A Flexible Coding Scheme Based on Block Krylov Subspace Approximation for Light Field Displays with Stacked Multiplicative Layers.

To create a realistic 3D perception on glasses-free displays, it is critical to support continuous motion parallax, greater depths of field, and wider fields of view. A new type of Layered or Tensor light field 3D display has attracted greater attention these days. Using only a few light-attenuating pixelized layers (e.g., LCD panels), it supports many views from different viewing directions that can be displayed simultaneously with a high resolution. This paper presents a novel flexible scheme for efficient layer-based representation and lossy compression of light fields on layered displays. The proposed scheme learns stacked multiplicative layers optimized using a convolutional neural network (CNN). The intrinsic redundancy in light field data is efficiently removed by analyzing the hidden low-rank structure of multiplicative layers on a Krylov subspace. Factorization derived from Block Krylov singular value decomposition (BK-SVD) exploits the spatial correlation in layer patterns for multiplicative layers with varying low ranks. Further, encoding with HEVC eliminates inter-frame and intra-frame redundancies in the low-rank approximated representation of layers and improves the compression efficiency. The scheme is flexible to realize multiple bitrates at the decoder by adjusting the ranks of BK-SVD representation and HEVC quantization. Thus, it would complement the generality and flexibility of a data-driven CNN-based method for coding with multiple bitrates within a single training framework for practical display applications. Extensive experiments demonstrate that the proposed coding scheme achieves substantial bitrate savings compared with pseudo-sequence-based light field compression approaches and state-of-the-art JPEG and HEVC coders.

Ionotronic Halide Perovskite Drift-Diffusive Synapses for Low-Power Neuromorphic Computation.

Emulation of brain-like signal processing is the foundation for development of efficient learning circuitry, but few devices offer the tunable conductance range necessary for mimicking spatiotemporal plasticity in biological synapses. An ionic semiconductor which couples electronic transitions with drift-diffusive ionic kinetics would enable energy-efficient analog-like switching of metastable conductance states. Here, ionic-electronic coupling in halide perovskite semiconductors is utilized to create memristive synapses with a dynamic continuous transition of conductance states. Coexistence of carrier injection barriers and ion migration in the perovskite films defines the degree of synaptic plasticity, more notable for the larger organic ammonium and formamidinium cations than the inorganic cesium counterpart. Optimized pulsing schemes facilitates a balanced interplay of short- and long-term plasticity rules like paired-pulse facilitation and spike-time-dependent plasticity, cardinal for learning and computing. Trained as a memory array, halide perovskite synapses demonstrate reconfigurability, learning, forgetting, and fault tolerance analogous to the human brain. Network-level simulations of unsupervised learning of handwritten digit images utilizing experimentally derived device parameters, validates the utility of these memristors for energy-efficient neuromorphic computation, paving way for novel ionotronic neuromorphic architectures with halide perovskites as the active material.

Influence of laser parameters on selective retinal treatment using single-phase heat transfer analyses.

Selective thermal treatment to retina is induced by short pulsed lasers to denaturize retinal pigment epithelium (RPE) selectively, while sparing the sensitive photoreceptors. The problem associated with the usage of short pulsed laser is the difficulty in determining the correct dosimetry parameters. This study quantifies the influence of laser parameters over the therapeutic range. The laser-tissue interaction is numerically investigated by analyzing the transient temperature in ocular tissues during the treatment. The rate process analysis for thermal injury is employed to estimate the selective damage of retina. The contours of Arrhenius integral value (Omeg/ Omegamax) presented in this study show both the area and magnitude of damage caused by various laser parameters. Results reveal that the 2 micros pulsed laser with green wavelength and Gaussian profile is relatively more effective for selective retinal treatment. The repetition frequency of 100 Hz is found to produce selectively RPE damage, while higher frequencies produce collateral damage to neural retina and choroid located within 2 microm from the RPE interface.

Characterization of substrate preference for Slc1p and Cst26p in Saccharomyces cerevisiae using lipidomic approaches and an LPAAT activity assay.

BACKGROUND: Phosphatidic acid (PA) is a key regulated intermediate and precursor for de novo biosynthesis of all glycerophospholipids. PA can be synthesized through the acylation of lysophosphatidic acid (LPA) by 1-acyl-3-phosphate acyltransferase (also called lysophosphatidic acid acyltransferase, LPAAT). Recent findings have substantiated the essential roles of acyltransferases in various biological functions. METHODOLOGIES/PRINCIPAL FINDINGS: We used a flow-injection-based lipidomic approach with approximately 200 multiple reaction monitoring (MRM) transitions to pre-screen fatty acyl composition of phospholipids in the yeast Saccharomyces cerevisiae mutants. Dramatic changes were observed in fatty acyl composition in some yeast mutants including Slc1p, a well-characterized LPAAT, and Cst26p, a recently characterized phosphatidylinositol stearoyl incorporating 1 protein and putative LPAAT in S. cerevisiae. A comprehensive high-performance liquid chromatography-based multi-stage MRM approach (more than 500 MRM transitions) was developed and further applied to quantify individual phospholipids in both strains to confirm these changes. Our data suggest potential fatty acyl substrates as well as fatty acyls that compensate for defects in both Cst26p and Slc1p mutants. These results were consistent with those from a non-radioactive LPAAT enzymatic assay using C17-LPA and acyl-CoA donors as substrates. CONCLUSIONS: We found that Slc1p utilized fatty acid (FA) 18:1 and FA 14:0 as substrates to synthesize corresponding PAs; moreover, it was probably the only acyltransferase responsible for acylation of saturated short-chain fatty acyls (12:0 and 10:0) in S. cerevisiae. We also identified FA 18:0, FA 16:0, FA 14:0 and exogenous FA 17:0 as preferred substrates for Cst26p because transformation with a GFP-tagged CST26 restored the phospholipid profile of a CST26 mutant. Our current findings expand the enzymes and existing scope of acyl-CoA donors for glycerophospholipid biosynthesis.

A 64-channel readout ASIC for nanowire biosensor array with electrical calibration scheme.

A 1.8-mW, 18.5-mm(2) 64-channel current readout ASIC was implemented in 0.18-µm CMOS together with a new calibration scheme for silicon nanowire biosensor arrays. The ASIC consists of 64 channels of dedicated readout and conditioning circuits which incorporate correlated double sampling scheme to reduce the effect of 1/f noise and offset from the analog front-end. The ASIC provides a 10-bit digital output with a sampling rate of 300 S/s whilst achieving a minimum resolution of 7 pA(rms). A new electrical calibration method was introduced to mitigate the issue of large variations in the nano-scale sensor device parameters and optimize the sensor sensitivity. The experimental results show that the proposed calibration technique improved the sensitivity by 2 to 10 times and reduced the variation between dataset by 9 times.

A low-power high-performance accelerometer ASIC for high-end medical motion sensing.

A 170µW readout IC for a capacitive MEMS acceleration sensor was implemented in a 1.5V 0.13µm CMOS for high-end medical motion sensing applications. The accelerometer achieves a 45µg/vHz noise floor and a dynamic range larger than 87dB for a 400Hz bandwidth. Power reduction is achieved by introducing reset and common-mode feedback circuit techniques based on a non-unity-gain feedback configuration.