Many artificial intelligence (AI) edge devices use nonvolatile memory (NVM) to store the weights for the neural network (trained off-line on an AI server), and require low-energy and fast I/O ...accesses. The deep neural networks (DNN) used by AI processors 1,2 commonly require p-layers of a convolutional neural network (CNN) and q-layers of a fully-connected network (FCN). Current DNN processors that use a conventional (von-Neumann) memory structure are limited by high access latencies, I/O energy consumption, and hardware costs. Large working data sets result in heavy accesses across the memory hierarchy, moreover large amounts of intermediate data are also generated due to the large number of multiply-and-accumulate (MAC) operations for both CNN and FCN. Even when binary-based DNN 3 are used, the required CNN and FCN operations result in a major memory I/O bottleneck for AI edge devices.
In plants, the key enzyme in ethylene biosynthesis is 1-aminocyclopropane-1 carboxylic acid (ACC) synthase (ACS), which catalyzes
-adenosyl-L-methionine (SAM) to ACC, the precursor of ethylene. ...Ethylene binds to its receptors, such as ethylene response 1 (ETR1), to switch on ethylene signal transduction. To understand the function of
and
in orchids,
(
) and
(
) from
Gower Ramsey were functionally analyzed in
. 35S::
caused late flowering and anther indehiscence phenotypes due to its effect on GA-DELLA signaling pathways. 35S::
repressed GA biosynthesis genes (
,
, and
), which caused the upregulation of DELLA
(
),
(
), and
expression. The increase in DELLAs not only suppressed
(
) expression and caused late flowering but also repressed the jasmonic acid (JA) biosynthesis gene
and caused anther indehiscence by downregulating the endothecium-thickening-related genes
,
, and
. The ectopic expression of an
dominant-negative mutation (
) caused both ethylene and JA insensitivity in
. 35S::
delayed flower/leaf senescence by suppressing downstream genes in ethylene signaling, including
and
, and in JA signaling, including
and
. JA signaling repression also resulted in indehiscent anthers
the downregulation of
,
,
, and
. These results not only provide new insight into the functions of
and
orthologs but also uncover their functional interactions with other hormone signaling pathways, such as GA-DELLA and JA, in plants.
As the growing demand on artificial intelligence (AI) Internet-of-Things (IoT) devices, smart vision sensors with energy-efficient computing capability are required. This article presents a low-power ...and low-voltage dual mode 0.5-V computational CMOS image sensor (C 2 IS) with array-parallel computing capability for feature extraction using convolution. In the feature extraction mode, by applying the pulsewidth modulation (PWM) pixel and switch-current integration (SCI) circuit, the in-sensor eight-directional matrix-parallel multiply-accumulate (MAC) operation is realized. Furthermore, the analog-domain convolution-on-readout (COR) operation, the programmable <inline-formula> <tex-math notation="LaTeX">3\times3 </tex-math></inline-formula> kernel with ±3-bit weights, and the tunable-resolution column-parallel analog-to-digital converter (ADC) (1-8 bit) are implemented to achieve the real-time feature extraction without using additional memory and sacrificing frame rate. In the image capturing mode, the sensor provides the linear-response 8-bit raw image data. The C 2 IS prototype has been fabricated in the TSMC 0.18-<inline-formula> <tex-math notation="LaTeX">\mu \text{m} </tex-math></inline-formula> standard process technology and verified to demonstrate the raw and feature images at 480 frames/s with a power consumption of 77/<inline-formula> <tex-math notation="LaTeX">117~\mu \text{W} </tex-math></inline-formula> and the resultant FoM of 9.8/14.8 pJ/pixel/frame, respectively. The prototype sensor is used as a real-time edge feature detection frond-end camera and accompanied with a simplified convolutional neural network (CNN) architecture to demonstrate the hand gesture recognition. The prototype system achieves more than 95% validation accuracy.
