RESUMEN
Cross-modal clustering aims to cluster the high-similar cross-modal data into one group while separating the dissimilar data. Despite the promising cross-modal methods have developed in recent years, existing state-of-the-arts cannot effectively capture the correlations between cross-modal data when encountering with incomplete cross-modal data, which can gravely degrade the clustering performance. To well tackle the above scenario, we propose a novel incomplete cross-modal clustering method that integrates canonical correlation analysis and exclusive representation, named incomplete Cross-modal Subspace Clustering (i.e., iCmSC). To learn a consistent subspace representation among incomplete cross-modal data, we maximize the intrinsic correlations among different modalities by deep canonical correlation analysis (DCCA), while an exclusive self-expression layer is proposed after the output layers of DCCA. We exploit a l1,2 -norm regularization in the learned subspace to make the learned representation more discriminative, which makes samples between different clusters mutually exclusive and samples among the same cluster attractive to each other. Meanwhile, the decoding networks are employed to reconstruct the feature representation, and further preserve the structural information among the original cross-modal data. To the end, we demonstrate the effectiveness of the proposed iCmSC via extensive experiments, which can justify that iCmSC achieves consistently large improvement compared with the state-of-the-arts.
RESUMEN
This study investigates the randomness detection of cryptographic algorithms in network security. To effectively test and verify the security of ZigBee protocol in the Internet of Things, the study combines with the characteristics of ZigBee networks, and it rationally organizes and divides test modes based on the binary matrix rank theory test. Then this paper proposes a randomness detection method of ZigBee protocol in a wireless sensor network. The proposed method solves the one-sidedness that the binary matrix rank test simply assesses random sequences by linear correlation. The proposed assessment method can effectively appraise whether the ZigBee protocol has an encryption mechanism and encryption strength. Simulation results show that this method has the characteristics of fewer errors and high reliability.
RESUMEN
BACKGROUND: The aim of this study is to determine the efficacy and feasibility of a novel folate receptor (FR)-based circulating tumor cell (CTC) detection method in the diagnosis of lung cancer. CTCs were collected from 3 mL of blood based on negative enrichment by immunomagnetic beads and then labeled by a conjugate of a tumor-specific ligand folate and an oligonucleotide. METHODS: After washing off redundant conjugates, the bound conjugates were removed and analyzed by quantitative polymerase chain reaction. RESULTS: The CTC levels of 97 patients with lung cancer were significantly higher than that of the controls (18 patients with benign lung diseases; P<0.001). With a threshold of 8.7 Folate units, the method showed a sensitivity of 82.5% and a specificity of 72.2% in the diagnosis of lung cancer, especially a sensitivity of 86.8% in stage I disease. Compared with the existing clinical biomarkers such as neuron-specific enolase (NSE), carcinoembryonic antigen (CEA), and CYFRA21-1, the method showed the highest diagnostic efficiency in lung cancer (area under the curve, 0.859; 95%CI: 0.779-0.939) and stage I lung cancer (area under the curve, 0.912; 95%CI: 0.829-0.994). For future work, the CTC levels of 5 lung cancer patients higher than 8.7 Folate units/3 mL in their postoperation. CONCLUSIONS: FR-positive CTCs were feasible diagnostic biomarkers in patients with lung cancer, as well as in early-stage tumors.