Medical image segmentation is a critical application of computer vision in the analysis of medical images. Its primary objective is to isolate regions of interest in medical images from the background, thereby assisting clinicians in accurately identifying lesions, their sizes, locations, and their relationships with surrounding tissues. However, compared to natural images, medical images present unique challenges, such as low resolution, poor contrast, inconsistency, and scattered target regions. Furthermore, the accuracy and stability of segmentation results are subject to more stringent requirements. In recent years, with the widespread application of Convolutional Neural Networks (CNNs) in computer vision, deep learning-based methods for medical image segmentation have become a focal point of research. This paper categorizes, reviews, and summarizes the current representative methods and research status in the field of medical image segmentation. A comparative analysis of relevant experiments is presented, along with an introduction to commonly used public datasets, performance evaluation metrics, and loss functions in medical image segmentation. Finally, potential future research directions and development trends in this field are predicted and analyzed.
17 September 2024 in Biomimetic Intelligence and Robotics
In recent years, deep learning techniques for processing 3D point cloud data have seen significant advancements, given their unique ability to extract relevant features and handle unstructured data. These techniques find wide-ranging applications in fields like robotics, autonomous vehicles, and various other computer-vision applications. This paper reviews the recent literature on key tasks, including 3D object classification, tracking, pose estimation, segmentation, and point cloud completion. The review discusses the historical development of these methods, explores different model architectures, learning algorithms, and training datasets, and provides a comprehensive summary of the state-of-the-art in this domain. The paper presents a critical evaluation of the current limitations and challenges in the field, and identifies potential areas for future research. Furthermore, the emergence of transformative methodologies like PoinTr and SnowflakeNet is examined, highlighting their contributions and potential impact on the field. The potential cross-disciplinary applications of these techniques are also discussed, underscoring the broad scope and impact of these developments. This review fills a knowledge gap by offering a focused and comprehensive synthesis of recent research on deep learning techniques for 3D point cloud data processing, thereby serving as a useful resource for both novice and experienced researchers in the field.
With the continuous development of technology, blockchain has been widely used in various fields by virtue of its decentralization, data integrity, traceability, and anonymity. However, blockchain still faces many challenges, such as scalability and security issues. Artificial intelligence, with its powerful data processing capability, pattern recognition ability, and adaptive optimization algorithms, can improve the transaction processing efficiency of blockchain, enhance the security mechanism, and optimize the privacy protection strategy, thus effectively alleviating the limitations of blockchain in terms of scalability and security. Most of the existing related reviews explore the application of AI in blockchain as a whole but lack in-depth classification and discussion on how AI can empower the core aspects of blockchain. This paper explores the application of artificial intelligence technologies in addressing core challenges of blockchain systems, specifically in terms of scalability, security, and privacy protection. Instead of claiming a deep theoretical integration, we focus on how AI methods, such as machine learning and deep learning, have been effectively adopted to optimize blockchain consensus algorithms, improve smart contract vulnerability detection, and enhance privacy-preserving mechanisms like federated learning and differential privacy. Through comprehensive classification and discussion, this paper provides a structured overview of the current research landscape and identifies potential directions for further technical collaboration between AI and blockchain technologies.
13 January 2025 in IEEE Robotics and Automation Letters
Haptic feedback systems play a critical role in enriching the user experience in human-robot interaction. However, existing devices designed for evoking haptic sensations often face limitations owing to their low degree of freedom of deformation. In this study, we introduce the Funabot-Sleeve, a haptic device based on McKibben artificial muscles, and investigate its potential to evoke a range of haptic sensations using both steady-state and transient air pressure patterns. Our investigation examines the influence of these patterns on evoking distinct haptic sensations and identifies four specific sensations that can be evoked: Embraced, Pinched, a combination of Embraced and Pressed, and Twisted sensations. Across all participants, the evoked sensations showed positive correlations, with most correlations exceeding a value of 0.4, indicating a high degree of agreement in the sensations felt by the subjects. Our research lays the groundwork for the design of fabric actuators, capable of replicating specific stimuli and skin surface effects, thereby enabling a more sophisticated and personalized haptic feedback experience.