Application of Optimal-Jerk Trajectory Planning in Gait-balance Training Robot

To accommodate the gait and balance disorder of the elderly with age progression and the occurrence of various senile diseases, this paper proposes a novel gait balance training robot (G-Balance) based on a six degree-of-freedom parallel platform. Using the platform movement and IMU wearable sensors, two training modes, i.e., active and passive, are developed to achieve vestibular stimulation. Virtual reality technology is applied to achieve visual stimulation. In the active training mode, the elderly actively exercises to control the posture change of the platform and the switching of the virtual scene. In the passive training mode, the platform movement is combined with the virtual scene to simulate bumpy environments, such as earthquakes, to enhance the human anti-interference ability. To achieve a smooth switching of the scene, continuous speed and acceleration of the platform motion are required in some scenarios, in which a trajectory planning algorithm is applied. This paper describes the application of the trajectory planning algorithm in the balance training mode and the optimization of jerk (differential of acceleration) based on cubic spline planning, which can reduce impact on the joint and enhance stability.

Temporal face feature progression with cycle GAN

The aging process creates significant changes in the appearances of people’s faces. When compared to other causes of variation in face imaging, aging-related variation has specific distinct properties. Facial Aging variations, for example, is unique for each person; it occurs gradually and is significantly influenced by other characteristics including health, gender, and life-style. As a result, the proposed effort will use Generative Adversarial Networks to address these critical concerns (GANs). Generative Adversarial Networks (GAN’s) is made up of a generator and a discriminator network. The generator model generates images that a discriminator model analyses to determine if they are real or fake. This paper provides a Temporal Face Feature Progressive framework with Cycle GAN, which maintains the initial appearance and identity in the elderly aspect of their facial structure. To address aging concerns, our goal is to transform an initial age category image into a targeted age with age progression. We show that our temporal face features progressive cycle GAN learns and transfers facial traits from the source group to the targeted group by training various images. The IMDB-WIKI Face dataset has been used to obtain the results for the same.

Extreme Learning Machine-Based Age-Invariant Face Recognition With Deep Convolutional Descriptors

The principal intention of this paper is to study face recognition across age progression at two levels: feature extraction and classification. In other words, this work aims to prove the benefit of replacing the Softmax layer of the Deep-Convolutional Neural Networks (CNN) by Extreme Learning Machine (ELM) classifier based on deep features computed from fully-connected layer of pre-trained AlexNet CNN model, in a context of age-invariant face recognition. Experimental results indicate that the ELM classifier combined with feature extracted by the pre-trained AlexNet CNN model worked effectively for face recognition across age progression. As significant highest mean accuracy rates are always obtained using ELM classifier. These results are more significant, following a 95% confidence level hypothesis test.

Ovarian Aging: Role of Pituitary-Ovarian Axis Hormones and ncRNAs in Regulating Ovarian Mitochondrial Activity

The number of mitochondria in the oocyte along with their functions (e.g., energy production, scavenger activity) decline with age progression. Such multifaceted functions support several processes during oocyte maturation, ranging from energy supply to synthesis of the steroid hormones. Hence, it is hardly surprising that their impairment has been reported in both physiological and premature ovarian aging, wherein they are crucial players in the apoptotic processes that arise in aged ovaries. In any form, ovarian aging implies the progressive damage of the mitochondrial structure and activities as regards to ovarian germ and somatic cells. The imbalance in the circulating hormones and peptides (e.g., gonadotropins, estrogens, AMH, activins, and inhibins), active along the pituitary-ovarian axis, represents the biochemical sign of ovarian aging. Despite the progress accomplished in determining the key role of the mitochondria in preserving ovarian follicular number and health, their modulation by the hormonal signalling pathways involved in ovarian aging has been poorly and randomly explored. Yet characterizing this mechanism is pivotal to molecularly define the implication of mitochondrial dysfunction in physiological and premature ovarian aging, respectively. However, it is fairly difficult considering that the pathways associated with ovarian aging might affect mitochondria directly or by altering the activity, stability and localization of proteins controlling mitochondrial dynamics and functions, either unbalancing other cellular mediators, released by the mitochondria, such as non-coding RNAs (ncRNAs). We will focus on the mitochondrial ncRNAs (i.e., mitomiRs and mtlncRNAs), that retranslocate from the mitochondria to the nucleus, as active players in aging and describe their role in the nuclear-mitochondrial crosstalk and its modulation by the pituitary-ovarian hormone dependent pathways. In this review, we will illustrate mitochondria as targets of the signaling pathways dependent on hormones and peptides active along the pituitary/ovarian axis and as transducers, with a particular focus on the molecules retrieved in the mitochondria, mainly ncRNAs. Given their regulatory function in cellular activities we propose them as potential diagnostic markers and/or therapeutic targets.

Cardiovascular Risk Assessment in Hypertensive patients: A Perspective Observative Study

Congestive heart disease (CHD) is considered to be the leading cause of mortality and morbidity in both gender groups in developed and developing countries. Hypertension is one of the main mortality risks and is attributed to over 45% of all deaths from CHD. The main objective of our work was to evaluate cardiovascular risk in hypertensive patients attending a tertiary care hospital in the Khammam region. The study was a prospective observational study conducted over an 8-month period from June 2019 to January 2020. 192 subjects were selected based on the inclusion criteria. CVD risk was assessed using Q Risk 3 software and the results were presented as CVD risk and relative risk. The same number of men and women (96) was selected in the study to evaluate the influence of gender on CVD risk. Other risk factors such as BMI, marital status, literacy rate, occupation, physical activity and lifestyle were assessed to determine CVD risk. Abnormal HTN values were found in 66 men and 63 women. Age progression was found to be an important factor in CVD risk in both men and women. Social status and literacy rates in patients over 50 have also been found to cause CVD risk. Our study showed that physical inactivity, eating habits, obesity, smoking, alcohol and hypertension had a direct effect on cardiovascular risk.

AgeFlow: Conditional Age Progression and Regression with Normalizing Flows

Age progression and regression aim to synthesize photorealistic appearance of a given face image with aging and rejuvenation effects, respectively. Existing generative adversarial networks (GANs) based methods suffer from the following three major issues: 1) unstable training introducing strong ghost artifacts in the generated faces, 2) unpaired training leading to unexpected changes in facial attributes such as genders and races, and 3) non-bijective age mappings increasing the uncertainty in the face transformation. To overcome these issues, this paper proposes a novel framework, termed AgeFlow, to integrate the advantages of both flow-based models and GANs. The proposed AgeFlow contains three parts: an encoder that maps a given face to a latent space through an invertible neural network, a novel invertible conditional translation module (ICTM) that translates the source latent vector to target one, and a decoder that reconstructs the generated face from the target latent vector using the same encoder network; all parts are invertible achieving bijective age mappings. The novelties of ICTM are two-fold. First, we propose an attribute-aware knowledge distillation to learn the manipulation direction of age progression while keeping other unrelated attributes unchanged, alleviating unexpected changes in facial attributes. Second, we propose to use GANs in the latent space to ensure the learned latent vector indistinguishable from the real ones, which is much easier than traditional use of GANs in the image domain. Experimental results demonstrate superior performance over existing GANs-based methods on two benchmarked datasets. The source code is available at https://github.com/Hzzone/AgeFlow.