Repairing damaged articular cartilage is challenging due to the limited regenerative capacity of hyaline cartilage. In this study, we fabricated a bilayered poly (lactic-co-glycolic acid) (PLGA) ...scaffold with small (200⁻300 μm) and large (200⁻500 μm) pores by salt leaching to stimulate chondrocyte differentiation, cartilage formation, and endochondral ossification. The scaffold surface was treated with tyramine to promote scaffold integration into native tissue. Porcine chondrocytes retained a round shape during differentiation when grown on the small pore size scaffold, and had a fibroblast-like morphology during transdifferentiation in the large pore size scaffold after five days of culture. Tyramine-treated scaffolds with mixed pore sizes seeded with chondrocytes were pressed into three-mm porcine osteochondral defects; tyramine treatment enhanced the adhesion of the small pore size scaffold to osteochondral tissue and increased glycosaminoglycan and collagen type II (Col II) contents, while reducing collagen type X (Col X) production in the cartilage layer. Col X content was higher for scaffolds with a large pore size, which was accompanied by the enhanced generation of subchondral bone. Thus, chondrocytes seeded in tyramine-treated bilayered scaffolds with small and large pores in the upper and lower parts, respectively, can promote osteochondral regeneration and integration for articular cartilage repair.
Osteochondral lesions in weight-bearing joints often fail to heal due to their poor intrinsic repair capacity and may lead to severe complications such as osteoarthritis. This study investigated the ...chondrogenesis, osteogenesis, and regenerative potential of an injectable, thermoresponsive chitosan-graft-poly(N-isopropylacrylamide) (CS-g–PNIPAAm, or CSPN) hydrogel combined with infrapatellar fat pad-derived mesenchymal stem cell (IFPSC) and platelet-rich plasma (PRP) with regard to osteochondral regeneration. CSPN hydrogels were prepared and characterized using several procedures. Optimal CSPN hydrogel (CSPN hydrogel with 67% of PNIPAAm, or CSPN05) was examined following a biochemical assay and animal implantation. PRP incorporated CSPN05 hydrogel, CSPN05-PRP, was fabricated. We found that rabbit IFPSCs (rIFPSCs) cultured in the CSPN05-PRP hydrogel (rIFPSC–CSPN05–PRP) enhanced the expression of chondrogenic and osteogenic genes after 1, 3, 5, and 7 days of culture and significantly increased sulfated glycosaminoglycan content, alkaline phosphatase activity, and mineralization after 5 and 7 days of culture compared with PN and CSPN05. In addition, PRP induced the expression of specific signaling molecules for chondrogenesis and osteogenesis in its signaling pathway. A preliminary in vivo study showed that rIFPSC-CSPN05–PRP could effectively regenerate the osteochondral defect at 4 weeks. Thus, the combination of IFPSC, the CSPN hydrogel, and PRP can synergistically enhance chondrogenic and osteogenic gene expression, cartilage- and bone-specific extracellular matrix production, hyaline cartilage regeneration, and mineralized osseous tissue formation. CSPN scaffolds with PRP show potential as biomaterial by providing mechanical support for cell growth and inducing stem cells to differentiate into osteochondral lineages for application in regenerative medicine.
Display omitted
Silver nanoplates (AgP) were prepared and used in a colorimetric method for the evaluation of Xanthine (Xan) in blood plasma and fish meat. The detection mechanism for Xan was observed to occur via ...etching of AgP particles/aggregation/fusion steps, resulting in a color change from blue to grey. First, the basic Xan solution is adsorbed through partial substitution of capping molecules around the AgP with Xan, and then intermolecular hydrogen bonds form between AgP and AgP. Subsequently, the titrant Xan solution further etches the AgP and finally fuses particles together. Owing to the step by step mechanism, the response range towards Xan has two linear regression ranges: 0.15-0.60 μM and 0.61-3.00 μM, respectively. The detection limit in the range of 0.15-0.60 μM is 0.011 μM (S/N = 3). AgP exhibits good selectivity for Xan over other potential interferents such as amino acids and blood proteins. AgP achieves rapid detection of Xan and can be applied to the satisfactory determination of Xan in blood plasma and fish meat. This colorimetric sensor is easy to use, cost effective, fast, selective and user friendly.
