11 April 2024 in IEEE Transactions on Medical Imaging
Adapting a medical image segmentation model to a new domain is important for improving its cross-domain transferability, and due to the expensive annotation process, Unsupervised Domain Adaptation (UDA) is appealing where only unlabeled images are needed for the adaptation. Existing UDA methods are mainly based on image or feature alignment with adversarial training for regularization, and they are limited by insufficient supervision in the target domain. In this paper, we propose an enhanced Filtered Pseudo Label (FPL+)-based UDA method for 3D medical image segmentation. It first uses cross-domain data augmentation to translate labeled images in the source domain to a dual-domain training set consisting of a pseudo source-domain set and a pseudo target-domain set. To leverage the dual-domain augmented images to train a pseudo label generator, domain-specific batch normalization layers are used to deal with the domain shift while learn the domain-invariant structure features, generating high-quality pseudo labels for target-domain images. We then combine labeled source-domain images and target-domain images with pseudo labels to train a final segmentor, where image-level weighting based on uncertainty estimation and pixel-level weighting based on dual-domain consensus are proposed to mitigate the adverse effect of noisy pseudo labels. Experiments on three public multi-modal datasets for Vestibular Schwannoma, brain tumor and whole heart segmentation show that our method surpassed ten state-of-the-art UDA methods, and it even achieved better results than fully supervised learning in the target domain in some cases.
1 January 2020 in IEEE Transactions on Industrial Electronics
Micro-displacement detection plays a vital role in the position diagnosis and accurate calibration of MEMS actuators. An optical micro-displacement detector proposed in this paper integrated a vertical cavity surface emitting laser (VCSEL) and nine photodiodes (PDs) to detect the sub-micrometer translation and sub-degree rotation of the piezoelectric micro vibratory stage under the limited space. The highly accurate detection was mainly realized by PDs design and quality improvement. Nine PDs layout with high resolution optical sensors and corresponding calculation guaranteed the detector to precisely acquire out-of-plane displacement and two axis rotation of vibratory platform. And the introduction of the light absorbing layer and the light shielding layer further improved the sensing quality to reach a higher measurement resolution. The micro stage was fabricated by MEMS technology and the experiments of measurement system indicated that the out-of-plane displacement detection accuracy was 150nm in the range of 500μm and the angle detection accuracy reached to 0.1° in the range of 2°.
22 March 2021 in IEEE/ASME Transactions on Mechatronics
The ultrasonic motors (USM) based on microelectromechanical systems (MEMS) technology utilize the inverse piezoelectric effects to drive the rotor through the friction force to realize the torque output. Therefore, the generation of a stable traveling wave within the stator is a key to USM performance, which heavily depends on the quality factor ( Q ) and the mode match level of a stator resonator. The current ring stators of micromotor exhibited an apparent mode mismatch and a relatively low Q value due to the displacement constraints and anchor loss of supporting beams. This article, through the modal analysis and energy dissipation evaluation of stators, proposes an elaborate design to improve the mode match and Q value by positioning the supporting beams at the saddle points of the B 13 modal shape and introducing a buffering annular frame between beams and substrate. The designed stator has the natural frequency of 101.63 and 100.90 kHz and Q value of 89.5 obtained from the finite element method. And the prototype experiment shows the natural frequencies of two modes are 100.76 and 101.5 kHz, respectively, the maximum deflection of the stator under driving voltage of 0.5 and 1 V are 3.96 and 4.43 μ m, respectively, and the Q value is 84.58. Both the frequency and rotation tests indicate the proposed motor displays, a much higher Q , and extremely better mode match than current ones.
8 August 2022 in IEEE Journal of Oceanic Engineering
Autonomous underwater vehicles (AUVs) have been widely implemented to explore marine resources or perform marine missions; meanwhile, complex mission planning has been expected to enhance the intelligence level, improve energy efficiency, and expand AUVs' application areas. The mission planning model needs to consider energy consumption, mission quality, and mission quantity of long-term working in the marine environment. In this article, one unified and robust model was proposed based on the above requirements in the complex marine environment to improve the automaticity of AUV. However, solving this mission planning model is a nontrivial problem, especially when high robustness, high efficiency, fast response, and near-optimal results are required. Therefore, we propose a novel hyperheuristic algorithm based on evolutionary strategy (ES-HH). The proposed ES-HH combines a metaheuristic framework with a selection function to evaluate the performance of low-level heuristic operators online. This evolutionary strategy endows the hyperheuristic algorithm with the online learning feature, giving the ES-HH algorithm better computing efficiency, robustness, and near-optimal results. The experiment results show that the ES-HH algorithm can achieve better convergence and higher robustness than other algorithms, such as ant colony optimization and biogeography-based optimization algorithms. Compared with ACO and BBO algorithms, the proposed ES-HH algorithm can improve the mission completion rate by 3.42% and reduce the average energy consumption of a single task by 8.02%.