Self-Supervised Open-Set Speaker Recognition with Laguerre-Voronoi Descriptors.

Speaker recognition is a challenging problem in behavioral biometrics that has been rigorously investigated over the last decade. Although numerous supervised closed-set systems inherit the power of deep neural networks, limited studies have been made on open-set speaker recognition. This paper proposes a self-supervised open-set speaker recognition that leverages the geometric properties of speaker distribution for accurate and robust speaker verification. The proposed framework consists of a deep neural network incorporating a wider viewpoint of temporal speech features and Laguerre-Voronoi diagram-based speech feature extraction. The deep neural network is trained with a specialized clustering criterion that only requires positive pairs during training. The experiments validated that the proposed system outperformed current state-of-the-art methods in open-set speaker recognition and cluster representation.

Human Micro-Expressions in Multimodal Social Behavioral Biometrics.

The advent of Social Behavioral Biometrics (SBB) in the realm of person identification has underscored the importance of understanding unique patterns of social interactions and communication. This paper introduces a novel multimodal SBB system that integrates human micro-expressions from text, an emerging biometric trait, with other established SBB traits in order to enhance online user identification performance. Including human micro-expression, the proposed method extracts five other original SBB traits for a comprehensive representation of the social behavioral characteristics of an individual. Upon finding the independent person identification score by every SBB trait, a rank-level fusion that leverages the weighted Borda count is employed to fuse the scores from all the traits, obtaining the final identification score. The proposed method is evaluated on a benchmark dataset of 250 Twitter users, and the results indicate that the incorporation of human micro-expression with existing SBB traits can substantially boost the overall online user identification performance, with an accuracy of 73.87% and a recall score of 74%. Furthermore, the proposed method outperforms the state-of-the-art SBB systems.

Transcriptome Analysis of Human Glioblastoma Cells Susceptible to Infection with the Leningrad-16 Vaccine Strain of Measles Virus.

Glioblastoma multiforme (GBM) accounts for almost half of all primary malignant brain tumors in adults and has a poor prognosis. Here we demonstrated the oncolytic potential of the L-16 vaccine strain of measles virus (MV) against primary human GBM cells and characterized the genetic patterns that determine the sensitivity of primary human GBM cells to oncolytic therapy. MV replicated in all GBM cells, and seven out of eight cell lines underwent complete or partial oncolysis. RNA-Seq analysis identified about 1200 differentially expressed genes (FDR < 0.05) with at least two-fold expression level change between MV-infected and uninfected cells. Among them, the most significant upregulation was observed for interferon response, apoptosis and cytokine signaling. One out of eight GBM cell lines was defective in type I interferon production and, thus, in the post-interferon response, other cells lacked expression of different cellular defense factors. Thus, none of the cell lines displayed induction of the total gene set necessary for effective inhibition of MV replication. In the resistant cells, we detected aberrant expression of metalloproteinase genes, particularly MMP3. Thus, such genes could be considered intriguing candidates for further study of factors responsible for cell sensitivity and resistance to L-16 MV infection.

Machine learning prediction of atrial fibrillation in cardiovascular patients using cardiac magnetic resonance and electronic health information.

Background: Atrial fibrillation (AF) is a commonly encountered cardiac arrhythmia associated with morbidity and substantial healthcare costs. While patients with cardiovascular disease experience the greatest risk of new-onset AF, no risk model has been developed to predict AF occurrence in this population. We hypothesized that a patient-specific model could be delivered using cardiovascular magnetic resonance (CMR) disease phenotyping, contextual patient health information, and machine learning. Methods: Nine thousand four hundred forty-eight patients referred for CMR imaging were enrolled and followed over a 5-year period. Seven thousand, six hundred thirty-nine had no prior history of AF and were eligible to train and validate machine learning algorithms. Random survival forests (RSFs) were used to predict new-onset AF and compared to Cox proportional-hazard (CPH) models. The best performing features were identified from 115 variables sourced from three data domains: (i) CMR-based disease phenotype, (ii) patient health questionnaire, and (iii) electronic health records. We evaluated discriminative performance of optimized models using C-index and time-dependent AUC (tAUC). Results: A RSF-based model of 20 variables (CIROC-AF-20) delivered an overall C-index of 0.78 for the prediction of new-onset AF with respective tAUCs of 0.80, 0.79, and 0.78 at 1-, 2- and 3-years. This outperformed a novel CPH-based model and historic AF risk scores. At 1-year of follow-up, validation cohort patients classified as high-risk of future AF by CIROC-AF-20 went on to experience a 17.3% incidence of new-onset AF, being 24.7-fold higher risk than low risk patients. Conclusions: Using phenotypic data available at time of CMR imaging we developed and validated the first described risk model for the prediction of new-onset AF in patients with cardiovascular disease. Complementary value was provided by variables from patient-reported measures of health and the electronic health record, illustrating the value of multi-domain phenotypic data for the prediction of AF.

