Real-Time Action Recognition System for Elderly People Using Stereo Depth Camera.

Smart technologies are necessary for ambient assisted living (AAL) to help family members, caregivers, and health-care professionals in providing care for elderly people independently. Among these technologies, the current work is proposed as a computer vision-based solution that can monitor the elderly by recognizing actions using a stereo depth camera. In this work, we introduce a system that fuses together feature extraction methods from previous works in a novel combination of action recognition. Using depth frame sequences provided by the depth camera, the system localizes people by extracting different regions of interest (ROI) from UV-disparity maps. As for feature vectors, the spatial-temporal features of two action representation maps (depth motion appearance (DMA) and depth motion history (DMH) with a histogram of oriented gradients (HOG) descriptor) are used in combination with the distance-based features, and fused together with the automatic rounding method for action recognition of continuous long frame sequences. The experimental results are tested using random frame sequences from a dataset that was collected at an elder care center, demonstrating that the proposed system can detect various actions in real-time with reasonable recognition rates, regardless of the length of the image sequences.

HMM-Based Action Recognition System for Elderly Healthcare by Colorizing Depth Map.

Addressing the problems facing the elderly, whether living independently or in managed care facilities, is considered one of the most important applications for action recognition research. However, existing systems are not ready for automation, or for effective use in continuous operation. Therefore, we have developed theoretical and practical foundations for a new real-time action recognition system. This system is based on Hidden Markov Model (HMM) along with colorizing depth maps. The use of depth cameras provides privacy protection. Colorizing depth images in the hue color space enables compressing and visualizing depth data, and detecting persons. The specific detector used for person detection is You Look Only Once (YOLOv5). Appearance and motion features are extracted from depth map sequences and are represented with a Histogram of Oriented Gradients (HOG). These HOG feature vectors are transformed as the observation sequences and then fed into the HMM. Finally, the Viterbi Algorithm is applied to recognize the sequential actions. This system has been tested on real-world data featuring three participants in a care center. We tried out three combinations of HMM with classification algorithms and found that a fusion with Support Vector Machine (SVM) had the best average results, achieving an accuracy rate (84.04%).

Evaluation of Main Factors for Improvement of the Scandium Leaching Process from Russian Bauxite Residue (Red Mud) in Carbonate Media.

The carbonate leaching of scandium from the landfilled bauxite residue (red mud) of the Bogoslovsky Aluminum Plant (Russia) and samples of red mud (RM) after alkaline pretreatment has been investigated. The results of kinetic studies allowing to compare and evaluate the effectiveness of different conditions and intensification factors in the process of scandium leaching from RM in carbonate/bicarbonate media are presented. It was determined that for 2.0 mol L-1 Na2CO3 leaching solution ultrasonic treatment under gas (CO2) carbonation conditions in the pH range of 9.5-10.0 allows reducing the scandium leaching time by two times and reaching 40-45% scandium extraction. Leaching of RM in carbonate/bicarbonate media is accompanied by secondary processes (adsorption, hydrolysis, and coprecipitation) leading to a decrease in scandium extraction. The obtained results allow improving understanding of scandium chemical behavior in complex aqueous carbonate/bicarbonate systems and can be used for the optimization of the alternative carbonate process for scandium extraction from RM.

Review of MDA registers for Lymphatic Filariasis: Findings, and potential uses in addressing the endgame elimination challenges.

BACKGROUND: Lymphatic filariasis (LF) is endemic in Ghana, and the country has implemented the GPELF strategy since 2000 with significant progress made in the control of the disease. However, after several years of mass drug administration (MDA) implementation, there is persistent transmission in 17 of the 98 endemic districts in the country. Current approaches to surveillance are clearly unable to target untreated individuals and new strategies are required to address the endgame challenges to enhance LF elimination as a public health problem in endemic countries. Community registers are used during MDAs to enumerate community members, their age, gender, house numbers, and records of their participation in MDAs. These MDA registers represent an untapped opportunity to identify and characterize non-compliance and inform appropriate programmatic actions. In this study, we analyzed the data presented in the registers to assess the coverage and individuals' compliance in MDA. METHODS: The information in the MDA registers were assessed to verify the reported coverages obtained from the district. The community registers were obtained from the district health offices and the data from each individual record was entered into a database. A simple questionnaire was used to cross-check the participation of study participants in the 2017 MDA. The questionnaire solicited data on: participation in the 2017 MDA, reasons for not taking part in the MDA, adverse events experienced, what was done for the adverse events, and willingness to participate in subsequent MDAs. RESULTS: We found that 40.1% of the population in the registers missed at least one MDA in 3 years (2016-2018) and the majority of them were between 10-30 years of age. The results of the questionnaire assessment indicated that 13.8% of the respondents did not receive treatment in 2017 for various reasons, the most prominent among them being "absence/travel" (37.1%). Data in the registers were used to verify the treatment coverage for the years 2017 and 2018, and reviewed against the reported coverage obtained from the district. Significant differences between the reported and verified coverages were only observed in four communities. However, the assessment also revealed that the reported coverage was only accurate in 33.3% of cases. CONCLUSIONS: The MDA registers allow for the identification of eligible individuals who were not reached during any MDA round. Thus, the MDA registers could be utilized at the community and programme levels to identify missing and untreated individuals, appropriately address their non-compliance to MDA, and thereby improve MDA coverage in each implementation unit and monitor the progress towards elimination of LF. The challenges observed through the review of the registers also offer opportunities to improve the training given to the community drug distributors.