Multiclass Fetal Abnormality Detection Using Ensemble Deep Learning Techniques R. Ramya, M. Krishnamoorthi Information Technology and Control, 2026 Classifying fetal cardiotocography data is essential in the efficient prenatal risk assessment due to its potential for identifying errors or abnormalities during pregnancy. Traditional fetal heart rate (FHR) analysis frameworks, which unfortunately still rely on manual interpretation of results, subsequently lead to the inefficient use of human resources and sometimes require more time for abnormality detection. With the implementation of Machine Learning (ML) algorithms, automatic analysis and early detection of abnormalities are now possible. The model’s performance is directly influenced by the retrieval of features and the optimal management of class imbalance in the dataset. In this regard, we introduce a feature-based innovative strategy for multi-class classification in fetal cardiograph datasets based on feature importance analysis. The proposed model utilizes Random Forest (RF) for feature extraction, which employs two distinct target importance analyses: 1. class imbalance, and 2. class weights. In Phase 1, an artificial neural network and an improved TabNet model were utilised for classifying three classes: Normal, Suspect, and Pathology (NSP), with SMOTE balancing. In Phase 2, we identify the features of classes that contribute to NSP classification, and we consider nine additional features based on class weight for various cardiotography features, such as baseline, ASTV, ALTV, etc. In Phase 2, NSP classification is performed by including class 1-9 features (A, B, C, D, E, AD, DE…) and assigning class weights. Using our proposed ensemble deep learning model, the accuracy of prediction is improved. The RF model retrieves primary features from the fetal cardiograph, and complex relationships among these features enhance the representation of information. The next step is the classification stage, which applies an attention-based deep learning model, TabNet. Due to the nature of the TABNET model in handling tabular data, it can selectively focus on relevant features while ensuring explainability. The proposed model is evaluated using different performance metrics for two novel feature importance analyses. The RF+TabNet+LSTM achieves a maximum accuracy of 97% with SMOTE in NSP target classification (phase 1), while including Class weight in class1-9 features, the model achieves classification accuracy of 92% (phase II) and proves the importance of features contributing to prediction and classification. All code and the curated dataset for Multiclass Fetal Abnormality Detection are available at https://github.com/rrramyaresea/Multiclass-Fetal-Abnormality-Detection, enabling the reproducibility of our findings.
Modified ensemble machine learning-based plant leaf disease detection model with optimized K-Means clustering Vijayaganth Viswanathan, Krishnamoorthi Murugasamy Network Bristol England, 2026 In the farming sector, the automatic detection of plant leaf disease is considered a vital landmark. Farmers move long distances to consult pathologists to observe the disease, which is expensive and time-consuming. Moreover, detection of disease in a premature period is a difficult process in the existing model. Thus, all these challenges motivate us to develop an inventive plant leaf disease detection model. In the developed model, the data is gathered initially and given as input to the pre-processing step using Contrast Limited Adaptive Histogram Equalization (CLAHE). Next, the leaves are segmented from the pre-processed images, and then abnormality segmentation is done by the K-means clustering system. Here, parameters are optimized using the Opposition-based Bird Swarm Algorithm (O-BSA). Further, features were extracted from abnormality-segmented images in feature extraction. The extracted features are given in the classification step, where leaf disease detection is carried out using Optimized Ensemble Machine Learning (OEML), where, parameter optimization is done by O-BSA. Finally, the developed plant leaf detection approach is evaluated with various performance metrics, and given an accuracy of up to 92.26. These findings show that the developed model is promising over conventional methods and its effectiveness in detecting plant leaf disease.
Multimodal Deep Learning for Early Prediction of Neurodegenerative Diseases: A Comprehensive Review P. Gouthami, M. Krishnamoorthi Proceedings of the 2026 6th International Conference on Image Processing and Capsule Networks Icipcn 2026, 2026 Major neurodegenerative diseases such as Alzheimer's disease, Parkinson's disease, Amyotrophic Lateral Sclerosis and Huntington's disease are characterized by the gradual degeneration or loss of brain neurons, which drastically affects memory, thinking and movement. Though the damaged neurons are irreversible, early detection may help to slow down the disease progression and improve therapeutic outcomes. In recent years, Deep Learning, a powerful field of AI, has proved to be an effective tool due to its ability to derive intricate patterns from substantial and varied biomedical datasets. This review compiles the current deep learning methods for the early detection of neurodegenerative disorders, based on structural and functional neuroimaging, EEG signals and genetic data. With a focus on accuracy, interpretability and clinical value, it examines the contribution of key models, including convolutional neural networks, transformers, graph neural networks and recurrent neural networks. Major issues like data asymmetry, the inability to create annotated datasets and reduced model generalizability across patient groups are also highlighted and discussed about future research directions that could pave the way for better early diagnosis and treatments.
