Search

Most encrypted speech retrieval algorithms are over-optimized for discriminability and robustness, which leads to poor security and efficiency. And it's inefficient to compute large amounts of data in a single machine. Therefore, in this paper, based on the traditional model of ciphertext speech retrieval system, an efficient encrypted speech retrieval based on Hadoop cluster under SW CPU is proposed. The study uses the SW CPU as a cloud and introduces the Hadoop cluster technology. In the proposed algorithm, firstly, the peak frequency and spectral crest factor of the speech are extracted and fused. Secondly, the hyper chaotic measurement matrix is generated by the key so that it is iterated with the feature vectors and further binarized to generate the BioHashing sequence. A pseudo-random sequence is generated by the key, mapping encryption is performed on the speech segments to generate the encrypted speech, and linear-feedback shift register (LSFR) encryption is performed on the BioHashing sequence to generate the hash index. Finally, the hash index and encrypted speech are uploaded to the cloud via WinSCP. In the SW CPU, multi-processor simultaneous operation can speed up the processing of large amounts of data. The experimental results show that the proposed BioHashing algorithm has a good compromise relationship and the proposed system model has a good security. Moreover, the Hadoop cluster technology effectively improves the retrieval performance. [ABSTRACT FROM AUTHOR]

Existing biohashing authentication schemes usually store bioinformation in the cloud, whose storage method can face the problems of bioinformation leakage and tampering, and single point of failure. To address the above problems, a ciphertext speech biohashing authentication scheme based on chaotic system and improved public chain is proposed. First, the fusion features are extracted and generated with the inner product of 3D chaotic system to generate the biohashing sequence. Second, the original speech is encrypted and uploaded to the InterPlanetary file system and then, uploaded to the blockchain after the data are encrypted. Finally, the server matches the hash sequence queried on the chain with the to-be-authenticated hash at the user's end using the Hamming distance. The experimental results show that the biohashing algorithm in this paper has better distinguishability, robustness, security and time efficiency. The key space of the speech encryption algorithm is 2 256 , which can well resist the exhaustive attack and ensure the security during speech authentication. In this paper, the combination of on-chain and off-chain storage of speech information can prevent the problem of bioinformation leakage and tampering, as well as avoiding the problem of single point of failure. [ABSTRACT FROM AUTHOR]

To present the clinical characteristics and the misdiagnosis rate of acute coronary syndrome manifested primarily as throat discomfort, we conducted a multicentric and retrospective study in the cardiology and otorhinolaryngology departments. Records of patients with primary complaint of throat discomfort, absence of chest pain at onset, and an ultimate diagnosis of acute coronary syndrome, as well as patients with pharyngitis (as controls) were collected from May 2015 to April 2016. The patients' main manifestations were compared. Logistic regression results showed that chest tightness, dyspnea, perspiring, and exertional throat symptoms were significantly associated with acute coronary syndrome, with odds ratios of 8.3 (95% CI 2.2–31.5), 10.9 (95% CI 1.8–66.9), 25.4 (95% CI 3.6–179.9), and 81.2 (95% CI 13.0–506.7). A total of 25 (56.82%) out of 44 acute coronary syndrome patients, who were first admitted to the otorhinolaryngology department, were misdiagnosed, with a 12% (3/25) mortality rate. Throat discomfort can be the principal manifestation of acute coronary syndrome. Such patients exhibit high misdiagnosis and mortality rates. Exertional throat symptoms, chest tightness, perspiring, and dyspnea were important indicators of acute coronary syndrome in patients whose main complaint was throat discomfort. The awareness of this condition will result in prompt diagnosis and reduce morbidity and mortality. [ABSTRACT FROM AUTHOR]

Automatic pain intensity estimation has great potential in current rehabilitation medicine, and patients' health status information can be obtained through the analysis of facial images. At present, deep convolutional neural networks (CNNs) have made great progress in many fields, including natural language processing, image classification and action recognition. Motivated by the current achievements, a novel end-to-end hybrid network is proposed to extract multidimensional features from image sequences, which is composed of 3D convolution, 2D convolution and 1D convolution. Specifically, the 3D convolutional neural network (3D CNN) is designed to capture the spatiotemporal features, and the 2D convolutional neural network (2D CNN) is designed to capture the spatial features, while the 1D convolutional neural network (1D CNN) is mainly used to capture the geometric information from facial landmarks. Finally, the features obtained by the three different networks are fused together for regression. The proposed HybNet is evaluated on UNBC-McMaster Shoulder Pain Expression Archive Database, and the experimental results show that it can effectively extract the discriminative high-level features and can achieve competitive performance with the state-of-the-art methods. [ABSTRACT FROM AUTHOR]