Abstract: An electrolytic polishing treatment method for a nickel-based alloy workpiece made by lamination manufacturing comprises the following steps. Step (A) comprises performing a sandblasting treatment on the nickel-based alloy workpiece, followed by ultrasonic oscillation of the sandblasted nickel-based alloy workpiece in an oxalic acid solution. Step (B) comprises placing the nickel-based alloy workpiece in an electrolyte solution containing methanol, sulfuric acid, and perchloric acid and performing electrolytic polishing on the nickel-based alloy workpiece at a constant voltage after step (A). The processes of oxalic acid activation and electrolytic polishing are used to avoid the problems of residual stress and processing directionality caused by conventional mechanical processing and make the surface properties of the entire workpiece uniform.

Abstract: A memory device and method of making the same are disclosed. The memory device includes transistor devices located in both a memory region and a logic region of the device. Transistor devices in the memory region include sidewall spacers having a first oxide layer over a side surface of a gate structure, a first nitride layer over the first oxide layer, a second oxide layer over the first nitride layer, and a second nitride layer over the second oxide layer. Transistor devices in the logic region include sidewall spacers having a first oxide layer over a side surface of a gate structure, a first nitride layer over the first oxide layer, and a second nitride layer over the first nitride layer.

Abstract: A terminal device includes a housing and a metal line that is disposed on an outer surface of the housing or is embedded in the housing, and the metal line is configured to receive or send an electromagnetic wave signal. The metal line acts as an antenna of the terminal device.

Abstract: An arcuate display device includes a plurality of display units each having has a plurality of pixels, a virtual axis, and a plurality of driving devices. Each pixel includes first, second, and third light-emitting elements respectively disposed at first, second, and third positions. The driving devices corresponding to the display units having the same minimum distance from the virtual axis have the same circuit layout design. The first, second, and third positions are sequentially arranged in a direction away from the virtual axis. Optical properties of the first light-emitting elements and the third light-emitting elements in at least a part of the pixels disposed at a first side of the virtual axis are respectively substantially the same as optical properties of the third light-emitting elements and the first light-emitting elements in at least a part of the pixels disposed at a second side of the virtual axis.

Abstract: An optical system affixed to an electronic apparatus is provided, including a first optical module, a second optical module, and a third optical module. The first optical module is configured to adjust the moving direction of a first light from a first moving direction to a second moving direction, wherein the first moving direction is not parallel to the second moving direction. The second optical module is configured to receive the first light moving in the second moving direction. The first light reaches the third optical module via the first optical module and the second optical module in sequence. The third optical module includes a first photoelectric converter configured to transform the first light into a first image signal.

Abstract: A driving device comprises a first complementary metal-oxygen-semiconductor circuit and a first comparator. The first complementary metal-oxygen-semiconductor circuit is configured for outputting a power signal or a pull-down signal according to the first input signal. The first comparator comprises a first non-inverting input terminal and a first inverting input terminal. The first non-inverting input terminal is coupled to the first complementary metal-oxygen-semiconductor circuit, and is configured to receive the power signal or the pull-down signal. The first inverting input terminal is configured for receiving a first reference signal, and the first comparator is configured to compare one of the power signal and the pull-down signal and the first reference signal to provide a first driving signal.

Abstract: The present disclosure provides a dressing configured to cover a wound. The dressing includes a substrate layer, a contact layer and a readable pattern layer. The substrate layer is light-permeable. The contact layer is located at a side of the substrate layer, and the contact layer is configured to contact the wound or skin around the wound. The readable pattern layer is disposed on the contact layer. The readable pattern layer includes a plurality of positioning marks and a plurality of colored cells. The positioning marks are configured to define an information area. The colored cells are located in the information areas, wherein at least part of the colored cells includes a biological indicator, and the positioning marks and the colored cells are visible through the substrate layer.

Abstract: A device structure according to the present disclosure includes a passivation layer, a first conductor plate layer disposed on the passivation layer, a second conductor plate layer disposed over the first conductor layer, a third conductor plate layer disposed over the second conductor layer, and a fourth conductor plate layer disposed over the third conductor layer. The second conductor plate layer encloses the first conductor plate layer and the fourth conductor plate layer encloses the third conductor plate layer. The device structure, when used in a back-end-of-line passive device, reduces leakage and breakdown due to corner discharge effect.

Abstract: A capacitive biosensor is provided. The capacitive biosensor includes: a transistor, an interconnect structure on the transistor, and a passivation layer on the interconnect structure. The interconnect structure includes a first metal structure on the transistor, a second metal structure on the first metal structure, and a third metal structure on the second metal structure. The third metal structure includes a first conductive layer, a second conductive layer, and a third conductive layer that are sequentially stacked. The passivation has an opening exposing a portion of the third metal structure. The capacitive biosensor further includes a sensing region on the interconnect structure. The sensing region includes a first sensing electrode and a second sensing electrode. The first sensing electrode is formed of the third conductive layer, and the second sensing electrode is disposed on the passivation layer.

Abstract: A method of reconstructing transcranial images using a dual-mode ultrasonic phased array includes steps of: controlling channels to emit energy toward an intracranial target point of a patient; respectively generating backscattered radiofrequency (RF) data by using the channels to receive backscattered energy reflected from the intracranial target; and reconstructing an acoustic distribution image based on those backscattered RF data in real-time. Compared with Pre-Treatment Ray Tracing Method, the present invention can display intracranial pressure distribution in real-time; compared with MR Thermometry, the present invention can be applied to low-energy applications without temperature change; and compared with Passive Cavitation Imaging, the present invention can stably present acoustic distribution images without relying on microbubbles.

