Telesurgery is a new medical technology wherein the surgeon is situated remotely and operates through computers on surgically interactive robotic equipment connected through high speed communications networks. Optical fibers are necessary for this process as they have low latency and high capacity for realtime control of the process and exchange of data. Nevertheless, transmission is vulnerable to disruption by optical fiber faults which is disruptive to surgery safety and accuracy. This paper introduces TeleOps framework to enhance communication over optical fibers and fault management in telesurgery systems. Deep Learning (DL) models like Feedforward Neural Network (FFNN), 1D Convolutional Neural Network (1D-CNN), and Recurrent Neural Network (RNN) were implemented to detect and classify faults in Optical Time-Domain Reflectometer (OTDR) trace sequences. The RNN with Adam optimization achieved the highest detection accuracy. The proposed TeleOps framework also includes a blockchain-based smart contract to ensure transparent and encrypted tracking of faults, decentralized storage, and entity management in terms of surgical outcomes. This dual strategy provides reliable communication and efficient fault monitoring to minimize downtime of the remote surgical action. Thus the proposed TeleOps framework promises to enable safer and more effective solutions for remote healthcare by significantly enhancing the security and reliability of telesurgery systems.

The study develops an analysis framework built with deep learning techniques that extensively tests various architectures of modern Convolutional Neural Network (CNN) structures. The ResNet-18 model demonstrated the most successful implementation by reaching a training accuracy of 97.56% and validation accuracy of 86.92% at the same time. The corresponding training and validation losses amounted to 0.0754 and 0.6071, respectively. The LIME and Grad-CAM techniques of the XAI, and Occlusion were used to enhance the transparency of the model while providing components to better understand model decisions. The combination of accurate CNN models with interpretability tools produces successful explainable and robust classification in real-world application scenarios.

The emergence of smart healthcare has transformed the delivery of medical services. It increased the reliability issues related to telesurgery system. Robotic technologies integrate to achieve surgical precision in countless medical fields. But, this increased connectivity has raised concerns about vulnerability to denial of service (DoS) attacks that can disrupt operations and affect users' safety. In this paper, we propose an explainable machine learning (ML) framework for detecting DoS attacks in telesurgery environments. Our framework utilizes several ML algorithms, including K-Nearest Neighbors (KNN), Decision Tree (DT), XGBoost (XGB), and Logistic Regression (LR), to analyze network traffic and distinguish between legitimate and malicious signals. Here, XGB achieves the highest accuracy at 0.8967, making it a good choicefor this application. We also used the concepts of Explainable AI (XAI) such as SHAP and LIME to enhance decision-making process of our models. We aim to build trust among healthcare providers and ensures the safe conduct of robotic telesurgery operations and smart healthcare solutions by improving the interpretation of research and providing information on performance metrics such as accuracy, precision, recall, F1 scores, and ROC curves.

The growing use of smart technologies in homes has changed the way we manage power consumption. In this paper, CHEFS an innovative framework is proposed for smart home power consumption analysis, using Explainable AI (XAI) to improve clarity and accuracy with prediction and continuous decision-making by edge computing. Devices like the Internet of Things (IoT) and smart meters are used in smart homes to observe energy usage precisely. Such devices are also used for cutting energy costs and reducing environmental impacts. They often lack clarity in decision-making which makes it difficult, whereas the traditional machine learning models offer accurate predictions. To show this, we use XAI methods like SHAP and LIME to get a better understanding of the power usage patterns and forecasts. Edge computing reduces delays and bandwidth usage while also enhancing response time. This approach enhances clarity and dependability in Artificial Intelligence (AI) models providing reliable insights into household power consumption. A comprehensive evaluation helps create a more transparent and effective system for creating smarter energy solutions.

Telesurgery is transforming healthcare by enabling surgeons to perform operations remotely through robotic systems connected to high-speed networks. The reliability and safety of these procedures depend on seamless communication, often secured using Proof-of-Stake (PoS) blockchain technology to ensure data integrity and validate transactions. However, malicious nodes within PoS blockchain networks pose significant risks by introducing delays, invalidating legitimate transactions, or colluding to compromise the system. This paper proposes a Deep Learning (DL) based framework to detect malicious nodes in PoS-based blockchain applications, ensuring secure and reliable operations. Using a dataset of node activity, DL models—LSTM, 1D-CNN, and FFNN—were trained with optimizers including Adam, Nadam, and RMSprop. Among these, the LSTM model with RMSprop achieved the highest detection accuracy of 87.37%. The framework enhances security by enabling real-time malicious node detection and communication monitoring, addressing key challenges in blockchain integrity and operational precision, ultimately ensuring the security and reliability of blockchain-integrated telesurgical systems.