Template aging in eye movement-driven biometrics

Cell aging in vivo and in vitro

In this paper, we present a Mixture Linear Prediction based approach for robust Gammatone Cepstral Coefficients extraction (MLPGCCs). The proposed method provides performance improvement of Automatic Speaker Verification (ASV) using i-vector and Gaussian Probabilistic Linear Discriminant Analysis GPLDA modeling under transmission channel noise. The performance of the extracted MLPGCCs was evaluated using the NIST 2008 database where a single channel microphone recorded conversational speech. The system is analyzed in the presence of different channel transmission noises such as Additive White Gaussian (AWGN) and Rayleigh fading at various Signals to Noise Ratio (SNR) levels. The evaluation results show that the MLPGCCs features are a promising way for the ASV task. Indeed, the speaker verification performance using the MLPGCCs proposed features is significantly improved compared to the conventional Gammatone Frequency Cepstral Coefficients (GFCCs) and Mel Frequency Cepstral Coefficients (MFCCs) features. For speech signals corrupted with AWGN noise at SNRs ranging from (-5 dB to 15 dB), we obtain a significant reduction of the Equal Error Rate (EER) ranging from 9.41% to 6.65% and 3.72% to 1.50%, compared with conventional MFCCs and GFCCs features respectively. In addition, when the test speech signals are corrupted with Rayleigh fading channel we achieve an EER reduction ranging from 23.63% to 7.8% and from 10.88% to 6.8% compared with conventional MFCCs and GFCCs, respectively. We also found that the combination of GFCCs and MLPGCCs gives the highest performance of speaker verification system. The best performance combination achieved is around EER from 0.43% to 0.59% and 1.92% to 3.88%.

The scope and emphasis of a public health program are necessarily influenced by the changing characteristics of the population it serves. The rate of population growth affects long-range planning of community health and medical facilities. Alterations in age composition, internal migration of racial or industrial groups, changes in population density and urban-rural movement require current adaptation of the health program to solve the new problems thus created. Among the various characteristics of recent population trends, aging of the population is one of the most fundamental in its bearing on national health. The social and economic effects of an aging population have long been recognized. Dr. Louis I. Dublin appraised the problem of old age in some detail in 1926, when the provision of economic security for the aged was the dominant theme of contemporary discussion.1 The passage of the Social Security Act in 1935 represented the fruits of the efforts of this early period. Adjustment of national policy with respect to the health problems associated with aging of the population has been slower in development. Under the terms of the Social Security Act, a limited expansion of activities designed to promote the health of older adults—control of cancer and pneumonia, and industrial hygiene services—has been made possible in the cooperating States. However, the Act makes no provision for the solution of such fundamental problems as invalidity insurance and medical care of the aged. During the past five years, the health aspects of old age have received increasing attention in the discussions of public health administrators. It therefore seems appropriate to resurvey this general problem, and to consider, in particular, the nature of future trends in mortality, morbidity, and the receipt of medical care which may be expected solely as a result of changing age structure of the population.

Hospitalization Patterns for Psychiatric Disorders Across the Lifespan in Australia From July 1998 to June 2005

We propose a new approach to the problem of estimating the hyperparameters which define the interspeaker variability model in joint factor analysis. We tested the proposed estimation technique on the NIST 2006 speaker recognition evaluation data and obtained 10%-15% reductions in error rates on the core condition and the extended data condition (as measured both by equal error rates and the NIST detection cost function). We show that when a large joint factor analysis model is trained in this way and tested on the core condition, the extended data condition and the cross-channel condition, it is capable of performing at least as well as fusions of multiple systems of other types. (The comparisons are based on the best results on these tasks that have been reported in the literature.) In the case of the cross-channel condition, a factor analysis model with 300 speaker factors and 200 channel factors can achieve equal error rates of less than 3.0%. This is a substantial improvement over the best results that have previously been reported on this task.

Person re-identification plays an important role in the construction of the smart city. A reliable person re-identification system relieves users from the inefficient work of identifying the specific individual from enormous numbers of photos or videos captured by different surveillance devices. The most existing methods either focus on local discriminative features without global contextual information or scatter global features while ignoring the local features, resulting in ineffective attention to irregular pedestrian zones. In this article, a novel Transformer-CNN Coupling Network (TCCNet) is proposed to capture the fluctuant body region features in a heterogeneous feature-aware manner. We employ two bridging modules, the Low-level Feature Coupling Module (LFCM) and the High-level Feature Coupling Module (HFCM), to improve the complementary characteristics of the hybrid network. It is significantly helpful to enhance the capacity to distinguish between foreground and background features, thereby reducing the unfavorable impact of cluttered backgrounds on person re-identification. Furthermore, the duplicate loss for the two branches is employed to incorporate semantic information from distant preferences of the two branches into the resulting person representation. The experiments on two large-scale person re-identification benchmarks demonstrate that the proposed TCCNet achieves competitive results compared with several state-of-the-art approaches. The mean Average Precision (mAP) and Rank-1 identification rate on the MSMT17 dataset achieve 66.9% and 84.5%, respectively.

