S, Naresh Babu Muppalaneni
    9th 2023 International Conference on Control Decision and Information Technologies Codit 2023, 2023
  • Breast cancer segmentation using a hybrid AttendSeg architecture combined with a gravitational clustering optimization algorithm using mathematical modelling
    Liping Yu, S. Berlin Shaheema, J. Sunil, Vediyappan Govindan, P. Mahimiraj, Yijie Li, Wasim Jamshed, Ahmed M. Hassan
    Open Physics, 2023
  • Panoptic Image Segmentation through Unet combined with Melody Search Optimization Algorithm for the Realistic Scene Image Understanding
    Berlin Shaheema S, Naresh Babu Muppalaneni
    2022 IEEE International Conference for Women in Innovation Technology and Entrepreneurship Icwite 2022 Proceedings, 2022
  • Benign and Malignant Brain Tumor Segmentation Using a Melody-Search Optimization Algorithm with an Extreme Softplus Learning
    Berlin Shaheema S, Naresh Babu Muppalaneni, Jasper J
    Proceedings 2022 IEEE Silchar Subsection Conference Silcon 2022, 2022
  • Natural image enhancement using a biogeography based optimization enhanced with blended migration operator
    J. Jasper, S. Berlin Shaheema, S. Berlin Shiny
    Mathematical Problems in Engineering, 2014

RECENT SCHOLAR PUBLICATIONS

  • Lung Cancer Diagnosis: Visualizing Deep Learning Decision Pathways Using Yolov8
    SB Shaheema, IE Albert, GG Jerith
    2026 IEEE Madhya Pradesh Section Conference (MPCON), 495-500 , 2026
    2026
  • IoT based Smart Waste Bins for Waste Collection, Waste segregation And disposal
    S Siddharth, T Vickneshwari, SB Shaheema, N Sujitha, GG Jerith, AA Arul
    2026 IEEE International Students' Conference on Electrical, Electronics and … , 2026
    2026
  • Accurate Leaf-Based Identification of Plant Diseases and Classification Using Shape Attentive U-Net
    V T, A S, KRN Aswini, S N, S Selvi, SB Shaheema
    2026 World Conference on Computational Science and Technology (WcCST … , 2026
    2026
  • A Federated Learning Based Automatic Brain Tumor Segmentation and Feature Extraction in MRI Images
    SB Shaheema, JK S J, S Selvi, M Supriya, RI Sajan, AA Arul
    2026 World Conference on Computational Science and Technology (WcCST … , 2026
    2026
  • Explainable AI for Diabetic Retinopathy Screening: Enhancing Clinician Trust in Deep Learning Predictions
    SB Shaheema, S N., A N K, S C K, SS Kumar, SB Shiny
    2026 World Conference on Computational Science and Technology (WcCST … , 2026
    2026
  • Brain Tumor Segmentation and Grade Classification using Deep Learning Models and Explainable AI
    B Shaheema
    National Institute of Technology Silchar , 2025
    2025
  • A Hybrid Dense-Gated U-Net with an Enhanced Crow Search (ECS)-Based Cyber-Attack Detection and Classification in a Smart Grid (2025-07-01)
    B Shaheema
    Lecture Notes in Electrical Engineering ((LNEE,volume 1371)), pp 39–50 , 2025
    2025
  • An explainable Liquid Neural Network combined with path aggregation residual network for an accurate brain tumor diagnosis
    NBM S. Berlin Shaheema , Suganya Devi K.
    Computers and Electrical Engineering 122, 23 , 2025
    2025
    Citations: 10
  • XAI Enhanced GCNN-HSA Framework for Anomaly Detection in Smart Grids
    BS Jasper J,Praveen B.M
    J.Electrical Systems 21 (1), 870-891 , 2025
    2025
  • An explainable deep learning-based panoptic segmentation for brain tumor diagnosis.
