Low resource federated learning for classification of nail disease by deploying cross-silo and heterogeneously dataset distributions Vikas Khullar, Mohamed Abbas, Isha Kansal, Amel Ksibi, Gifty Gupta, Deepali Gupta, Sapna Juneja, Ali Nauman Scientific Reports, 2026 Nail diseases, including such common conditions as fungus, and more serious issues like melanoma, may be important clues to the overall health and require a clear diagnosis to be treated. The purpose of the paper is to create a nail disease detection system based on the advanced machine learning methods, including transfer learning and federated learning. The research seeks to show how machine learning and federated learning can be combined to detect nail disease performance with high accuracy without having to share data. The data include pictures of diverse nail conditions including Acral Lentiginous Melanoma, Onychogryphosis, and Pitting among others that are checked to maintain the quality of data in a uniform manner to facilitate the effective training of the models. The most common feature extraction models are ResNet152V2, DenseNet201, MobileNetV2, and InceptionResNetV2 that produce between 1,280 and 2,048 features per image. These characteristics are then pooled to create a unified feature space of 6,784 dimensions which is further narrowed to five major characteristics with Linear Discriminant Analysis (LDA) to create an efficient form of classification. A range of classification models, including Deep Neural Networks (DNN) and Long Short-Term Memory (LSTM) and Bidirectional LSTM (Bi-LSTM) are compared, with the last one reaching the highest classification accuracy of 91.8%. The federated learning strategy enables the joint training of DL models by different clients to ensure data-privacy and has validation-accuracy rates exceeding 99-percent in both uniformly random and structured data distributions. The proposed federated learning-based models resulted high in both uniformly random and structured data distributions.
Artificial Intelligence Architectures and Algorithms in Natural Language Processing Ecosystems Sangeetha Annam, Vikas Khullar Natural Language Processing and Artificial Intelligence A Perspective Towards Current Trends Challenges and Applications, 2026 The goal of natural language processing (NLP), which is positioned midway between computational linguistics and computer science, is to transform spoken and written natural human languages into organized, mineable data. NLP can generate responses that resemble those of a human being or interpret textual data using statistical, linguistic, and artificial intelligence (AI) techniques. Within the AI field, NLP deals with creating and putting into practice algorithms and systems that can communicate with one another using natural language. Nowadays, the performance of NLP applications has increased to an unparalleled degree. This chapter will give you a general introduction and tutorial on AI architecture and its algorithms in NLP. Also, we outline how NLP has evolved historically, as well as a summary of some of the most notable aspects, such as architectures and algorithms of AI in NLP ecosystems. This chapter includes an overview of AI techniques applied to various NLP sub-problems. Then we go into the design of contemporary AI architectures and AI algorithms in NLP ecosystems, concluding with a summary.
FSL-TM: Review on the Integration of Federated Split Learning with TinyML in the Internet of Vehicles Meenakshi Aggarwal, Vikas Khullar, Nitin Goyal Computers Materials and Continua, 2026 The Internet of Vehicles, or IoV, is expected to lessen pollution, ease traffic, and increase road safety. IoV entities’ interconnectedness, however, raises the possibility of cyberattacks, which can have detrimental effects. IoV systems typically send massive volumes of raw data to central servers, which may raise privacy issues. Additionally, model training on IoV devices with limited resources normally leads to slower training times and reduced service quality. We discuss a privacy-preserving Federated Split Learning with Tiny Machine Learning (TinyML) approach, which operates on IoV edge devices without sharing sensitive raw data. Specifically, we focus on integrating split learning (SL) with federated learning (FL) and TinyML models. FL is a decentralised machine learning (ML) technique that enables numerous edge devices to train a standard model while retaining data locally collectively. The article intends to thoroughly discuss the architecture and challenges associated with the increasing prevalence of SL in the IoV domain, coupled with FL and TinyML. The approach starts with the IoV learning framework, which includes edge computing, FL, SL, and TinyML, and then proceeds to discuss how these technologies might be integrated. We elucidate the comprehensive operational principles of Federated and split learning by examining and addressing many challenges. We subsequently examine the integration of SL with FL and various applications of TinyML. Finally, exploring the potential integration of FL and SL with TinyML in the IoV domain is referred to as FSL-TM. It is a superior method for preserving privacy as it conducts model training on individual devices or edge nodes, thereby obviating the necessity for centralised data aggregation, which presents considerable privacy threats. The insights provided aim to help both researchers and practitioners understand the complicated terrain of FL and SL, hence facilitating advancement in this swiftly progressing domain.
Natural Language Processing and Artificial Intelligence: A Perspective Toward Current Trends, Challenges, and Applications Mohit Angurala, Shikha Gupta, Abhineet Anand, Vikas Khullar Natural Language Processing and Artificial Intelligence A Perspective Towards Current Trends Challenges and Applications, 2026 This chapter digs into the dynamic and transformational convergence of natural language processing (NLP) and artificial intelligence (AI), offering a thorough examination of current trends, problems, and applications in this emerging subject. We investigate the impact of deep learning (DL) and neural networks, highlighting their critical significance in tasks ranging from language translation to sentiment analysis. Transfer learning (TL) appears as an important trend, allowing pre-trained models to be fine-tuned for specific NLP applications. The integration of multimodal AI, which combines text with visuals and audio, is being investigated for its influence on a variety of applications. However, difficulties exist, such as linguistic ambiguity, ethical problems, and the interpretability of complicated models. Despite these challenges, the applications of NLP and AI are extensive, including chatbots, sentiment analysis, and language translation.
