scispace - formally typeset
Search or ask a question
Author

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.


Papers
More filters
Journal ArticleDOI
TL;DR: Nanoporous silicon (Si) networks with controllable porosity and thickness are fabricated by a simple and scalable electrochemical process, and then released from Si wafers and transferred to flexible and conductive substrates to serve as high performance Li-ion battery electrodes.
Abstract: Nanoporous silicon (Si) networks with controllable porosity and thickness are fabricated by a simple and scalable electrochemical process, and then released from Si wafers and transferred to flexible and conductive substrates. These nanoporous Si networks serve as high performance Li-ion battery electrodes, with an initial discharge capacity of 2570 mA h g−1, above 1000 mA h g−1 after 200 cycles without any electrolyte additives.

65 citations

Proceedings ArticleDOI
07 Nov 2017
TL;DR: In this article, the authors proposed an approach which has better EEG interpretation ability via raw Electroencephalography (EEG) signal analysis for multi-person and multi-class brain activity recognition.
Abstract: An electroencephalography (EEG) based brain activity recognition is a fundamental field of study for a number of significant applications such as intention prediction, appliance control, and neurological disease diagnosis in smart home and smart healthcare domains. Existing techniques mostly focus on binary brain activity recognition for a single person, which limits their deployment in wider and complex practical scenarios. Therefore, multi-person and multi-class brain activity recognition has obtained popularity recently. Another challenge faced by brain activity recognition is the low recognition accuracy due to the massive noises and the low signal-to-noise ratio in EEG signals. Moreover, the feature engineering in EEG processing is time-consuming and highly relies on the expert experience. In this paper, we attempt to solve the above challenges by proposing an approach which has better EEG interpretation ability via raw Electroencephalography (EEG) signal analysis for multi-person and multi-class brain activity recognition. Specifically, we analyze inter-class and inter-person EEG signal characteristics, based on which to capture the discrepancy of inter-class EEG data. Then, we adopt an Autoencoder layer to automatically refine the raw EEG signals by eliminating various artifacts. We evaluate our approach on both a public and a local EEG datasets and conduct extensive experiments to explore the effect of several factors (such as normalization methods, training data size, and Autoencoder hidden neuron size) on the recognition results. The experimental results show that our approach achieves a high accuracy comparing to competitive state-of-the-art methods, indicating its potential in promoting future research on multi-person EEG recognition.

65 citations

Journal ArticleDOI
TL;DR: In this article, an acoustic metasurface with 1/12 wavelength thickness was introduced to realize an acoustic carpet cloak for a randomly rapid-change surface and a virtual acoustic diffuser from a flat surface using a set of Helmholtz resonators.
Abstract: Recently developed metasurfaces have been used for surface engineering applications. However, the ability to cloak or mimic reflective surfaces with a large in-plane phase gradient remains unexplored. One major challenge is that even with a small incidence angle, the strong acoustic impedance variation induced by the random height profile creates additional scattering and increases the complexity of the analysis and design of the metasurface. Here, we introduce an acoustic metasurface with 1/12 wavelength thickness to realize an acoustic carpet cloak for a randomly rapid-change surface and a virtual acoustic diffuser from a flat surface using a set of Helmholtz resonators. The limitation of the metasurface for large phase gradient application is explored based on a nonlocal model that considers the contributions from neighboring surface profiles. This study extends the integration of smart acoustic surface and may find applications of surface engineering such as in architectural acoustics.

65 citations

Posted ContentDOI
27 Nov 2020-bioRxiv
TL;DR: This work demonstrated that the bone microenvironment facilitates breast and prostate cancer cells to further metastasize and establish multi-organ secondary metastases, and suggested a stable reprogramming process that engenders cancer cells more metastatic.
Abstract: Metastasis has been considered as the terminal step of tumor progression. However, recent clinical studies suggest that many metastases are seeded from other metastases, rather than primary tumors. Thus, some metastases can further spread, but the corresponding pre-clinical models are lacking. By using several approaches including parabiosis and an evolving barcode system, we demonstrated that the bone microenvironment facilitates breast and prostate cancer cells to further metastasize and establish multi-organ secondary metastases. Importantly, dissemination from the bone microenvironment appears to be more aggressive compared to that from mammary tumors and lung metastases. We further uncovered that this metastasis-promoting effect is independent from genetic selection, as single cell-derived cancer cell populations (SCPs) exhibited enhanced metastasis capacity after being extracted from the bone microenvironment. Taken together, our work revealed a previously unappreciated effect of the bone microenvironment on metastasis evolution, and suggested a stable reprogramming process that engenders cancer cells more metastatic.

64 citations

Journal ArticleDOI
TL;DR: In this paper, the authors studied the possible correlation between flux densities (FR, FK, FO, FX, and Fγ) in the radio, infrared, optical, X-ray, and γ-ray wave bands, in both the low and high states.
Abstract: Using multi-wave band data for 61 γ-ray-loud blazars (17 BL Lacertae objects and 44 flat-spectrum radio quasars [FSRQs]), we have studied the possible correlation between flux densities (FR, FK, FO, FX, and Fγ) in the radio, infrared, optical, X-ray, and γ-ray wave bands, in both the low and high states. For some blazars, it is hard to determine whether they are in a low or a high state because only one data point is available for each of them; initially, we exclude these blazars in our analysis. However, we include these blazars in later analysis by temporarily assuming them to be in a low state or a high state. Our main results are as follows. There are very strong correlations between FX and FO and between FO and FK in both low and high states. However, a strong correlation between FX and FK exists only in the low state. No definite correlation is found between γ-ray flux density and those of lower energy bands; however, there are hints of anticorrelation between Fγ and FX as well as Fγ and FO and a positive correlation between Fγ and FR. From these results we suggest that (1) photons from infrared to X-rays are emitted by the same particles via synchrotron radiation and (2) if γ-rays are mainly produced by an inverse Compton scattering mechanism, it seems that the up-scattered soft photons are from external photons rather than synchrotron photons.

63 citations


Cited by
More filters
Proceedings ArticleDOI
27 Jun 2016
TL;DR: In this article, the authors proposed a residual learning framework to ease the training of networks that are substantially deeper than those used previously, which won the 1st place on the ILSVRC 2015 classification task.
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.

123,388 citations

Proceedings Article
04 Sep 2014
TL;DR: This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small 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.
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.

55,235 citations

Journal ArticleDOI
04 Mar 2011-Cell
TL;DR: Recognition of the widespread applicability of these concepts will increasingly affect the development of new means to treat human cancer.

51,099 citations

Posted Content
TL;DR: This work presents a residual learning framework to ease the training of networks that are substantially deeper than those used previously, and provides comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth.
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.

44,703 citations

Proceedings ArticleDOI
07 Jun 2015
TL;DR: Inception as mentioned in this paper is a deep convolutional neural network architecture that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14).
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.

40,257 citations