    SD Berlin Shaheema, Naresh Babu
    Neural Comput & Applic (2025). , 2025
    2025
    Citations: 3
  • Multimodal brain image segmentation: a recent review, challenges and future perspectives
    NB Berlin Shaheema
    Multimedia Tools and Applications , 2025
    2025
    Citations: 3
  • Efficient Diabetes Detection using Hybrid Machine Learning Model
    AS Gowri, P Jose, SB Shaheema, KM Karuppasway, R Balamurugan
    2024 International Conference on Distributed Systems, Computer Networks and … , 2024
    2024
  • Explainability based Panoptic brain tumor segmentation using a hybrid PA-NET with GCNN-ResNet50
    NBM S. Berlin Shaheema, Suganya Devi K
    Biomedical Signal Processing and Control 94 (106334), 14 , 2024
    2024
    Citations: 28
  • Diabetic Retinopathy Lesions Severity Identification Using a Hybrid Dense-Gated 3D-UNet
    SB Shaheema, VP Kolanchinathan, S Rajalakshmi, SB Mohan, J Jasper
    2024 International Conference on Data Science and Network Security (ICDSNS), 1-6 , 2024
    2024
    Citations: 1
  • Res2-UNeXt Combined with Federated Learning for Cyber-Attack Detection and Classification in Multi Area Smart Grid Power System
    J Jasper
    2024 IEEE Silchar Subsection Conference (SILCON 2024), 1-6 , 2024
    2024
    Citations: 2
  • AI-Powered Traffic Surveillance: License Plate Recognition with Non-Helmet Detection Using YOLOv8
    B Shaheema
    IEEE International Conference on Signal Processing, Informatics … , 2024
    2024
    Citations: 3
  • Multi-tier authentication of user access in cloud storage– A survey
    S Shiny, J Jasper, RM Jasmine, SB Shaheema
    AIP Conference Proceedings 2587, 050033 (2023), Volume , Year 2023, Pages , 2023
    2023
    Citations: 1
  • Breast cancer segmentation using a hybrid AttendSeg architecture combined with a gravitational clustering optimization algorithm using mathematical modelling
    L Yu, SB Shaheema, J Sunil, V Govindan, P Mahimiraj, Y Li, W Jamshed, ...
    Open Physics 21 (1), 20230105 , 2023
    2023
    Citations: 8
  • Automated Multimodal Brain Tumor Classification Using a YOLOv7 Approach
    NB Muppalaneni
    2023 9th International Conference on Control, Decision and Information … , 2023
    2023
    Citations: 5
  • Panoptic image segmentation through unet combined with melody search optimization algorithm for the realistic scene image understanding
    NB Muppalaneni
    2022 IEEE International Conference for Women in Innovation, Technology … , 2022
    2022
    Citations: 4

MOST CITED SCHOLAR PUBLICATIONS

  • Explainability based Panoptic brain tumor segmentation using a hybrid PA-NET with GCNN-ResNet50
    NBM S. Berlin Shaheema, Suganya Devi K
    Biomedical Signal Processing and Control 94 (106334), 14 , 2024
    2024
    Citations: 28
  • Natural image enhancement using a biogeography based optimization enhanced with blended migration operator
    J Jasper, S Berlin Shaheema, S Berlin Shiny
    Mathematical Problems in Engineering 2014 (1), 232796 , 2014
    2014
    Citations: 14
  • An explainable Liquid Neural Network combined with path aggregation residual network for an accurate brain tumor diagnosis
    NBM S. Berlin Shaheema , Suganya Devi K.
    Computers and Electrical Engineering 122, 23 , 2025
    2025
    Citations: 10
  • Breast cancer segmentation using a hybrid AttendSeg architecture combined with a gravitational clustering optimization algorithm using mathematical modelling
    L Yu, SB Shaheema, J Sunil, V Govindan, P Mahimiraj, Y Li, W Jamshed, ...
