Hybrid Deep Learning-Ensemble Framework with Multi-Modal Feature Fusion for Early Detection of Juvenile Rheumatoid Arthritis: A Novel Predictive Analytics Approach Binny S, Dr. P. Sardar Maran International Journal of Drug Delivery Technology, 2026 Juvenile Rheumatoid Arthritis (JRA) represents a critical pediatric autoimmune condition requiring early intervention to prevent irreversible joint damage and systemic complications. Traditional diagnostic approaches suffer from delayed recognition due to atypical symptom presentations and reliance on subjective clinical assessments. This study introduces a novel hybrid deep learning-ensemble framework incorporating multi-modal feature fusion for automated JRA detection in adolescent populations aged 12-18 years. Our proposed methodology integrates clinical biomarkers, radiological imaging features, genetic predisposition indicators, and temporal symptom progression patterns through a sophisticated attentionbased neural architecture combined with ensemble learning techniques. The framework employs a three-stage pipeline: (1) multi-modal data preprocessing with advanced feature extraction using convolutional neural networks for imaging data and transformer architectures for sequential clinical measurements, (2) adaptive feature selection through genetic algorithm-optimized recursive feature elimination, and (3) hybrid classification using stacked ensemble methods combining XGBoost, LightGBM, and deep neural networks with uncertainty quantification. Experimental validation on a comprehensive dataset of 2,847 adolescent patients demonstrates superior performance with 94.3% accuracy, 92.7% sensitivity, and 95.8% specificity, significantly outperforming traditional machine learning approaches and existing clinical diagnostic protocols. The proposed framework introduces several novel contributions including temporal biomarker trend analysis, multi-scale radiological feature extraction, and explainable AI components for clinical decision support. Real-world deployment simulations indicate potential for 40% reduction in diagnostic delays and 60% improvement in early intervention outcomes. This research establishes a new paradigm for AI-assisted pediatric rheumatology diagnosis with direct implications for precision medicine and personalized treatment strategies.
An Automated ML-Based Framework for Predicting Mutation-Induced Drug Resistance via ESMFold and Molecular Docking Bharath P, Dakshesh T, P. Sardar Maran Proceedings of the 12th International Conference on Biosignals Images and Instrumentation Icbsii 2026, 2026 Mutations in the genome that lead to drug resistance are major problem in treating cancer. The study proposes an automated computer pipeline that does in silico docking to figure out how mutations affect binding to drugs. The workflow integrates modules like Biopython, PubChemPy, and the ESMFold API. They use it to model the EGFR protein, specifically the wild type version NP 005219.2, and then the L858R mutant variant. With AutoDock Vina for the molecular docking, it showed a reduction in how stable the binding is for Gefitinib when in mutant form when compared to the wild type. But the predictions from it still need future experiments to validate them, to confirm they work in actual clinical situations.
Analysis of Lidar Signal Noise Reduction in Power Spectrum Density (PSD) estimation using Filters Anigo Merjora, Sardar Maran Poongavanam 2025 IEEE International Students Conference on Electrical Electronics and Computer Science Sceecs 2025, 2025 This paper designs and compares three different filtering methods to reduce the effects of noise in power density lidar signal such as Raleigh and Mie scattering refers primarily to the elastic scattering of light. The low pass filter excludes the noise at a low-level, moving average filter takes average values of the signal, wiener filtering adapts to the local noise level and can be effective in reducing noise while preserving signal features and decompose the signal using wavelet transforms and then apply thresholding to remove noise. Simulation results reveal that the Wiener filter appears to be the most effective method for this particular signal and noise scenario. It not only improved the SNR but did so substantially. Both moving average filtering and wavelet denoising showed poorer performance in terms of SNR. This could be due to inappropriate parameter settings or inherent limitations in handling the specific noise characteristics of the signal.
