arXiv subjects

Jiaqi Xu

Publications and source records attributed to Jiaqi Xu.

75 records · Page 5Linked to original sources

arXiv2020

A Novel Physics-based Channel Model for Reconfigurable Intelligent Surface-assisted Multi-user Communication Systems

The reconfigurable intelligent surface (RIS) is one of the promising technologies contributing to the next generation smart radio environment. A novel physics-based RIS channel model is proposed. Particularly, we consider the RIS and the scattering environment as a whole by studying the signal's multipath propagation, as well as the radiation pattern of the RIS. The model suggests that the RIS-assisted wireless channel can be approximated by a Rician distribution. Analytical expressions are derived for the shape factor and the scale factor of the distribution. For the case of continuous phase shifts, the distribution depends on the number of elements of the RIS and the observing direction of the receiver. For the case of continuous phase shifts, the distribution further depends on the quantization level of the RIS phase error. The scaling law of the average received power is obtained from the scale factor of the distribution. For the application scenarios where RIS functions as an anomalous reflector, we investigate the performance of single RIS-assisted multiple access networks for time-division multiple access (TDMA), frequency-division multiple access (FDMA) and non-orthogonal multiple access (NOMA). Closed-form expressions for the outage probability of the proposed channel model are derived. It is proved that a constant diversity order exists, which is independent of the number of RIS elements. Simulation results are presented to confirm that the proposed model applies effectively to the phased-array implemented RISs.

eess.SP↗

arXiv2020

Deep Mining External Imperfect Data for Chest X-ray Disease Screening

Deep learning approaches have demonstrated remarkable progress in automatic Chest X-ray analysis. The data-driven feature of deep models requires training data to cover a large distribution. Therefore, it is substantial to integrate knowledge from multiple datasets, especially for medical images. However, learning a disease classification model with extra Chest X-ray (CXR) data is yet challenging. Recent researches have demonstrated that performance bottleneck exists in joint training on different CXR datasets, and few made efforts to address the obstacle. In this paper, we argue that incorporating an external CXR dataset leads to imperfect training data, which raises the challenges. Specifically, the imperfect data is in two folds: domain discrepancy, as the image appearances vary across datasets; and label discrepancy, as different datasets are partially labeled. To this end, we formulate the multi-label thoracic disease classification problem as weighted independent binary tasks according to the categories. For common categories shared across domains, we adopt task-specific adversarial training to alleviate the feature differences. For categories existing in a single dataset, we present uncertainty-aware temporal ensembling of model predictions to mine the information from the missing labels further. In this way, our framework simultaneously models and tackles the domain and label discrepancies, enabling superior knowledge mining ability. We conduct extensive experiments on three datasets with more than 360,000 Chest X-ray images. Our method outperforms other competing models and sets state-of-the-art performance on the official NIH test set with 0.8349 AUC, demonstrating its effectiveness of utilizing the external dataset to improve the internal classification.

cs.CV↗

arXiv2019

IRNet: Instance Relation Network for Overlapping Cervical Cell Segmentation

Cell instance segmentation in Pap smear image remains challenging due to the wide existence of occlusion among translucent cytoplasm in cell clumps. Conventional methods heavily rely on accurate nuclei detection results and are easily disturbed by miscellaneous objects. In this paper, we propose a novel Instance Relation Network (IRNet) for robust overlapping cell segmentation by exploring instance relation interaction. Specifically, we propose the Instance Relation Module to construct the cell association matrix for transferring information among individual cell-instance features. With the collaboration of different instances, the augmented features gain benefits from contextual information and improve semantic consistency. Meanwhile, we proposed a sparsity constrained Duplicate Removal Module to eliminate the misalignment between classification and localization accuracy for candidates selection. The largest cervical Pap smear (CPS) dataset with more than 8000 cell annotations in Pap smear image was constructed for comprehensive evaluation. Our method outperforms other methods by a large margin, demonstrating the effectiveness of exploring instance relation.

cs.CV↗