Quantifying Resilience of CNN-Based Brain Tumor Classification Under FGSM, PGD, and BIM Attacks Reeti Jaswal, Vikas Khullar, Surya Narayan Panda, Krishnaraj Chadaga, Deepali Gupta, Sapna Juneja, Abeer A. Al-Masri IEEE Access, 2026 Medical image classification has been significantly improved by Convolutional Neural Networks (CNN), enabling efficient and accurate diagnosis, especially in detecting brain tumors. Despite their high success rate, AI infrastructure security of deep learning models remains susceptible to adversarial attacks, involving the careful creation of small imperceptible perturbations to the input data that cause the model to produce inaccurate predictions. This research investigates the resilience of a CNN-based brain tumor classification model in the presence of adversarial attack techniques such as the Fast Gradient Sign Method, Projected Gradient Descent, and the Basic Iterative Method. Trained baseline CNN to classify brain MRI images into four different categories: glioma, meningioma, pituitary tumor, and no tumor, attaining a maximum accuracy of 97.92 percent on clean data. After that, the model’s performance was tested against the three adversarial attacks. In addition, adversarial training was employed by incorporating attack-specific adversarial samples during training to improve model robustness against such perturbations. Fast Gradient Sign Method resulted in a significant accuracy drop to 67.62 percent, while Projected Gradient Descent and the Basic Iterative Method led to accuracies of 77.48 percent and 79.71 percent, respectively. Metrics such as precision, recall, PSNR, RMSE, and perturbation distances were evaluated to determine the severity of each attack. The findings indicate that although all attacks compromise the performance of the model, iterative methods like the Basic Iterative Method make attacks less noticeable and more challenging to defend. These findings highlight the need for AI security by adding robust defense mechanisms in medical AI systems, ensuring reliability and safety in real-world healthcare applications.
From Prototyping to Deployment: Human-Centered Design Practices in Responsible AI Innovation. Textual Intelligence Large Language Models and their Real World Applications, 2025
A federated learning approach to classify depression using audio dataset Applied Data Science and Smart Systems, 2024
Agriculture in Society 5.0 Meenakshi Aggarwal, Vikas Khullar, Nitin Goyal Artificial Intelligence and Society 5 0 Issues Opportunities and Challenges, 2024
A comprehensive review of federated learning: Methods, applications, and challenges in privacy-preserving collaborative model training Applied Data Science and Smart Systems, 2024
Efficient Skin Lesion based Classification System for Monkeypox Detection using VGG16 and Ensemble Learning Proceedings of the 17th Indiacom 2023 10th International Conference on Computing for Sustainable Global Development Indiacom 2023, 2023
Automatic Coal and Coal Gangue Image Recognition using Transfer Learning Preeti Sharma, Vikas Khullar, Preeti Gupta, Renu Popli, Isha Kansal 2023 IEEE International Conference on Research Methodologies in Knowledge Management Artificial Intelligence and Telecommunication Engineering Rmkmate 2023, 2023
Classifying Diverse Material Images based on Transfer Learning Isha Kansal, Vikas Khullar, Kinny Garg, Preeti Sharma, Renu Popli 2023 IEEE International Conference on Research Methodologies in Knowledge Management Artificial Intelligence and Telecommunication Engineering Rmkmate 2023, 2023
A Comprehensive Review of Techniques for Enhancing Lifetime of Wireless Sensor Network Smart and Power Grid Systems Design Challenges and Paradigms, 2022
Communication Jamming in Body Sensor Network: A Review Raj Gaurang Tiwari, Alok Misra, Ambuj Kumar Agarwal, Vikas Khullar Proceedings of the 2021 10th International Conference on System Modeling and Advancement in Research Trends Smart 2021, 2021
Transfer Learning Inspired Fish Species Classification Ambuj Kumar Agarwal, Raj Gaurang Tiwari, Vikas Khullar, Rajesh Kumar Kaushal Proceedings of the 8th International Conference on Signal Processing and Integrated Networks Spin 2021, 2021
Heart rate outlier detection for probable meltdown or tantrum state in autism spectrum disorder International Journal of Scientific and Technology Research, 2019
Perinatal hypoxia diagnostic system by using scalable machine learning algorithms CSE, CT Institute of Engineering Management, Technology, Jalandhar, India., Harmandeep Kaur*, Vikas Khullar, CSE, CT Institute of Engineering Management, Technology, Jalandhar, India., Harjit Pal Singh, ECE, CT Institute of Engineering Management, Technology, Jalandhar, India., Manju Bala, CSE, Khalsa College of Engineering, et al. International Journal of Innovative Technology and Exploring Engineering, 2019
Patient text feedback based optimized deep learned model to identify the impact of therapy CSE, CT Institute of Engineering, Management, Technology, Jalandhar, India., Jagriti*, Vikas Khullar, CSE, CT Institute of Engineering, Management, Technology, Jalandhar, India., Dr. Harjit pal singh, ECE., CT Institute of Engineering, Management, Technology, Jalandhar, India., Dr. Manju Bala, CSE, Khalsa College of Engineering, et al. International Journal of Innovative Technology and Exploring Engineering, 2019