An Integrated CNN-RNN Framework for Enhanced Rheumatoid Arthritis Detection and Diagnosis from Medical Imaging Binny S, Sardar Maran P Journal of Internet Services and Information Security, 2025 Rheumatoid arthritis (RA) encompasses damage to joint tissues, and its pathology can be diagnosed using available medical imaging modalities like X-rays, MRI, and Ultrasound. It is one of the lifelong progressive autoimmune diseases that require early detection and prompt treatment to improve patient outcomes. This study proposes an integrated convolutional neural network-recurrent neural network (CNN-RNN) framework to facilitate RA diagnosis by jointly using spatial and temporal features from sequentially acquired medical images. The model was trained and evaluated by using over 3000 radiographic images acquired from 500 patients, including patient-wise split into training (70%), validation (15%), and testing (15%) subsets. Intensive preprocessing included contrast enhancement, noise reduction, and normalization of image quality. The CNN module extracts discriminative spatial features from individual frames, while the RNN component captures the progress of the disease across several time points. Experimental assessment showed that this integrated model achieves a classification accuracy of 94.8%, precision-93.6%, recall-95.1%, and F1-score of 94.3%, significantly superior to traditional CNN- and SVM-based methods. The statistical significance was established using McNemar's test (p<0.01), thus confirming the superiority of the proposed approach. Together, these findings emphasize the potential of deep spatial-temporal learning towards the more accurate and earlier detection of RA in clinical settings.
Significant Analysis of Heart Disease Detection Using Artificial Intelligence Algorithms Varsha Varsha, P. Sardar Maran Journal of Neonatal Surgery, 2025 Lifestyle changes affect human health significantly, leading to heart disease and Sudden death in patients. The Healthcare Industry faces multiple challenges in managing and treating the increasing number of patients according to their disease severity level. The challenges for medical experts include early identification of symptoms & timely treatment to save them. The non-availability of efficient, effective, and automatic screening methods and limited medical data analytics augment these challenges. Some current medical applications have stated that Artificial intelligence algorithms provide efficient outcomes in predicting heart diseases. This paper conducted a detailed survey on various artificial intelligence algorithms used for heart disease prediction from various related datasets. The survey aims to identify the limitations and factual problem statements and identify solutions.
Investigation and Projection of Rheumatoid Arthritis Joint Pain Sickness Utilizing Artificial Neural Net Binny S, Sardar Maran P 2025 IEEE 1st International Conference on Innovations in Engineering and Next Generation Technologies for Sustainability Icinvents 2025, 2025 Rheumatoid Joint inflammation (RA) is a safe framework going after the human body and it will harm the bones in a more modest joint of fingers, legs, and wrist. If a person severely affected with rheumatoid arthritis disease and their usual life can loss and cause to discomfort more. Now a day, the initial diagnosis of the rheumatoid disease is very problematic. The preliminary diagnosis will help more for the patients for getting the treatment before it will affect to uncomfortable stage. The Artificial intelligence techniques is an advanced development of diagnosis of diseases in healthcare applications. This investigation study's rheumatoid joint pain conclusion is accomplished utilizing counterfeit brain organizations (ANN) innovation. In order to identify false brain organizations in the MATLAB environment, the Levenberg Marquardt back propagation technique was investigated. The framework reduces the quantity of square blunder capabilities and characterizes the data collected from those with rheumatoid arthritis and those without the disease. When compared to other back propagation processes, the ANN Levenberg Marquardt back propagation calculation provides the best display in the figure. The methodology achieves an accuracy of 85% in execution testing using the informational gathering; the perceptron technique's precision rate is 76%.