KinectGaitNet: Kinect-Based Gait Recognition Using Deep Convolutional Neural Network.

Over the past decade, gait recognition had gained a lot of attention in various research and industrial domains. These include remote surveillance, border control, medical rehabilitation, emotion detection from posture, fall detection, and sports training. The main advantages of identifying a person by their gait include unobtrusiveness, acceptance, and low costs. This paper proposes a convolutional neural network KinectGaitNet for Kinect-based gait recognition. The 3D coordinates of each of the body joints over the gait cycle are transformed to create a unique input representation. The proposed KinectGaitNet is trained directly using the 3D input representation without the necessity of the handcrafted features. The KinectGaitNet design allows avoiding gait cycle resampling, and the residual learning method ensures high accuracy without the degradation problem. The proposed deep learning architecture surpasses the recognition performance of all state-of-the-art methods for Kinect-based gait recognition by achieving 96.91% accuracy on UPCV and 99.33% accuracy on the KGB dataset. The method is the first, to the best of our knowledge, deep learning-based architecture that is based on a unique 3D input representation of joint coordinates. It achieves performance higher than previous traditional and deep learning methods, with fewer parameters and shorter inference time.

Motion Capture Sensor-Based Emotion Recognition Using a Bi-Modular Sequential Neural Network.

Motion capture sensor-based gait emotion recognition is an emerging sub-domain of human emotion recognition. Its applications span a variety of fields including smart home design, border security, robotics, virtual reality, and gaming. In recent years, several deep learning-based approaches have been successful in solving the Gait Emotion Recognition (GER) problem. However, a vast majority of such methods rely on Deep Neural Networks (DNNs) with a significant number of model parameters, which lead to model overfitting as well as increased inference time. This paper contributes to the domain of knowledge by proposing a new lightweight bi-modular architecture with handcrafted features that is trained using a RMSprop optimizer and stratified data shuffling. The method is highly effective in correctly inferring human emotions from gait, achieving a micro-mean average precision of 0.97 on the Edinburgh Locomotive Mocap Dataset. It outperforms all recent deep-learning methods, while having the lowest inference time of 16.3 milliseconds per gait sample. This research study is beneficial to applications spanning various fields, such as emotionally aware assistive robotics, adaptive therapy and rehabilitation, and surveillance.

A multimethod approach to the differentiation of enthesis bone microstructure based on soft tissue type.

Whereas there is a wealth of research studying the nature of various soft tissues that attach to bone, comparatively little research focuses on the bone's microscopic properties in the area where these tissues attach. Using scanning electron microscopy to generate a dataset of 1600 images of soft tissue attachment sites, an image classification program with novel convolutional neural network architecture can categorize images of attachment areas by soft tissue type based on observed patterns in microstructure morphology. Using stained histological thin section and liquid crystal cross-polarized microscopy, it is determined that soft tissue type can be quantitatively determined from the microstructure. The primary diagnostic characters are the orientation of collagen fibers and heterogeneity of collagen density throughout the attachment area thickness. These determinations are made across broad taxonomic sampling and multiple skeletal elements.

FN3 protein fragment containing two type III fibronectin domains from B. longum GT15 binds to human tumor necrosis factor alpha in vitro.

Most species of the genus Bifidobacterium contain the gene cluster PFNA, which is presumably involved in the species-specific communication between bacteria and their hosts. The gene cluster PFNA consists of five genes including fn3, which codes for a protein containing two fibronectin type III domains. Each fibronectin domain contains sites similar to cytokine-binding sites of human receptors. Based on this finding we assumed that this protein would bind specifically to human cytokines in vitro. We cloned a fragment of the fn3 gene (1503 bp; 501 aa) containing two fibronectin domains, from the strain B. longum subsp. longum GT15. After cloning the fragment into the expression vector pET16b and expressing it in E. coli, the protein product was purified to a homogenous state for further analysis. Using the immunoferment method, we tested the purified fragment's ability to bind the following human cytokines: IL-1ß, IL-6, IL-10, TNFα. We developed a sandwich ELISA system to detect any specific interactions between the purified protein and any of the studied cytokines. We found that the purified protein fragment only binds to TNFα.

