A5100324492- Surgical Simulation and Training
- Medical Imaging and Analysis
- Augmented Reality Applications
- Soft Robotics and Applications
- Robotics and Sensor-Based Localization
- Robotic Path Planning Algorithms
- Spine and Intervertebral Disc Pathology
- Medical Image Segmentation Techniques
- Brain Tumor Detection and Classification
- Nanoplatforms for cancer theranostics
- Cancer-related molecular mechanisms research
- Anatomy and Medical Technology
- EEG and Brain-Computer Interfaces
- Genetic Mapping and Diversity in Plants and Animals
- AI-based Problem Solving and Planning
- RFID technology advancements
- Indoor and Outdoor Localization Technologies
- Formal Methods in Verification
- Robot Manipulation and Learning
- Advanced Vision and Imaging
- Musculoskeletal pain and rehabilitation
- Metal Extraction and Bioleaching
- Genetic and phenotypic traits in livestock
- Real-Time Systems Scheduling
- Reinforcement Learning in Robotics
Politecnico di Milano
2021-2025
Ministry of Agriculture and Rural Affairs
2023-2025
Sichuan Agricultural University
2023-2025
University of California, Berkeley
2025
University of Science and Technology of China
2019-2024
Sun Yat-sen University
2022-2024
Fifth Affiliated Hospital of Sun Yat-sen University
2022-2024
Kingmed Diagnostics
2022-2023
China Nonferrous Metal Mining (China)
2021-2022
Guangdong University of Technology
2022
Accurate tumor identification is essential in cancer management. Incomplete excision of tissue, however, negatively affects the prognosis patient. To accomplish radical radiotracers can be used that target tissue and detected using a gamma probe during surgery. Intraoperative fluorescence imaging could allow accurate real-time delineation. Herein, novel dual-modal platform base-catalyzed double addition thiols into propiolamide scaffold has been developed, allowing for highly efficient...
Modeling trajectory data is a building block for many smart-mobility initiatives. Existing approaches apply shallow models such as Markov chain and inverse reinforcement learning to model trajectories, which cannot capture the long-term dependencies. On other hand, deep Recurrent Neural Network (RNN) have demonstrated their strength of modeling variable length sequences. However, directly adopting RNN trajectories not appropriate because unique topological constraints faced by trajectories....
In the human–robot interaction, especially when hand contact appears directly on robot arm, dynamics of human arm presents an essential component in interaction and object manipulation. Modeling estimation show great potential for achieving more natural safer interaction. To enrich dexterity guarantee accuracy manipulation, mapping motor functionality muscle using biosignals becomes a popular topic. this article, novel algorithm was constructed deep learning to explore model between surface...
Robot-assisted surgery is rapidly developing in the medical field, and integration of augmented reality shows potential to improve operation performance surgeons by providing more visual information. In this paper, we proposed a markerless framework enhance safety avoiding intra-operative bleeding, which high risk caused collision between surgical instruments delicate blood vessels (arteries or veins). Advanced stereo reconstruction segmentation networks are compared find best combination...
Aim Accurate severity grading of lumbar spine disease by magnetic resonance images (MRIs) plays an important role in selecting appropriate treatment for the disease. However, interpreting these complex MRIs is a repetitive and time-consuming workload clinicians, especially radiologists. Here, we aim to develop multi-task classification model based on artificial intelligence automated disc herniation (LDH), central canal stenosis (LCCS) nerve roots compression (LNRC) at axial MRIs. Methods...
Abstract Background The severity assessment of lumbar disc herniation (LDH) on MR images is crucial for selecting suitable surgical candidates. However, the interpretation time‐consuming and requires repetitive work. This study aims to develop evaluate a deep learning‐based diagnostic model automated LDH detection classification axial T2‐weighted images. Methods A total 1115 patients were analyzed in this retrospective study; both development dataset (1015 patients, 15 249 images) an...
Surgical task automation in robotics can improve the outcomes, reduce quality-of-care variance among surgeons and relieve surgeons' fatigue. Reinforcement learning (RL) methods have shown considerable performance robot autonomous control complex environments. However, existing RL algorithms for surgical robots do not consider any safety requirements, which is unacceptable automating tasks. In this work, we propose an approach called Safe Experience Reshaping (SER) that be integrated into...
Mechanical vibration sensing/monitoring plays a critical role in today's industrial Internet of Things (IoT) applications. Existing solutions usually involve directly attaching sensors to the target objects, which is invasive and may affect operations delicate devices. Non-invasive approaches such as video laser methods have drawbacks that, former incurs poor performance low light conditions, while latter has difficulties monitor multiple objects simultaneously.In this work, we design...
Social genetic effects (SGEs) refer to how the genotypes of other individuals impact an individual’s phenotype within a population. These significantly influence feeding behavior and production performance in pigs, though their mechanisms are not well understood. This study examined two pig groups with extreme SGE values for residual feed intake (RFI), analyzing molecular involved using transcriptomics proteomics analysis liver ileum tissues. Pigs higher exhibited distinct patterns, spending...
4D medical image interpolation is essential for improving temporal resolution and diagnostic precision in clinical applications. Previous works ignore the problem of distribution shifts, resulting poor generalization under different distribution. A natural solution would be to adapt model a new test distribution, but this cannot done if input comes without ground truth label. In paper, we propose novel time training framework which uses self-supervision requiring any labels. Indeed, before...
Automatic vertebral osteophyte recognition in Digital Radiography is of great importance for the early prediction degenerative disease but still a challenge because tiny size and high inter-class similarity between normal vertebrae. Meanwhile, common sampling strategies applied Convolution Neural Network could cause detailed context loss. All these lead to an incorrect positioning predicament. In this paper, based on important pathological priors, we define set potential lesions each...
Automatic identification of Alzheimer's Disease (AD) through magnetic resonance imaging (MRI) data can effectively assist to doctors diagnose and treat Alzheimer's. Current methods improve the accuracy AD recognition, but they are insufficient address challenge small interclass large intraclass differences. Some studies attempt embed patch-level structure in neural networks which enhance pathologic details, enormous size time complexity render these unfavorable. Furthermore, several...
The generous application of robot-assisted minimally invasive surgery (RAMIS) promotes human-machine interaction (HMI). Identifying various behaviors doctors can enhance the RAMIS procedure for redundant robot. It bridges intelligent robot control and activity recognition strategies in operating room, including hand gestures human activities. In this paper, to identification a dynamic situation, we propose multimodal data fusion framework provide multiple information accuracy enhancement....
This study introduces an advanced human-robot interface designed to discern and execute manipulation tasks based solely on visual cues. The combines eye-tracking technology robotic manipulation, facilitating actions like grasping or pick-and-place tasks. We have developed a head-mounted device for tracking eye movements, allowing the system determine user's focus initiate sight-driven manipulation. Enhancing efficiency, incorporates transformer-based model, utilizing attention blocks feature...
Although the da Vinci surgical system enhances manipulation dexterity and restores 3D vision in robotic surgery, it requires surgeons to asynchronously control instruments endoscope, which hinders a smooth operation. Surgeons frequently position endoscope maintain good field of view during operation, potentially increasing time workload. In this paper, Human-Out-Of-The-Loop (HOOTL) navigation with assistance context awareness is proposed enhance autonomy. A comprehensive comparison study...
Advanced developments in the medical field have gradually increased public demand for surgical skill evaluation. However, this assessment always depends on direct observation of experienced surgeons, which is time-consuming and variable. The introduction robot-assisted surgery provides a new possibility evaluation paradigm. This paper aims at evaluating surgeon performance automatically with novel metrics based different data.Urologists ([Formula: see text]) from hospital were requested to...