ConnectX: AI-Powered Social Networking with Dynamic Content Curation K Anusha, Krishnamoorthi M, Ramya R, Naveena Nayaki R, Suryaprakatheesh M Proceedings of 5th International Conference on Communication Computing and Electronics Systems Iccces 2026, 2026 With the rapid growth of social media platforms, several critical challenges have emerged. These include the rise of toxic and abusive content, the quick spread of fake news and misinformation, and the increasing presence of fake or automated user profiles that manipulate online interactions. Traditional content moderation systems, which rely heavily on manual reviews, are often slow and inefficient. They struggle to handle the enormous volume of real-time data generated by users, making them ineffective at maintaining a safe digital environment. To overcome these challenges, ConnectX introduces an AI-driven social networking platform that aims to create a secure, reliable, and meaningful online experience for users. ConnectX uses powerful technologies like a strong language model, and sentiment analysis to automatically detect harmful and misleading content found online. It also identifies bot-generated and suspicious automated posts, which helps reduce fraudulent practices and maintain content authenticity. By filtering and prioritizing meaningful interactions intelligently, the platform improves the overall quality of communication among users. By combining automated moderation with necessary human over- sight, ConnectX tackles important social networking challenges such as safety, trust, and usability. This ultimately encourages a more transparent, secure, and positive digital environment.
AI-Powered Anonymization for Secure Cloud-Based Medical Data Sharing: A Comprehensive Survey N. Suganya, M. Krishnamoorthi Proceedings of 2025 International Conference on Signal Processing Computation Electronics Power and Telecommunication Iconscept 2025, 2025 The large-scale implementation of Electronic Health Records (EHRs) and IoT-based medical devices in medical healthcare data has transformed current healthcare into an information-driven framework. However, digital health infrastructure mostly relies on cloud-based storage and data- sharing platforms, there exists severe security and privacy concerns as the medical data is sensitive and the cloud systems are more complex. When working with complex medical data from multiple sources, conventional anonymization approaches like suppression, generalization, k-anonymity and l-diversity often fails in securing patient privacy yet retaining the usefulness of the data. To address these problems, new innovations in Artificial Intelligence (AI) have introduced smart anonymization techniques that can automatically balance data privacy and analytical utility. This paper gives a comparative analysis of AI-powered anonymization approaches for securely sharing medical data using cloud systems. It also investigates research gaps, challenges in interoperability, transparency and explainability. To illustrate the strength and weakness of each techniques a comparative analysis framework along with visual analysis are introduced. The results show that federated learning (FL) and diffusion model (DM) provides high privacy protection and transformer- based models demonstrate strong adaptability and data utility. The survey concludes by presenting the future research directions that focuses on explainable AI, adherence to regulations and hybrid methods for anonymity in order to create healthcare networks that are reliable and protect patient privacy.
Optimizing Stock Predictions With Bi-Directional LSTM and Levy Flight Fuzzy Social Spider Optimization (LFFSSO): LSTM Model Deepa Raghunathan, M. Krishnamoorthi International Journal on Semantic Web and Information Systems, 2025 Stock Market Prediction (SMP) has developed into a significant area of research, especially in recent decades. Major novelty of the work is to develop an Evolutionary Bidirectional Long Short-Term Memory (EBi-LSTM) framework depends on investors' sentiment in tweets to Stock Market (SM). In addition, three feature selectors: the Chi-Square Test (CST), Analysis Of VAriance (ANOVA) technique and Mutual Information (MI) method are introduced for the selecting most important features. Levy Flight Fuzzy Social Spider Optimization (LFFSSO) algorithm is used for optimal tuning of parameters in the Bi-LSTM classifier. EBi-LSTM algorithm has been worked on datasets like Twitter, Stock, Weather, and Coronavirus disease (COVID-19). The proposed model extends the Valence Aware Dictionary and sEntiment Reasoner (VADER), TextBlob, and robustly optimized Bidirectional Encoder Representations from Transformers Retraining Approach (RoBERTa) for sentiment analysis. Highest results of 88.26%, 90.43%, 89.33% and 92.63% for precision, recall, F1-score and accuracy has been attained by proposed system.
