This paper demonstrates a new method to porate single suspension living cell membrane using cavitation bubble-induced high speed jet flow. A microfluidic chip with an array of single cell trapping structures is design...
详细信息
This paper demonstrates a new method to porate single suspension living cell membrane using cavitation bubble-induced high speed jet flow. A microfluidic chip with an array of single cell trapping structures is designed and fabricated to trap sing cell and induce the asymmetric collapse of the cavitation bubble. Myeloma cells suspended in Trypan blue saline solution is tested and single Myeloma cells can be trapped in the trapping structures. The dynamic process is recorded using a high speed camera. This method has great potential in biomedical applications and easy to be integrated to other microfluidic system.
In this paper, an exact step-coupling theory is developed to describe the modes coupling behavior and lightwaves propagation for the power exchange between the waveguide modes in geometrical variation photonic crystal...
In this paper, an exact step-coupling theory is developed to describe the modes coupling behavior and lightwaves propagation for the power exchange between the waveguide modes in geometrical variation photonic crystal waveguide. The exact step-coupling theory provides a general description of the mode-coupling mechanism for light waveguide with geometrical variation with complete set of equations and solutions. The coupling equations of the exact step theory are derived and compared with the scattering matrix method, where simulation results show good agreement with an error of less than 2.2%. Subsequently, the coupling equations are applied to different case studies such as slab tapered waveguide and lossy “turn-on” waveguide. The transmission spectrum and field pattern distribution show that the lossy waveguide has a large radiation loss with an average transmission efficiency of less than 5%. The slab tapered waveguide can have more than 90% transmission efficiency with the convex curvature. The exact step-coupling theory can be applied to a vast range of geometrical variation photonic crystal based waveguides and it has quick and accurate convergence simulation results.
This paper presents a micro machined tunable optical filter using micro-electro-mechanical-system (MEMS) technology. The tunable filter is comprised of Fabry-Parot cavity consisting of an in-plane-movable high reflect...
详细信息
This paper presents a micro machined tunable optical filter using micro-electro-mechanical-system (MEMS) technology. The tunable filter is comprised of Fabry-Parot cavity consisting of an in-plane-movable high reflection mirror by bulk micromachining and single mode fiber. The length of the cavity is tuned by adjusting the driving voltage to the comb drive actuator who controls the position of the mirror. This micro optical filter is silicon on insulator (SOI) based so that the process is simple and tuning range is great. The optical fibers are easily aligned with the mirror on the substrate thanks to the fiber groove obtained by the high-resolution lithography and deep etching process. The finesse obtained experimentally is 40 and 2.1 and the insertion loss is 12.0 dB and 6.6 dB corresponding to 140 μm and 35 μm original cavity length.
Few-shot learning (FSL) aims to classify a novel object into a specific category under limited training samples. This is a challenging task since (1) the features expressed by pre-trained knowledge introduce perceived...
详细信息
Few-shot learning (FSL) aims to classify a novel object into a specific category under limited training samples. This is a challenging task since (1) the features expressed by pre-trained knowledge introduce perceived bias and then constrain the classification space, and (2) the use of general hallucination techniques based on global features fails to escape the limited classification space, resulting in suboptimal improvements. To solve these issues, this paper proposes an interventional feature generation (IFG) method. Specifically, we first use the relations of the categories or instances as interventional operations to implicitly constrain the feature representations (pre-trained knowledge) into different classification subsets. Then, we employ a parameter-free feature generation strategy to enrich each subset’s training samples of the support category. In other words, IFG provides a multi-subsets learning strategy to reduce the influence of perceived bias, enrich the diversity of generated features, and improve the robustness of the few-shot classifier. We apply our method to four benchmark datasets and observe state-of-the-art performance across all experiments. Specifically, compared to the baseline on the Mini-ImageNet dataset, our approach yields accuracy improvements of 6.03% and 3.46% for 1 and 5 support training samples, respectively. Furthermore, the proposed interventional feature generation technique can improve classifier performance in other FSL methods, demonstrating its versatility and potential for broader applications. The code is available at https://***/ShuoWangCS/IFG-FSL/.
暂无评论