Food Wastage Management Application using Android Studio Katakam Sudheepa, Padmashetti Rashmitha, P. Sardar Maran Proceedings of the 2nd International Conference on Applied Artificial Intelligence and Computing Icaaic 2023, 2023
Analysis of the Water Quality Monitoring System L. Lakshmanan, A Jesudoss, A Sivasangari, Sardar Maran, M Mercy Theresa Proceedings of the 2020 IEEE International Conference on Communication and Signal Processing Iccsp 2020, 2020
Meteorological tower wind shear characteristics, vertical wind speed profile, and surface roughness analysis near the coastline of Chennai Songklanakarin Journal of Science and Technology, 2019
A survey on crop recommendation using machine learning International Journal of Recent Technology and Engineering, 2019
Investigation and Projection of Rheumatoid Arthritis Joint Pain Sickness Utilizing Artificial Neural Net SMP Binny S International Conference on Innovations in Engineering and Next-Generation … , 2026 2026
Enhanced approach with edge feature guidance for LiDAR signal denoising AA Merjora, PS Maran Computer Vision and Image Understanding, 104609 , 2025 2025
A review-rheumatoid arthritis using machine learning PS Binny Sujatha SECOND INTERNATIONAL CONFERENCE ON INNOVATIONS IN ELECTRONICS AND … , 2025 2025
Symbolic Sequence Frameworks for Ultra-High Fidelity with Hierarchical Representation in Multi-Stage Processing Pipelines R Rajalakshmi, G Kavitha, G Sheena, PS Maran, B IS 2025 International Conference on Sustainability, Innovation & Technology … , 2025 2025
Thunderstorm prediction over Eastern state of India during pre-monsoon season in 2018 using artificial neural network model N Vanganuru, PS Maran, L Lakshmanan, AK Mishra, A Sandeep Theoretical and Applied Climatology 156 (4), 227 , 2025 2025
Analysis of Lidar Signal Noise Reduction in Power Spectrum Density (PSD) estimation using Filters A Merjora, SM Poongavanam 2025 IEEE International Students' Conference on Electrical, Electronics and … , 2025 2025
An integrated CNN-RNN framework for enhanced rheumatoid arthritis detection and diagnosis from medical imaging S Binny, PS Maran Journal of Internet Services and Information Security 15 (2), 103-124 , 2025 2025 Citations: 6
A Major Role of Artificial Intelligence in Modern Healthcare Industries Varsha, P Sardar Maran International Conference on Advances in Artificial Intelligence and Machine … , 2024 2024
Significant Analysis of Heart Disease Detection Using Artificial Intelligence Algorithms PSM Varsha Frontiers in Health Informatics 13 (4), 273-284 , 2024 2024
Optimized shufe attention based Lidar signal denoising and temperature retrievals in the middle atmosphere PSM A. Anigo Merjora Optical and Quantum Electronics 56 , 2024 2024 Citations: 4
Real-Time Deep Learning Methodology For Pothole Diagnosis V Kalaichelvan, V Kasirajan, PS Maran 2024
MULTI-LAYERED ARCHITECTURE CONVOLUTION NEURAL NETWORKS FOR DIAGNOSING AND PREDICTING HEART DISEASES ON MULTI-MODAL PS Maran Malaysian Journal of Computer Science, 29-43 , 2023 2023 Citations: 3
SVM Algorithm based Fitness App Development with Age Classification using Andriod Studio L Flavya, V Shreya, PS Maran 2023 7th International Conference on Intelligent Computing and Control … , 2023 2023 Citations: 3
Food wastage management application using android studio K Sudheepa, P Rashmitha, PS Maran 2023 2nd International Conference on Applied Artificial Intelligence and … , 2023 2023 Citations: 11
Analysis of CBC and FCMC clustering approaches for skin melanoma segmentation using dermoscopic images SP Maniraj, PS Maran Research Journal of Pharmacy and Technology 15 (10), 4807-4811 , 2022 2022 Citations: 2
A hybrid deep learning approach for skin cancer diagnosis using subband fusion of 3D wavelets SP Maniraj, PS Maran The Journal of Supercomputing 78 (10), 12394-12409 , 2022 2022 Citations: 42
Weather Data Visualization Using IoT and Cloud P Sardar Maran, D Krishna Vamsi, D Vidya Shankar Cognitive Informatics and Soft Computing: Proceeding of CISC 2020, 849-857 , 2021 2021 Citations: 1
An integrated and Dynamic Wireless Intrusion Exposure Solutions based on Neural Network SLJ Shabu, J Refonaa, S Maran, S Dhamodaran, Vedanarayanan Journal of Physics: Conference Series 1770 (1), 012016 , 2021 2021
3d-Wavelet Transform Based Skin Cancer Classification Of Vgg-16 Network Model SP Maniraj, PS Maran Indian Journal of Computer Science and Engineering 12 (5), 1510-1518 , 2021 2021 Citations: 2