Web-based Plant Leaf Disease Detection and Prescription Module Kalamani M, Krishnamoorthi M, Hiruthik Bose G, Charan S, Hariharan A, Dhilipan S Proceedings 3rd International Conference on Advancement in Computation and Computer Technologies Incacct 2025, 2025 Plant disease (PD) is a persistent issue for farmers that reduces both their income and food production. Even for subject matter specialists, it might be challenging to stop the disease from scattering to other portions of the plant once it has begun. Although it takes time and requires a close inspection of the sick region, experts may identify and diagnose the ailment. Smallholder farmers do not have simple access to these experts while constituting a major section of our nation. To identify the plant leaf disease and to evaluate its phases, a proper approach is thus required. Thanks to the present internet revolution and advancements in computer vision models, computer vision can now be used in agriculture. Recently, the Convolutional Neural Networks (CNN) are regarded as the most advanced method for classifying pictures and are capable of making a conclusive diagnosis. To diagnose the illness using a web-based application module created using deep learning techniques through the Transfer Learning approach is used in this research paper. A pre-trained method is built on photos of diverse plant leaves from the Plant Village dataset. Various efficiency metrics such as f1-score, Precision, accuracy, recall calculations and tracking are simulated to validate the efficiency of the suggested approach. The accuracy of ResNet50 with the Plant Village Dataset was around 97.5%, and it will provide information on the disease spread level on the scanned leaf. This study will improve farming practices for producing food and reduce early-stage crop loss.
Hybrid approach of deep feature extraction using BERT– OPCNN & FIAC with customized Bi-LSTM for rumor text classification Nithya K, Krishnamoorthi M, Sathishkumar Veerappampalayam Easwaramoorthy, Dhivyaa C R, Seohyun Yoo, Jaehyuk Cho Alexandria Engineering Journal, 2024 Rumor data classification in social media seems essential research due to its dependency on digital communications, and this rumor data makes social media unstable. In Natural Language Processing, word embedding, feature extraction, and classification techniques are challenging in rumor detection research. In this research, two phases of feature extraction techniques are involved for performing word embedding and deep feature extraction. In the first phase, Bidirectional Encoder Representations from Transformers with the Optimized Convolutional Neural Network (BERT-OPCNN) techniques are integrated, and in the second phase, the fastText with information gain of an ant colony optimization technique (FIAC) is proposed. Finally, the vectors formed using BERT-OPCNN and the FIAC embedding model are classified using a customized Bi-LSTM. The experiment is computed and compared with existing techniques for balanced and unbalanced datasets. The results evaluation shows that our proposed FIAC embedding with BERT-OPCNN outperforms all other existing techniques using the customized Bi-LSTM classifier.
AI Approaches in Intelligent Transportation Systems K. Nithya, S. Mythili, M. Kalamani, M. Krishnamoorthi Artificial Intelligence for Future Intelligent Transportation Smarter and Greener Infrastructure Design, 2024
Foundation of cognitive computing R. Remya, M.G. Sumithra, M. Krishnamoorthi Deep Learning for Cognitive Computing Systems Technological Advancements and Applications, 2022
Virtual Mouse Using YOLO M Krishnamoorthi, S Gowtham, K Sanjeevi, R Revanth Vishnu Proceedings of the 2022 International Conference on Innovative Computing Intelligent Communication and Smart Electrical Systems Icses 2022, 2022
Plant diseases detection using an improved CNN model Vijayaganth V, Srithar S, Krishnamoorthi M, Vijaya Kumar T Proceedings of the 2022 International Conference on Innovative Computing Intelligent Communication and Smart Electrical Systems Icses 2022, 2022
Early-Stage Depression Detector Using IoMT I Jeya Daisy, B. Vinoth Kumar, M. Krishnamoorthy 2021 International Conference on Advancements in Electrical Electronics Communication Computing and Automation Icaeca 2021, 2021
A comprehensive survey on virtual migration techniques in cloud computing International Journal of Recent Technology and Engineering, 2019
A survey on ultrasound image segmentation algorithm for detection of female pelvic masses International Journal of Recent Technology and Engineering, 2019
Tweets: Review of micro-blog based recommendation systems (RS) for news recommendation (NR) International Journal of Recent Technology and Engineering, 2019
Contact and non-contact methods of photo plethysmography International Journal of Engineering and Advanced Technology, 2018
Design of compact antenna for ultra wide band applications International Journal of Engineering and Advanced Technology, 2018
A two-stage hybrid model for intrusion detection Krishnamoorthi, N.V. Subba Reddy, U. Dinesh Acharya Proceedings 2006 14th International Conference on Advanced Computing and Communications Adcom 2006, 2006