Biometric Identification from Human Aesthetic Preferences.

In recent years, human-machine interactions encompass many avenues of life, ranging from personal communications to professional activities. This trend has allowed for person identification based on behavior rather than physical traits to emerge as a growing research domain, which spans areas such as online education, e-commerce, e-communication, and biometric security. The expression of opinions is an example of online behavior that is commonly shared through the liking of online images. Visual aesthetic is a behavioral biometric that involves using a person's sense of fondness for images. The identification of individuals using their visual aesthetic values as discriminatory features is an emerging domain of research. This paper introduces a novel method for aesthetic feature dimensionality reduction using gene expression programming. The proposed system is capable of using a tree-based genetic approach for feature recombination. Reducing feature dimensionality improves classifier accuracy, reduces computation runtime, and minimizes required storage. The results obtained on a dataset of 200 Flickr users evaluating 40,000 images demonstrate a 95% accuracy of identity recognition based solely on users' aesthetic preferences.

The Susceptibility of Human Melanoma Cells to Infection with the Leningrad-16 Vaccine Strain of Measles Virus.

Oncolytic viruses, including live attenuated measles virus (MV) vaccine strains, have recently been shown as promising therapeutic agents against human malignancies. In this study, the oncolytic potential of the attenuated vaccine strain Leningrad-16 (L-16) of MV was evaluated in a panel of human metastatic melanoma cell lines. The L-16 measles virus was shown to replicate within melanoma cells mediating direct cell killing of tumor cells, although all melanoma cell lines varied in regard to their ability to respond to L-16 MV infection, as revealed by the different pattern of the Interferon Stimulated Gene expression, cytokine release and mechanisms of cell death. Furthermore, the statistically significant L-16 measles virus related tumor growth inhibition was demonstrated in a melanoma xenograft model. Therefore, L-16 MV represents an appealing oncolytic platform for target delivery of therapeutic genes along with other attenuated measles virus strains.

Multidimensional evaluation of a radio frequency identification wi-fi location tracking system in an acute-care hospital setting.

Real-time locating systems (RTLS) have the potential to enhance healthcare systems through the live tracking of assets, patients and staff. This study evaluated a commercially available RTLS system deployed in a clinical setting, with three objectives: (1) assessment of the location accuracy of the technology in a clinical setting; (2) assessment of the value of asset tracking to staff; and (3) assessment of threshold monitoring applications developed for patient tracking and inventory control. Simulated daily activities were monitored by RTLS and compared with direct research team observations. Staff surveys and interviews concerning the system's effectiveness and accuracy were also conducted and analyzed. The study showed only modest location accuracy, and mixed reactions in staff interviews. These findings reveal that the technology needs to be refined further for better specific location accuracy before full-scale implementation can be recommended.

Multimodal biometric system using rank-level fusion approach.

In many real-world applications, unimodal biometric systems often face significant limitations due to sensitivity to noise, intraclass variability, data quality, nonuniversality, and other factors. Attempting to improve the performance of individual matchers in such situations may not prove to be highly effective. Multibiometric systems seek to alleviate some of these problems by providing multiple pieces of evidence of the same identity. These systems help achieve an increase in performance that may not be possible using a single-biometric indicator. This paper presents an effective fusion scheme that combines information presented by multiple domain experts based on the rank-level fusion integration method. The developed multimodal biometric system possesses a number of unique qualities, starting from utilizing principal component analysis and Fisher's linear discriminant methods for individual matchers (face, ear, and signature) identity authentication and utilizing the novel rank-level fusion method in order to consolidate the results obtained from different biometric matchers. The ranks of individual matchers are combined using the highest rank, Borda count, and logistic regression approaches. The results indicate that fusion of individual modalities can improve the overall performance of the biometric system, even in the presence of low quality data. Insights on multibiometric design using rank-level fusion and its performance on a variety of biometric databases are discussed in the concluding section.