Aim: The research described is intended to give a description of articulation dynamics as a correlate of the kinematic behavior of the jaw-tongue biomechanical system, encoded as a probability distribution of an absolute joint velocity. This distribution may be used in detecting and grading speech from patients affected by neurodegenerative illnesses, as Parkinson Disease. Hypothesis: The work hypothesis is that the probability density function of the absolute joint velocity includes information on the stability of phonation when applied to sustained vowels, as well as on fluency if applied to connected speech. Methods: A dataset of sustained vowels recorded from Parkinson Disease patients is contrasted with similar recordings from normative subjects. The probability distribution of the absolute kinematic velocity of the jaw-tongue system is extracted from each utterance. A Random Least Squares Feed-Forward Network (RLSFN) has been used as a binary classifier working on the pathological and normative datasets in a leave-one-out strategy. Monte Carlo simulations have been conducted to estimate the influence of the stochastic nature of the classifier. Two datasets for each gender were tested (males and females) including 26 normative and 53 pathological subjects in the male set, and 25 normative and 38 pathological in the female set. Results: Male and female data subsets were tested in single runs, yelding equal error rates under 0.6% (Accuracy over 99.4%). Due to the stochastic nature of each experiment, Monte Carlo runs were conducted to test the reliability of the methodology. The average detection results after 200 Montecarlo runs of a 200 hyperplane hidden layer RLSFN are given in terms of Sensitivity (males: 0.9946, females: 0.9942), Specificity (males: 0.9944, females: 0.9941) and Accuracy (males: 0.9945, females: 0.9942). The area under the ROC curve is 0.9947 (males) and 0.9945 (females). The equal error rate is 0.0054 (males) and 0.0057 (females). Conclusions: The proposed methodology avails that the use of highly normalized descriptors as the probability distribution of kinematic variables of vowel articulation stability, which has some interesting properties in terms of information theory, boosts the potential of simple yet powerful classifiers in producing quite acceptable detection results in Parkinson Disease.

Currently, the use of biometric identification, automated or semiautomated, is a reality. For this reason, the number of attacks has increased in such systems. One of the most common biometric attacks is the presentation attack (PA) because it is relatively easy to perform. Automated border control (ABC) is a clear target for phishers. Concerning biometric attacks, morphing is one of the most threatening attacks because authentication systems are usually unable to correctly detect them. In this attack, a fake face is generated with the morphing and blending of two different subjects (genuine and phisher), and the image result is stored in the passport. These attacks can generate risky situations in cases of border crossings where an ABC system should perform identification tasks. This research work proposes a de-morphing architecture that is founded on a convolutional neural network (CNN) architecture. This technique is based on the use of two images: the potentially morphed image stored in the passport, and the snapshot of the person located in the ABC system. The goal of the de-morphing process is to unravel the chip image. If the chip image is a morphed one, the revealing process between the in vivo image and the morphed chip image will return a different facial identity to the person located in the ABC system, and the impostor will be uncovered in situ. If the chip image is a non-morphing image, the resulting image will be similar to a genuine passenger. Therefore, the information obtained is considered at the border crossing. The equal error rate (EER) achieved is very low compared to the literature values published to date. The accomplished outcomes endorse a robust method that provides high accuracy rates without taking into account the quality of images used. This key point is crucial to plausible deployment plans in areas such as ABC.

Speech articulation is produced by the movements of muscles in the larynx, pharynx, mouth and face. Therefore speech shows acoustic features as formants which are directly related with neuromotor actions of these muscles. The first two formants are strongly related with jaw and tongue muscular activity. Speech can be used as a simple and ubiquitous signal, easy to record and process, either locally or on e-Health platforms. This fact may open a wide set of applications in the study of functional grading and monitoring neurodegenerative diseases. A relevant question, in this sense, is how far speech correlates and neuromotor actions are related. This preliminary study is intended to find answers to this question by using surface electromyographic recordings on the masseter and the acoustic kinematics related with the first formant. It is shown in the study that relevant correlations can be found among the surface electromyographic activity (dynamic muscle behavior) and the positions and first derivatives of the first formant (kinematic variables related to vertical velocity and acceleration of the joint jaw and tongue biomechanical system). As an application example, it is shown that the probability density function associated to these kinematic variables is more sensitive than classical features as Vowel Space Area (VSA) or Formant Centralization Ratio (FCR) in characterizing neuromotor degeneration in Parkinson’s Disease.

Gender classification of human faces is a trending topic and a remarkable biometric task. This research area has useful applications in several fields, such as automated border control (ABC) and forensic work. There are many approaches to gender classification in the literature; the classical approaches usually use real faces. Although good performances have been achieved, data collection remains a problem. Additionally, the privacy of individuals must be included in many existing works. These drawbacks can be overcome by using fake faces. Recently, the creation of a robust fake face corpus using machine learning has become possible. Our main contribution in the present paper is to experimentally investigate the ability of an artificial deepfake corpus to be a substitute for real corpora in facial gender classification tasks. We propose a deep learning-based approach using convolutional neural networks trained with fake faces and tested on real faces. By exploiting artificial faces, data collection obstacles are resolved for the training step, and privacy is highly preserved. Four classifiers based on popular convolutional neural network architectures were implemented. In the test phase, we used faces of real identities extracted from well-known experimental databases such as Face Recognition Technology (FERET), Faculdade de Engenharia Industrial (FEI) faces, Face Recognition and Artificial Vision (FRAV) and Labeled Faces in the Wild (LFW). The results achieved are very promising. We obtained high accuracy rates and low EER scores. They are similar to those of research works using real faces. As a result of this work, we propose a gender-labeled deepfake facial dataset containing more than 200k deepfake corpora that we will make available upon request for research purposes.