Xiang Zhang

Bio: Xiang Zhang is an academic researcher from Baylor College of Medicine. The author has contributed to research in topics: Medicine & Computer science. The author has an hindex of 154, co-authored 1733 publications receiving 117576 citations. Previous affiliations of Xiang Zhang include University of California, Berkeley & University of Texas MD Anderson Cancer Center.

All-Liquid Photonic Microcavity Stabilized by Quantum Dots

Abstract: We demonstrate two simple methods to fabricate QD-stabilized toluene microdroplets in water as whispering gallery mode microscale resonators in an all-liquid phase. The toluene microdroplets show size-dependently high Q-factors up to 5100 resulting from the stable QD-loaded microdroplets. The highly QD-stabilized toluene microdroplet resonators in the all-liquid phase would be promising for multiple all-liquid lasers.

Relative Impact of Pain and Fatigue on Work Productivity in Patients with Rheumatoid Arthritis from the RA-BEAM Baricitinib Trial

Abstract: Introduction To explore the relationship of pain and fatigue with daily activity and work productivity in rheumatoid arthritis (RA) patients from the baricitinib clinical trial, RA-BEAM.

A signal processing analysis of Purkinje cells in vitro

Abstract: Cerebellar Purkinje cells in vitro fire recurrent sequences of Sodium and Calcium spikes. Here, we analyze the Purkinje cell using harmonic analysis, and our experiments reveal that its output signal is comprised of three distinct frequency bands, which are combined using Amplitude and Frequency Modulation (AM/FM). We find that the three characteristic frequencies - Sodium, Calcium and Switching - occur in various combinations in all waveforms observed using whole-cell current clamp recordings. We found that the Calcium frequency can display a frequency doubling of its frequency mode, and the Switching frequency can act as a possible generator of pauses that are typically seen in Purkinje output recordings. Using a reversibly photo-switchable kainate receptor agonist, we demonstrate the external modulation of the Calcium and Switching frequencies. These experiments and Fourier analysis suggest that the Purkinje cell can be understood as a harmonic signal oscillator, enabling a higher level of interpretation of Purkinje signaling based on modern signal processing techniques.

Ring-Contraction Disproportionation/Spirocyclization Cascade Reaction of Isochromeno[4,3-b]indol-5(11H)-ones: Synthesis of N-Unsubstituted Spirocycles.

Abstract: An efficient ring-contraction reaction of isochromeno[4,3-b]indol-5(11H)-ones via a nucleophile-induced disproportionation/spirocyclization cascade process has been developed under mild conditions. The process realized the conversion of isochromeno[4,3-b]indol-5(11H)-ones into N-unsubstituted spiro[indoline-2,1′-isobenzofuran]-3,3′-diones and spiro[indoline-2,1′-isoindoline]-3,3′-diones in the absence of a transition-metal catalyst or oxidant. Gram-scale reaction further demonstrated the practicability of the protocol.

Selective switch made from a graded nanosandwich chain

Abstract: The magnetic plasmon (MP) modes in a metallic nanosandwich chain with a linearly increasing spacing along the chain has been investigated. Because of the graded coupling between nanosandwiches, the MP gradon with special field localization and large field amplitude can be found in the chain as well as the extended mode, which differs from the case of periodic chain. Using this property, we can precisely control the field in the chain and guide it to different ports at different frequencies, which works as a selective switch and may have potential application in integrated optics. Finally, we give out the underlying physical mechanism to help better understand and apply this graded chain.

Deep Residual Learning for Image Recognition

Abstract: Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers—8× deeper than VGG nets [40] but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions1, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.

Very Deep Convolutional Networks for Large-Scale Image Recognition

Abstract: In this work we investigate the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting. Our main contribution is a thorough evaluation of networks of increasing depth using an architecture with very small (3x3) convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers. These findings were the basis of our ImageNet Challenge 2014 submission, where our team secured the first and the second places in the localisation and classification tracks respectively. We also show that our representations generalise well to other datasets, where they achieve state-of-the-art results. We have made our two best-performing ConvNet models publicly available to facilitate further research on the use of deep visual representations in computer vision.

Deep Residual Learning for Image Recognition

Abstract: Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers---8x deeper than VGG nets but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.

Going deeper with convolutions

Abstract: We propose a deep convolutional neural network architecture codenamed Inception that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14). The main hallmark of this architecture is the improved utilization of the computing resources inside the network. By a carefully crafted design, we increased the depth and width of the network while keeping the computational budget constant. To optimize quality, the architectural decisions were based on the Hebbian principle and the intuition of multi-scale processing. One particular incarnation used in our submission for ILSVRC14 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.