Search for: All records

This study validates a fiber optics-based Surface Enhanced Raman Spectroscopy (SERS) sensor for detecting Salmonella in raw turkey samples. The sensor uses nanoantenna arrays on a side-polished optical fiber core with a fixed periodicity of 0.77 µm to maximize the SERS signal intensity. A 3D-printed chamber and microstructure optimize light reflection, improving sensitivity. The sensor detects Salmonella at 0.4-0.5 CFU/ml in 10 minutes, offering cost-effective, portable pathogen detection. 

This paper studies information theoretic secure aggregation in federated learning, which involves K distributed nodes and a central server. For security, the server can only recover aggregated updates of locally trained models, without any other information about the local users’ data being leaked. The secure aggregation process typically consists of two phases: the key sharing phase and the model aggregation phase. In previous research, a constraint on keys known as “uncoded groupwise keys” was introduced, and we adopt this constraint during the key sharing phase, where each set of S -users shares an independent key. During the model aggregation phase, each user transmits its encrypted model results to the server. To tolerate user dropouts in secure aggregation (i.e., some users may not respond), where up to K−U users may drop out and the identity of the surviving users is unpredictable in advance, at least two rounds of transmission are required in the model aggregation phase. In the first round, users send the masked models. Then, in the second round, based on the identity of the surviving users after the first round, these surviving users send additional messages that assist the server in decrypting the sum of the users’ trained models. Our goal is to minimize the number of transmissions in the two rounds. Additionally, we consider the potential impact of user collusion, where up to T users may collude with the server. This requires the transmissions to meet stricter security constraints, ensuring that the server cannot learn anything beyond the aggregated model updates, even if it colludes with any set of up to T users. For this more challenging problem, we propose schemes that ensure secure aggregation and achieve the capacity region when S∈{2}∪[K−U+1:K−T] . Experimental results conducted on Tencent Cloud also show that the proposed secure aggregation schemes improve the model aggregation time compared to the benchmark scheme. 

The geometry of multisegmented thermo-responsive gel robots was manipulated to break symmetry and support locomotion. 

Microsphere Photolithography (MPL) is a nanopatterning technique that utilizes a self-assembled monolayer of microspheres as an optical element to focus incident radiation inside a layer of photoresist. The microspheres produces a sub-diffraction limited photonic-jet on the opposite side of each microsphere from the illumination. When combined with pattern transfer techniques such as etching/lift-off, MPL provides a versatile, low-cost fabrication method for producing hexagonal close-packed metasurfaces. This article investigates the MPL process for creating refractive index (RI) sensors on the cleaved tips of optical fiber. The resonant wavelength of metal elements on the surface is dependent on the local dielectric environment and allows the refractive index of an analyte to be resolved spectrally. A numerical study of hole arrays defined in metal films shows that the waveguide mode provides good sensitivity to the analyte refractive index. This can be readily tuned by adjusting the MPL exposure and the simulation results guide the fabrication of a defect tolerant refractive index sensor on the tip of a fiber tip with a sensitivity of 613 nm/RIU. The conformal nature of the microsphere monolayer simplifies the fabrication process and provides a viable alternative to direct-write techniques such as Focused Ion Beam (FIB) milling 

Changes in structure and function of small muscular arteries play a major role in the pathophysiology of pulmonary hypertension, a burgeoning public health challenge. Improved anatomically mimetic in vitro models of these microvessels are urgently needed because nonhuman vessels and previous models do not accurately recapitulate the microenvironment and architecture of the human microvascular wall. Here, we describe parallel biofabrication of photopatterned self-rolled biomimetic pulmonary arterial microvessels of tunable size and infrastructure. These microvessels feature anatomically accurate layering and patterning of aligned human smooth muscle cells, extracellular matrix, and endothelial cells and exhibit notable increases in endothelial longevity and nitric oxide production. Computational image processing yielded high-resolution 3D perspectives of cells and proteins. Our studies provide a new paradigm for engineering multicellular tissues with precise 3D spatial positioning of multiple constituents in planar moieties, providing a biomimetic platform for investigation of microvascular pathobiology in human disease.