Soft biometrics enable the identification of subjects based on semantic descriptions collected from eyewitnesses allowing people to search in surveillance databases. Although research has recently shown an increased interest in soft biometrics, not much of the work have used crowdsourcing, and it did not investigate the impact of feature selection on identification. In this paper, we introduce a new set of facial soft biometrics and labels with a novel description for the eyebrow region. Also, we examine the use of crowdsourcing for labelling the comparative facial soft biometrics and assess its impact on the identification. Moreover, we explore the impact of feature selection with our biometric measures and evaluate the effect of label scale compression. Experiments based on the Southampton biometric tunnel database demonstrate a 100% rank-1 identification rate using 20 features only.

Optimization of the parameters characterizing sigmoidal rate-level functions based on acoustic features

Most of the speaker recognition systems use system features for speaker recognition which are mostly spectral in nature. Recently, there has been significant work on using source features, viz., prosodies and pitch dynamics, glottal flow derivative, Linear Prediction (LP) residual and its phase, wavelet-domain representation of LP residual, etc for speaker recognition. In this paper, a new source-like feature set, viz., Teager Energy Operator (TEO) Phase is developed for speaker recognition. Proposed TEO Phase has several salient features as compared to that of LP residual phase, viz., less number of spurious peaks in Hilbert envelope plot, and better perceptual and sample correlation with speech signal. Furthermore, this avoids use of voiced/unvoiced detection, preprocessing techniques such as windowing and pre-emphasis and LP residual computation. Experiments have been carried out for speaker recognition task using TEO Phase and LP residual phase with polynomial classifier of 2 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">nd</sup> order approximation. For speaker verification, a reduction in equal error rate (EER) by 2.62 % over LP residual phase is achieved for proposed feature set.

We discuss the limitations of the i-vector representation of speech segments in speaker recognition and explain how Joint Factor Analysis (JFA) can serve as an alternative feature extractor in a variety of ways. Building on the work of Zhao and Dong, we implemented a variational Bayes treatment of JFA which accommodates adaptation of universal background models (UBMs) in a natural way. This allows us to experiment with several types of features for speaker recognition: speaker factors and diagonal factors in addition to i-vectors, extracted with and without UBM adaptation in each case. We found that, in text-independent speaker verification experiments on NIST data, extracting i-vectors with UBM adaptation led to a 10% reduction in equal error rates although performance did not improve consistently over the whole DET curve. We achieved a further 10% reduction (with a similar inconsistency) by using speaker factors extracted with UBM adaptation as features. In text-dependent speaker recognition experiments on RSR2015 data, we were able to achieve very good performance using a JFA model with diagonal factors but no speaker factors as a feature extractor. Contrary to standard practice, this JFA model was configured so as to model speakerphrase combinations (rather than speakers) and it was trained on utterances of very short duration (rather than whole recording sessions). We also present a variant of the length normalization trick inspired by uncertainty propagation which leads to substantial gains in performance over the whole DET curve.

This study proposes an i-Vector based approach to gender identification. Gender-labeled utterances from the Fisher English (FE) corpus are used to formulate an i-Vector extraction framework, and a Probabilistic Linear Discriminant Analysis (PLDA) back-end is employed to compute the scores for gender identification. A novel duration mismatch compensation strategy is also presented that offers very little degradation in identification accuracy even with a large reduction in the duration of the test-segment. The proposed method is shown to consistently outperform a GMM-UBM based gender-identification scheme on several test-sets created from a held-out portion of the FE corpus, and is able to achieve an identification accuracy of up to 97.63%. On the severely distorted and multilingual DARPA-RATS (Robust Automatic Transcription of Speech) corpora, the proposed approach achieves an identification accuracy of 76.48% using only the FE data in training. Next, a novel unsupervised domain adaptation strategy is also presented that utilizes only unlabeled RATS data to adapt the out-of-domain PLDA parameters derived from the FE training data. The strategy is able to offer a 6.8% relative improvement in identification accuracy, and a 14.75% relative reduction in Equal Error Rate (EER) compared to using the out-of-domain PLDA model on the RATS test-utterances. These improvements are significant since: 1) RATS test-utterances are severely distorted, 2) No labeled data of any kind is used for 4 of the 5 languages present in the test-utterances.

