Ruoyi Hao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Ruoyi Hao has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
- Hongliang Ren
The Chinese University of Hong Kong
5 shared publications - Jinyu Liu
Hubei Key Laboratory of Modern Manufacturing Quality Engineering, Hubei University of Technology, Nanhu Avenue 28, Wuhan 430068, China
2 shared publications Yang Zhang
2 shared publicationsYang Zhou
2 shared publicationsCatherine Po Ling Chan
Department of Otorhinolaryngology, Head and Neck Surgery, The Chinese University of Hong Kong, Hong Kong SAR, China
1 shared publication- Changsheng Li
School of Mechatronical Engineering Beijing Institute of Technology Beijing China
1 shared publication
Latest Publications
- Open Access
- 1 March 2026
- Published by KEAI Publishing in Biomimetic Intelligence and Robotics
Real-time glottis detection framework via spatial-decoupled feature learning for nasal transnasal intubation
- 1 March 2026
- Published by KEAI Publishing in Biomimetic Intelligence and Robotics
Open AccessReal-time glottis detection framework via spatial-decoupled feature learning for nasal transnasal intubation
- Open Access
- 13 February 2026
- Published by MDPI AG in Bioengineering
- Vol. 13 (2) , 217
Integrating Dynamic Representation and Multi-Priors for Transnasal Intubation via Visual Foundation Model
Accurate and real-time glottis localization is critical for ensuring intraoperative oxygenation and patient safety during nasotracheal intubation. However, representative foundation models exemplified by the Segment Anything Model exhibit notable limitations in medical applications, stemming from their rigid attention mechanisms, feature space misalignment, and insufficient generalization to complex glottal anatomies. To address these challenges, we propose Glottis-SAM, a lightweight and task-adaptive segmentation framework that integrates dynamic representation learning with multi-prior contextual modeling. Specifically, we introduce a hierarchical low-rank adaptation strategy that enables efficient fine-tuning of visual foundation models by preserving geometric priors while significantly reducing computational overhead. To further enhance semantic fusion and generalization, we design a feature aggregation module with dual-path dynamic feature pyramids, which enables complementary optimization from local textures to global semantic structures under varying anatomical conditions. Extensive experiments on three diverse datasets demonstrate that Glottis-SAM achieves state-of-the-art segmentation accuracy with 72.6% mDice, a compact 55.2 MB model size, and 44.3 FPS inference speed on clinical data. These results highlight the model’s robustness, efficiency, and potential for deployment in visual guidance systems for nasotracheal intubation.
- 13 February 2026
- Published by MDPI AG in Bioengineering
- Vol. 13 (2) , 217
Open AccessIntegrating Dynamic Representation and Multi-Priors for Transnasal Intubation via Visual Foundation Model
Accurate and real-time glottis localization is critical for ensuring intraoperative oxygenation and patient safety during nasotracheal intubation. However, representative foundation models exemplified by the Segment Anything Model exhibit notable limitations in medical applications, stemming from their rigid attention mechanisms, feature space misalignment, and insufficient generalization to complex glottal anatomies. To address these challenges, we propose Glottis-SAM, a lightweight and task-adaptive segmentation framework that integrates dynamic representation learning with multi-prior contextual modeling. Specifically, we introduce a hierarchical low-rank adaptation strategy that enables efficient fine-tuning of visual foundation models by preserving geometric priors while significantly reducing computational overhead. To further enhance semantic fusion and generalization, we design a feature aggregation module with dual-path dynamic feature pyramids, which enables complementary optimization from local textures to global semantic structures under varying anatomical conditions. Extensive experiments on three diverse datasets demonstrate that Glottis-SAM achieves state-of-the-art segmentation accuracy with 72.6% mDice, a compact 55.2 MB model size, and 44.3 FPS inference speed on clinical data. These results highlight the model’s robustness, efficiency, and potential for deployment in visual guidance systems for nasotracheal intubation.
- 5 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 73 (4) , 5758-5770
Asymmetric Cosine Jerk Trajectory Optimization for Robust Vibration Suppression
Rapid and precise movements of industrial equipment necessitate reference trajectory design to mitigate motion-induced vibrations, which becomes more challenging in the presence of system uncertainties. This article develops a novel trajectory planning method based on an improved cosine jerk model to robustly suppress residual vibration under actuator limits. Incorporating tunable asymmetry into the model allows trajectories to start faster and shorten motion duration. By applying a linear transformation in the complex domain, the vibration suppression capability is extended to accommodate both undamped and damped systems. The profile parameters are modulated to align the intrinsic trajectory zeros with the system poles, attenuating spectral components near the critical modal frequency that excite resonance. Unlike much current work that provides only partial solutions or requires numerical optimization, a complete analytical solution is derived for all possible trajectory types using a designed coefficient that manages the acceleration ratio. The effectiveness and robustness of the method are validated experimentally across different systems. The results show that the proposed method can achieve better vibration mitigation performance under modeling errors than existing benchmark methods.