Spam and fake spam message detection framework using machine learning algorithm SL Jany Shabu, V Bose, V Bandaru, S Maran, J Refonaa Journal of Computational and Theoretical Nanoscience 17 (8), 3444-3448 , 2020 2020 Citations: 1
MOST CITED SCHOLAR PUBLICATIONS
A hybrid deep learning approach for skin cancer diagnosis using subband fusion of 3D wavelets SP Maniraj, PS Maran The Journal of Supercomputing 78 (10), 12394-12409 , 2022 2022 Citations: 42
A Survey on Crop Recommendation Using Machine Learning PSM M.V.R. Vivek, D.V.V.S.S. Sri Harsha International Journal of Recent Technology and Engineering (IJRTE) 7 (Issue … , 2019 2019 Citations: 13
Meteorological tower wind shear characteristics, vertical wind speed profile, and surface roughness analysis near the coastline of Chennai. PS Maran Songklanakarin Journal of Science & Technology 41 (4) , 2019 2019 Citations: 12
Food wastage management application using android studio K Sudheepa, P Rashmitha, PS Maran 2023 2nd International Conference on Applied Artificial Intelligence and … , 2023 2023 Citations: 11
Air quality prediction (IoT) using machine learning P Sardar Maran, BS Reddy, C Saiharshavardhan International Conference on Emerging Trends and Advances in Electrical … , 2020 2020 Citations: 9
Wind characteristics and Weibull parameter analysis to predict wind power potential along the south-east coastline of Tamil Nadu PS Maran, PM Velumurugan, BPD Batvari International Conference on Intelligent Information Technologies, 190-199 , 2018 2018 Citations: 9
A meteorological tower based wind speed prediction model using fuzzy logic. SM Poongavanam, PR Ponnusamy Ramalingam 2013 Citations: 8
An integrated CNN-RNN framework for enhanced rheumatoid arthritis detection and diagnosis from medical imaging S Binny, PS Maran Journal of Internet Services and Information Security 15 (2), 103-124 , 2025 2025 Citations: 6
Wind Direction Dependent Vertical Wind Shear and Surface Roughness Parameterization in Two different Coastal Environments A Bagavathsingh, CV Srinivas, PS Maran, R Baskaran, B Venkatraman International Journal of Geology, Agriculture and Environmental Sciences 4 (3) , 2016 2016 Citations: 6
Web enabled real time weather data analysis PS Maran, R Ponnusamy Indian Journal of Science and Technology 6 (11), 5507-5513 , 2013 2013 Citations: 6
Optimized shufe attention based Lidar signal denoising and temperature retrievals in the middle atmosphere PSM A. Anigo Merjora Optical and Quantum Electronics 56 , 2024 2024 Citations: 4
Skin Cancer - Computer Aided Diagnosis by Feature Analysis and Machine Learning: A Survey PSM S P Maniraj Indian Journal of Public Health Research & Development 9 (6), 544-549 , 2018 2018 Citations: 4
WIND ENERGY LOCATION PREDICTION BETWEEN METEOROLOGICAL STATIONS USING ANN SAB Maran P.S, Ponnusamy.R,VENKATESAN R Global NEST 16 (6), 1135-1144 , 2015 2015 Citations: 4
Wind power density estimation using meteorological tower data P Sardar Maran, R Ponnusamy Int. J. Renew. Sustain. Energy 2 (3), 110-114 , 2013 2013 Citations: 4
MULTI-LAYERED ARCHITECTURE CONVOLUTION NEURAL NETWORKS FOR DIAGNOSING AND PREDICTING HEART DISEASES ON MULTI-MODAL PS Maran Malaysian Journal of Computer Science, 29-43 , 2023 2023 Citations: 3
SVM Algorithm based Fitness App Development with Age Classification using Andriod Studio L Flavya, V Shreya, PS Maran 2023 7th International Conference on Intelligent Computing and Control … , 2023 2023 Citations: 3
Analysis of CBC and FCMC clustering approaches for skin melanoma segmentation using dermoscopic images SP Maniraj, PS Maran Research Journal of Pharmacy and Technology 15 (10), 4807-4811 , 2022 2022 Citations: 2
3d-Wavelet Transform Based Skin Cancer Classification Of Vgg-16 Network Model SP Maniraj, PS Maran Indian Journal of Computer Science and Engineering 12 (5), 1510-1518 , 2021 2021 Citations: 2
Prediction of extreme atmospheric temperature with meteorological tower observations using fuzzy logic technique P Sardar Maran, K Ashok Kumar, J Refonaa, J Shabu, A Jesudoss, ... Journal of Computational and Theoretical Nanoscience 17 (8), 3408-3411 , 2020 2020 Citations: 2
Detection of Land Uses Changes in the Drought Prone Overdrafted Hard Rock Terrain of Upper Thurinjalar Watershed of Ponnaiyar River Basin Tamil Nadu using Geospatial Techniques EVNMSM K. Santhanam, Marykutty Abraham, German Amali Jacintha International Journal of Earth and Atmospheric Sciences I 4 (3), 163-166 , 2017 2017 Citations: 2