In this paper appropriate confidence measures (CMs) are investigated for Mandarin command word recognition, both in the so-called target region and non-target region, respectively. Here the target region refers to the recognized speech part of command word while the non-target region refers to the recognized silence part. It shows that exploiting extra information in the non-target region can effectively complement the traditional CM which usually focus on the target region. Furthermore, when analyzing the non-target region in a more theoretical way, where Bayesian information criterion (BIC) is employed to locate more precise boundary in the non-target region, even more improvement is achieved. In two different Mandarin telephone command word tasks, more than 20% relative reduction of equal error rate (EER) is obtained.

It is widely known that the quality of confidence measure is critical for speech applications. In this paper, we present our recent work on improving word confidence scores by calibrating them using a small set of calibration data when only the recognized word sequence and associated raw confidence scores are made available. The core of our technique is the maximum entropy model with distribution constraints which naturally and effectively make use of the word distribution, the raw confidence-score distribution, and the context information. We demonstrate the effectiveness of our approach by showing that it can achieve relative 38% mean square error (MSE), 39% negative normalized likelihood (NNLL), and 23% equal error rate (EER) reduction on a voice mail transcription data set and relative 35% MSE, 45% NNLL, and 35% EER reduction on a command and control data set.

Occupational carcinogens, smoking, and obesity are believed to be the main causing agents of kidney cancer. China is undergoing rapid industrialization, and hence the people's lifestyles have witnessed tremendous changes. However, the trend of kidney cancer incidence during the late 20th and early 21st centuries remains unexplored in China. Data from the Global Burden of Diseases (GBD; 2019) was retrieved for the incidence of kidney cancer from 1990 to 2019. The rates of disease average annual percentage changes (AAPC) were assessed using joinpoint regression analysis. Age-period-cohort (APC) model was used to assess age, period, and cohort effects on the incidence of the disease simultaneously. An increase in age-standardized incidence rates (ASIR) of kidney cancer was observed from 1990 to 2019 in total residents (1.33 - 4.24), men (1.56 - 6.15), and women (1.11 - 2.31) per 100,000 population suggesting a more obvious increase in males than in females. A consistent increase in age effects was observed in all age groups, especially in males. On the other hand, the 70 - 74 age group in females showed greater age effects. In addition, the period effects analysis showed that the incidence of kidney cancer increased with time. Moreover, the analysis of cohort effects showed a decrease in the disease in birth cohorts, especially before 1940. The incidence of kidney cancer is increasing rapidly in China. The kidney cancer burden will rise in the next decades due to population aging, environmental pollution, occupation, food safety, and so on. Results of this study suggest that more etiological studies should be performed to identify the driving factors for kidney cancer trends, and appropriate preventive measures should be implemented for the age-, period-, and cohort-related factors in the population.

BackgroundBrazil leads the world in number of firearm deaths and ranks sixth by country in rate of firearm deaths per 100,000 people. This study aims to analyze trends in and burden of mortality by firearms, according to age and sex, for Brazil, and the association between these deaths and indicators of possession and carrying of weapons using data from the global burden of diseases, injuries, and risk factors study (GBD) 2017.MethodsWe used GBD 2017 estimates of mortality due to physical violence and self-harm from firearms for Brazil to analyze the association between deaths by firearms and explanatory variables.ResultsDeaths from firearms increased in Brazil from 25,819 in 1990 to 48,493 in 2017. Firearm mortality rates were higher among men and in the 20–24 age group; the rate was 20 times higher than for women in the same age group. Homicide rates increased during the study period, while mortality rates for suicides and accidental deaths decreased. The group of Brazilian federation units with the highest firearm collection rate (median = 7.5) showed reductions in the rate of total violent deaths by firearms. In contrast, the group with the lowest firearm collection rate (median = 2.0) showed an increase in firearm deaths from 2000 to 2017. An increase in the rate of voluntary return of firearms was associated with a reduction in mortality rates of unintentional firearm deaths (r = −0.364, p < 0.001). An increase in socio-demographic index (SDI) was associated with a reduction in all firearm death rates (r = −0.266, p = 0.008). An increase in the composite index of firearms seized or collected was associated with a reduction in rates of deaths by firearm in the subgroup of females, children, and the elderly (r = −0.269, p = 0.005).ConclusionsThere was a change in the trend of firearms deaths after the beginning of the collection of weapons in 2004. Federation units that collected more guns have reduced rates of violent firearm deaths.