    Open Physics 21 (1), 20230105 , 2023
    2023
    Citations: 8
  • Benign and Malignant brain tumor segmentation using a melody-search optimization algorithm with an extreme softplus learning
    NB Muppalaneni, J Jasper
    2022 IEEE silchar subsection conference (SILCON), 1-7 , 2022
    2022
    Citations: 7
  • Automated Multimodal Brain Tumor Classification Using a YOLOv7 Approach
    NB Muppalaneni
    2023 9th International Conference on Control, Decision and Information … , 2023
    2023
    Citations: 5
  • Panoptic image segmentation through unet combined with melody search optimization algorithm for the realistic scene image understanding
    NB Muppalaneni
    2022 IEEE International Conference for Women in Innovation, Technology … , 2022
    2022
    Citations: 4
  • An explainable deep learning-based panoptic segmentation for brain tumor diagnosis.
    SD Berlin Shaheema, Naresh Babu
    Neural Comput & Applic (2025). , 2025
    2025
    Citations: 3
  • Multimodal brain image segmentation: a recent review, challenges and future perspectives
    NB Berlin Shaheema
    Multimedia Tools and Applications , 2025
    2025
    Citations: 3
  • AI-Powered Traffic Surveillance: License Plate Recognition with Non-Helmet Detection Using YOLOv8
    B Shaheema
    IEEE International Conference on Signal Processing, Informatics … , 2024
    2024
    Citations: 3
  • Res2-UNeXt Combined with Federated Learning for Cyber-Attack Detection and Classification in Multi Area Smart Grid Power System
    J Jasper
    2024 IEEE Silchar Subsection Conference (SILCON 2024), 1-6 , 2024
    2024
    Citations: 2
  • Diabetic Retinopathy Lesions Severity Identification Using a Hybrid Dense-Gated 3D-UNet
    SB Shaheema, VP Kolanchinathan, S Rajalakshmi, SB Mohan, J Jasper
    2024 International Conference on Data Science and Network Security (ICDSNS), 1-6 , 2024
    2024
    Citations: 1
  • Multi-tier authentication of user access in cloud storage– A survey
    S Shiny, J Jasper, RM Jasmine, SB Shaheema
    AIP Conference Proceedings 2587, 050033 (2023), Volume , Year 2023, Pages , 2023
    2023
    Citations: 1
  • Lung Cancer Diagnosis: Visualizing Deep Learning Decision Pathways Using Yolov8
    SB Shaheema, IE Albert, GG Jerith
    2026 IEEE Madhya Pradesh Section Conference (MPCON), 495-500 , 2026
    2026
  • IoT based Smart Waste Bins for Waste Collection, Waste segregation And disposal
    S Siddharth, T Vickneshwari, SB Shaheema, N Sujitha, GG Jerith, AA Arul
    2026 IEEE International Students' Conference on Electrical, Electronics and … , 2026
    2026
  • Accurate Leaf-Based Identification of Plant Diseases and Classification Using Shape Attentive U-Net
    V T, A S, KRN Aswini, S N, S Selvi, SB Shaheema
    2026 World Conference on Computational Science and Technology (WcCST … , 2026
    2026
  • A Federated Learning Based Automatic Brain Tumor Segmentation and Feature Extraction in MRI Images
    SB Shaheema, JK S J, S Selvi, M Supriya, RI Sajan, AA Arul
    2026 World Conference on Computational Science and Technology (WcCST … , 2026
    2026
  • Explainable AI for Diabetic Retinopathy Screening: Enhancing Clinician Trust in Deep Learning Predictions
    SB Shaheema, S N., A N K, S C K, SS Kumar, SB Shiny
    2026 World Conference on Computational Science and Technology (WcCST … , 2026
    2026
  • Brain Tumor Segmentation and Grade Classification using Deep Learning Models and Explainable AI
    B Shaheema
    National Institute of Technology Silchar , 2025
    2025
  • A Hybrid Dense-Gated U-Net with an Enhanced Crow Search (ECS)-Based Cyber-Attack Detection and Classification in a Smart Grid (2025-07-01)
    B Shaheema
    Lecture Notes in Electrical Engineering ((LNEE,volume 1371)), pp 39–50 , 2025
    2025