Recently, functional magnetic resonance imaging (fMRI)-based brain networks have been shown to be an effective diagnostic tool with great potential for accurately detecting autism spectrum disorders (ASD). Meanwhile, the successful use of graph convolution networks (GCNs) methods based on fMRI information has improved the classification accuracy of ASD. However, many graph convolution-based methods do not fully utilize the topological information of the brain functional connectivity network (BFCN) or ignore the effect of non-imaging information. Therefore, we propose a hierarchical graph embedding model that leverage both the topological information of the BFCN and the non-imaging information of the subjects to improve the classification accuracy. Specifically, our model first use the Infomax Module to automatically identify embedded features in regions of interests (ROIs) in the brain. Then, these features, along with non-imaging information, is used to construct a population graph model. Finally, we design a graph convolution framework to propagate and aggregate the node features and obtain the results for ASD detection. Our model takes into account both the significance of the BFCN to individual subjects and relationships between subjects in the population graph. The model performed autism detection using the Autism Brain Imaging Data Exchange (ABIDE) dataset and obtained an average accuracy of 77.2% and an AUC of 87.2%. These results exceed those of the baseline approach. Through extensive experiments, we demonstrate the competitiveness, robustness and effectiveness of our model in aiding ASD diagnosis.

This paper presents a 12-bit current-steering digital-to-analog converter (DAC) with a multi-segment architecture implemented in GSMC $1\mathrm{P}-5\mathrm{M}0.11- \mu \mathrm{m}$ CMOS process. To reach high linearity and large-swing output voltage, a complementary switch array and a current source array with cascode devices are employed. Digital latches and cross-point adjustment circuits for switch control signal are introduced to suppress glitches. In order to facilitate the DAC to use as a stand alone module, a high-speed deserializer is integrated on the chip and an internal bandgap reference has also been implemented. The simulation results show that under the 1.5/3.3 V power supply voltage, DNL is below ± 0.08 LSB and INL is below ± 0.35 LSB. When the sampling rate of DAC is 50 MS/s and 100 MS/s, the spurious free dynamic range (SFDR) of DAC in the whole Nyquist domain is greater than 72 dB and 66 dB respectively, the effective number of bits (ENOB) of DAC is all greater than 11 bits.

BACKGROUND: With the rapid development of high-throughput sequencing technology, the cost of whole genome sequencing drops rapidly, which leads to an exponential growth of genome data. How to efficiently compress the DNA data generated by large-scale genome projects has become an important factor restricting the further development of the DNA sequencing industry. Although the compression of DNA bases has achieved significant improvement in recent years, the compression of quality score is still challenging. RESULTS: In this paper, by reinvestigating the inherent correlations between the quality score and the sequencing process, we propose a novel lossless quality score compressor based on adaptive coding order (ACO). The main objective of ACO is to traverse the quality score adaptively in the most correlative trajectory according to the sequencing process. By cooperating with the adaptive arithmetic coding and an improved in-context strategy, ACO achieves the state-of-the-art quality score compression performances with moderate complexity for the next-generation sequencing (NGS) data. CONCLUSIONS: The competence enables ACO to serve as a candidate tool for quality score compression, ACO has been employed by AVS(Audio Video coding Standard Workgroup of China) and is freely available at https://github.com/Yoniming/ACO.

Recently, the feature map recalibration (FMR) mechanism has been widely explored in single image super-resolution (SISR) and obtained remarkable performances. However, the existing FMR-based SISR methods directly incorporate the attention module into a deeper network structure (e.g. EDSR), while neglecting the differences between the low-level and high-level vision problems. In this paper, we design a low-level specific FMR mechanism for SISR task based on a new observation by examining current SISR methods, which all demonstrate a solid correlation between the SISR performance and the convolutional feature noise. Inspired by this, we extend the classic soft thresholding technique in the way of deep network, and develop an Adaptive Soft Thresholding (AST) module for feature noise suppression. Comparing to existing attention modules, AST is light-weighted and can be taken as an easy plug-in module in any SISR networks. To this end, we construct a adaptive Feature Denoising Super-Resolution (FDSR) network by combining the baseline EDSR and the proposed AST. Extensive experimental results show that the proposed FDSR network could achieve the state-of-the-art performances on SISR benchmarks, and significantly reduce the parameter (28.8% for EDSR, 74.6% for RCAN,s 76.0% for SAN) with respect to FMR module.

A 2×9-bit multiphase (MP) Class-G switched capacitor power amplifier (SCPA) prototype in a 65-nm CMOS with enhanced efficiency is proposed for digital intensive transmitters. In order to improve the peak output power, two independent power amplifier cells are employed rather than a shared cell, and the two cells are driven by two discrete fixed phases but not limited to adjacent phases. The two discrete fixed phases are determined by the proposed multiphase resolution of vector. Class-G operation is introduced to improve the efficiency within 6 dB back-off region. The simulated peak output power of MP Class-G SCPA is 27.17 dBm, a peak system efficiency of 41.73% and a 6 dB back-off system efficiency of 23.72%. Under the condition of 100 MHz 64-QAM signal with 7.2 dB PAPR, the PA achieves an average output power of 20. 43 dBm with an average system efficiency of 19.26%.

The paper proposes an automatic on-line signature verification system based on multiple models, a frequency function model, a shape-related parameter model, and a dynamics-related parameter model. This system uses fuzzy logic instead of the traditional Euclidean distance metric for comparison, and uses a personalized feature selection algorithm based on the DP function for parameter model building. From the statistical results it is concluded that fuzzy logic, the personalized feature selection algorithm based on the DP function and the multiple models method are helpful for automatic on-line signature verification.

For signature verification, there can be a large number of features available in a signature. However, not all these features are of use as some even could be unfavorable for verification of particular signatures. Finding an optimal personalized subset of all the possible features for a signer is crucial for signature verification systems, which are based on a parameter approach. This paper proposes a novel automated optimal personalized feature selection method based on genetic algorithms. Feature vectors for a specific signer are encoded into a population of genes or chromosomes. Through a process of genetic evolution with the application of specific genetic crossover and mutation operations an optimized personalized feature vector is obtained. An important characteristic of the method is that the number and type of features obtained for each signer through genetic evolution is not fixed and predetermined. Experimental results are presented to demonstrate the effectiveness of this novel approach to automated signature verification.