Abstract: A method for enhancing kernel reparameterization of a non-linear machine learning model includes providing a predefined machine learning model, expanding a kernel of the predefined machine learning model with a non-linear network for convolution operation of the predefined machine learning model to generate the non-linear machine learning model, training the non-linear machine learning model, reparameterizing the non-linear network back to a kernel for convolution operation of the non-linear machine learning model to generate a reparameterized machine learning model, and deploying the reparameterized machine learning model to an edge device.

Abstract: In aspects of the disclosure, a method, a system, and a computer-readable medium are provided. The method of building a kernel reparameterization for replacing a convolution-wise operation kernel in training of a neural network comprises selecting one or more blocks from tensor blocks and operations; and connecting the selected one or more blocks with the selected operations to build the kernel reparameterization. The kernel reparameterization has a dimension same as that of the convolution-wise operation kernel.

Abstract: A reparameterization method for initializing a machine learning model includes initializing a prefix layer of a first low dimensional layer in the machine learning model and a postfix layer of the first low dimensional layer, inverting the prefix layer to generate an inverse prefix layer of the first low dimensional layer, inverting the postfix layer to generate an inverse postfix layer of the first low dimensional layer, combining the inverse prefix layer, the first low dimensional layer and the inverse postfix layer to form a high dimensional layer, generating parallel operation layers from the high dimensional layer, and assigning initial weights to the parallel operation layers.

Abstract: A method for removing branches from trained deep learning models is provided. The method includes steps (i)-(v). In step (i), a trained model is obtained. The trained model has a branch structure involving one or more original convolutional layers and a shortcut connection. In step (ii), the shortcut connection is removed from the branch structure. In step (iii), a reparameterization model is built by linearly expanding each of the original convolutional layers into a reparameterization block in the reparameterization model. In step (iv), parameters of the reparameterization blocks are optimized by training the reparameterization model. In step (v), each of the optimized reparameterization blocks is transformed into a reparameterized convolutional layer to form a branchless structure that replaces the branch structure in the trained model.

Abstract: A method for preparing a fiber material from a banana pseudostem, which includes subjecting the banana pseudostem to a pressing treatment to obtain a pressed banana pseudostem, subjecting the pressed banana pseudostem to a fermentation reaction with Saccharomyces cerevisiae to obtain a fermented culture, subjecting the fermented culture to a simultaneous saccharification and fermentation process with pectinase to obtain a fermented product, subjecting the fermented product to a solid-liquid separation treatment to obtain a solid fraction, and subjecting the solid fraction to an alkali treatment, a water washing treatment, a bleaching treatment, a softening treatment, and a drying treatment in sequence, so as to obtain the fiber material.

Abstract: An antenna module includes a first metal plate and a frame body. The frame body surrounds the first metal plate. The frame body includes a first antenna radiator, a second antenna radiator, a third antenna radiator, a first breakpoint and a second breakpoint. The first antenna radiator includes a first feeding end and excites a first frequency band. The second antenna radiator includes a second feeding end and excites a second frequency band. The third antenna radiator includes a third feeding end and excites a third frequency band. The first breakpoint is located between the first antenna radiator and the second antenna radiator. The second breakpoint is located between the second antenna radiator and the third antenna radiator. An electronic device including the above-mentioned antenna module is also provided.

Abstract: An interactive simulation system with a stereoscopic image and a method for operating the system are provided. The system includes a three-dimensional display and a control host. The stereoscopic image display is used to display a stereoscopic image. The system includes an interactive sensor that is used to sense gesture of a user or an action that the user manipulates a haptic device so as to produce sensing data. In the meantime, the system detects eye positions of the user. The changes of coordinates with respect to the gesture or the haptic device can be determined and referred to for forming a manipulation track. An interactive instruction can be determined. The stereoscopic image can be updated according to the interactive instruction and the eye positions of the user.

Abstract: A gas flow accelerator may include a body portion, and a tapered body portion including a first end integrally formed with the body portion. The gas flow accelerator may include an inlet port connected to the body portion and to receive a process gas to be removed from a semiconductor processing tool by a main pumping line. The semiconductor processing tool may include a chuck and a chuck vacuum line to apply a vacuum to the chuck to retain a semiconductor device. The tapered body portion may be configured to generate a rotational flow of the process gas to prevent buildup of processing byproduct on interior walls of the main pumping line. The gas flow accelerator may include an outlet port integrally formed with a second end of the tapered body portion. An end portion of the chuck vacuum line may be provided through the outlet port.

Abstract: A battery protection charging method and a charging system thereof are provided. The battery protection charging method includes obtaining an operation parameter of a battery. When a first protection condition is satisfied, a charging voltage of the battery is reduced to a first voltage. When a second protection condition is satisfied, the charging voltage of the battery is reduced to a second voltage. The second voltage is lower than the first voltage. The first (second) protection condition includes the cycle-life count is higher than a first (second) cycle-life threshold, the high voltage cumulative time is higher than a first (second) high voltage time threshold, and the high voltage-temperature cumulative time is higher than a first (second) high voltage-temperature time threshold.

Abstract: Semiconductor devices and methods of forming the same are provided. In some embodiments, a method includes receiving a workpiece having a redistribution layer disposed over and electrically coupled to an interconnect structure. In some embodiments, the method further includes patterning the redistribution layer to form a recess between and separating a first conductive feature and a second conductive feature of the redistribution layer, where corners of the first conductive feature and the second conductive feature are defined adjacent to and on either side of the recess. The method further includes depositing a first dielectric layer over the first conductive feature, the second conductive feature, and within the recess. The method further includes depositing a nitride layer over the first dielectric layer. In some examples, the method further includes removing portions of the nitride layer disposed over the corners of the first conductive feature and the second conductive feature.