A CMOS time-of-flight (TOF) image sensor with in-pixel background light (BGL) cancellation and demodulation lock-in pixel is developed for both indoor and outdoor depth imaging applications. The ...proposed polarity switching integration technique using p+/n-well diode realizes the in-pixel BGL cancellation up to 180 klx. In addition, the developed phase-shift readout (PSR) effectively reduces the inherent correlated noise and column fixed-pattern noise (FPN) as well with sensitivity improvement. A prototyped pulse-modulation-based TOF sensing chip with a <inline-formula> <tex-math notation="LaTeX">64 \times 64 </tex-math></inline-formula> pixel array, <inline-formula> <tex-math notation="LaTeX">20\mu \text {m} \times 20\mu \text{m} </tex-math></inline-formula> pixel pitch, and 33% fill factor has been fabricated in TSMC standard 0.18-<inline-formula> <tex-math notation="LaTeX">\mu \text{m} </tex-math></inline-formula> CMOS process and verified. The achieved depth sensing capability ranges from 0.75 to 7.5 m with a linearity error below 1.1%, and the measured relative precision is 4.2% at a 7.5-m target distance.
Organometal halide perovskites are attracting a great deal of attention because of their long carrier diffusion lengths, wide wavelength tunability, and narrow-band emission. However, the toxicity of ...lead has caused considerable environmental and health concerns. In this work, lead-free cesium tin halide nanocrystals are synthesized and investigated. CsSnBr
3
and CsSnI
3
nanocrystals, 25 and 7 nm in size, are synthesized by a facile hot injection method. Absorption spectroscopy, photoluminescence spectroscopy, and X-ray diffraction were used to understand their structural and optical properties. CsSnBr
3
and CsSnI
3
nanocrystals show emission peaks at 683 and 938 nm, respectively. These nanocrystals show shelf stability for a few months. Temperature-dependent photoluminescence is utilized to know more about fundamental physical parameters, such as exciton binding energy, charge carrier-phonon interactions and band gap. Light-emitting diodes and color down-conversion films are also demonstrated using these lead free perovskite nanocrystals.
Organometal halide perovskites are attracting a great deal of attention because of their long carrier diffusion lengths, wide wavelength tunability, and narrow-band emission.
This article presents an intelligent vision sensor (IVS) with embedded tiny convolutional neural network (CNN) model and programmable processing-in-sensor (PIS) circuit for real-time inference ...applications of low-power edge devices. The proposed imager realizes the full computing functions of a customized three-layers tiny network, which includes a <inline-formula> <tex-math notation="LaTeX">3 \times 3</tex-math> </inline-formula> convolution layer (stride <inline-formula> <tex-math notation="LaTeX">=</tex-math> </inline-formula> 3) with activation function of rectified linear unit (ReLU), a <inline-formula> <tex-math notation="LaTeX">2 \times 2</tex-math> </inline-formula> maximum pooling (MP) layer (stride <inline-formula> <tex-math notation="LaTeX">=</tex-math> </inline-formula> 2), and a <inline-formula> <tex-math notation="LaTeX">1 \times 1</tex-math> </inline-formula> fully connected (FC) layer for inference. A 0.8 V <inline-formula> <tex-math notation="LaTeX">128 \times 128</tex-math> </inline-formula> IVS prototype was fabricated and verified in TSMC 0.18 <inline-formula> <tex-math notation="LaTeX">\mu</tex-math> </inline-formula>m standard CMOS technology. In normal image mode, it consumed 76.4 <inline-formula> <tex-math notation="LaTeX">\mu</tex-math> </inline-formula>W with full-resolution (<inline-formula> <tex-math notation="LaTeX">126 \times 126</tex-math> </inline-formula> active resolution) image output at 125 f/s. In CNN mode, it consumed 134.5 <inline-formula> <tex-math notation="LaTeX">\mu</tex-math> </inline-formula>W at 250 f/s and an achieved iFoMs of 33.8 pJ/pixel<inline-formula> <tex-math notation="LaTeX">\cdot</tex-math> </inline-formula>frame. Using the proposed mixed-mode PIS circuits, the prototype is configured to demonstrate a "human face or not detection" task with an achieved accuracy of 93.6%.
PDF