- 5 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 73 (4) , 5758-5770
Asymmetric Cosine Jerk Trajectory Optimization for Robust Vibration Suppression
Rapid and precise movements of industrial equipment necessitate reference trajectory design to mitigate motion-induced vibrations, which becomes more challenging in the presence of system uncertainties. This article develops a novel trajectory planning method based on an improved cosine jerk model to robustly suppress residual vibration under actuator limits. Incorporating tunable asymmetry into the model allows trajectories to start faster and shorten motion duration. By applying a linear transformation in the complex domain, the vibration suppression capability is extended to accommodate both undamped and damped systems. The profile parameters are modulated to align the intrinsic trajectory zeros with the system poles, attenuating spectral components near the critical modal frequency that excite resonance. Unlike much current work that provides only partial solutions or requires numerical optimization, a complete analytical solution is derived for all possible trajectory types using a designed coefficient that manages the acceleration ratio. The effectiveness and robustness of the method are validated experimentally across different systems. The results show that the proposed method can achieve better vibration mitigation performance under modeling errors than existing benchmark methods.
- Open Access
- 3 December 2025
- Published by Springer Nature in Scientific Data
- Vol. 12 (1) , 1907
Upper Airway Anatomical Landmark Dataset for Automated Bronchoscopy and Intubation
Bronchoscopy and intubation play crucial roles in respiratory disease diagnosis and treatment, yet the automation of their initial insertion phase remains limited. Advanced image analysis presents a viable solution to this challenge. However, insufficient comprehensive, publicly available datasets for training such models have hindered progress. We present a novel Upper Airway Anatomical Landmark (UAAL) Dataset, which annotates multiple anatomical landmark classes visualized through a bronchoscope, including the nose, nostril, channel, glottis, glottic aperture, vocal fold, and trachea, encompassing the entire upper respiratory tract from the nasal cavity to the trachea. It includes 3,814 clinical images from 82 patients with 10,330 annotations (4,910 instance segmentation masks and 5,420 bounding boxes) across 8 classes and 2,746 supplementary phantom images with 4,526 annotations (2,795 instance segmentation masks and 1,551 bounding boxes) across 9 classes. With its key contributions of diverse anatomical coverage, clinical data, supplementary phantom data, and public accessibility, this dataset will contribute to bronchoscopy and intubation automation systems, facilitating their transition from laboratory to clinical applications.
- 3 December 2025
- Published by Springer Nature in Scientific Data
- Vol. 12 (1) , 1907
Open AccessUpper Airway Anatomical Landmark Dataset for Automated Bronchoscopy and Intubation
Bronchoscopy and intubation play crucial roles in respiratory disease diagnosis and treatment, yet the automation of their initial insertion phase remains limited. Advanced image analysis presents a viable solution to this challenge. However, insufficient comprehensive, publicly available datasets for training such models have hindered progress. We present a novel Upper Airway Anatomical Landmark (UAAL) Dataset, which annotates multiple anatomical landmark classes visualized through a bronchoscope, including the nose, nostril, channel, glottis, glottic aperture, vocal fold, and trachea, encompassing the entire upper respiratory tract from the nasal cavity to the trachea. It includes 3,814 clinical images from 82 patients with 10,330 annotations (4,910 instance segmentation masks and 5,420 bounding boxes) across 8 classes and 2,746 supplementary phantom images with 4,526 annotations (2,795 instance segmentation masks and 1,551 bounding boxes) across 9 classes. With its key contributions of diverse anatomical coverage, clinical data, supplementary phantom data, and public accessibility, this dataset will contribute to bronchoscopy and intubation automation systems, facilitating their transition from laboratory to clinical applications.
- 3 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 71 (8) , 9328-9338
Modeling and Compensation of Stiffness-Dependent Hysteresis for Stiffness-Tunable Tendon-Sheath Mechanism in Flexible Endoscopic Robots
Robot-assisted gastrointestinal endoscopic surgery requires flexible manipulators to possess a compact dimension and stiffness tuning capability. Current stiffness-tunable miniature manipulators (STMM) using tendon-sheath mechanism (TSM) experience the problem of stiffness-influenced hysteresis. To address this, we propose the first stiffness-dependent M odified G eneralized P randtl– I shlinski (PI) model and a specific compensation strategy. First, we analyzed the stiffness tuning mechanism and extracted the stiffness parameter. Based on this, the analytical hysteresis model and its inverse function were built and verified in simulations and experiments, which increases the fitting accuracy of the hysteresis. Then, with the assistance of the compensation strategy, the real-time input–output relationship of the STMM's stiffness-tunable joints can be approximately linear. The average errors of trajectory tracking achieve a significant improvement of over 85%. This work provides a new method to model and compensate for the dynamic hysteresis in flexible endoscopic robots with the stiffness-tunable TSM.
- 3 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 71 (8) , 9328-9338
Modeling and Compensation of Stiffness-Dependent Hysteresis for Stiffness-Tunable Tendon-Sheath Mechanism in Flexible Endoscopic Robots
Robot-assisted gastrointestinal endoscopic surgery requires flexible manipulators to possess a compact dimension and stiffness tuning capability. Current stiffness-tunable miniature manipulators (STMM) using tendon-sheath mechanism (TSM) experience the problem of stiffness-influenced hysteresis. To address this, we propose the first stiffness-dependent M odified G eneralized P randtl– I shlinski (PI) model and a specific compensation strategy. First, we analyzed the stiffness tuning mechanism and extracted the stiffness parameter. Based on this, the analytical hysteresis model and its inverse function were built and verified in simulations and experiments, which increases the fitting accuracy of the hysteresis. Then, with the assistance of the compensation strategy, the real-time input–output relationship of the STMM's stiffness-tunable joints can be approximately linear. The average errors of trajectory tracking achieve a significant improvement of over 85%. This work provides a new method to model and compensate for the dynamic hysteresis in flexible endoscopic robots with the stiffness-tunable TSM.