Search

Remote photoplethysmography (rPPG) is an unobtrusive solution to heart rate monitoring in drivers. However, disturbances that occur during driving such as driver behavior, motion artifacts, and illuminance variation complicate the monitoring of heart rate. Faced with disturbance, one commonly used assumption is heart rate periodicity (or spectrum sparsity). Several methods improve stability at the expense of tracking sensitivity for heart rate variation. Based on statistical signal processing (SSP) and Monte Carlo simulations, the outlier probability is derived and ADaptive spectral filter banks (AD) is proposed as a new algorithm which provides an explicable tuning option for spectral filter banks to strike a balance between robustness and sensitivity in remote monitoring for driving scenarios. Moreover, we construct a driving database containing over 23 hours of data to verify the proposed algorithm. The influence on rPPG from driver habits (both amateurs and professionals), vehicle types (compact cars and buses), and routes are also evaluated. In comparison to state-of-the-art rPPG for driving scenarios, the mean absolute error in the Passengers, Compact Cars, and Buses scenarios is 3.43, 7.85, and 5.02 beats per minute, respectively. Moreover, AD also won the top third place in the first challenge on remote physiological signal sensing (RePSS) with relative low computational complexity.

Background: Papillary thyroid carcinoma (PTC) patients with anterior extrathyroidal extension (ETE) involving the strap muscle have a relatively better prognosis than those with posterior gross ETE involving the recurrent laryngeal nerve. Whether prophylactic central-compartment lymph node dissection (CLND) should be performed in PTCs with only strap muscle invasion (SMI) is still unclear.Methods: A retrospective cohort study was conducted in clinical N0 (cN0) PTC patients with SMI who underwent thyroid surgery from 2009 to 2017. A total of 152 patients were included, and predictive factors of central-compartment lymph node metastasis (CLNM) were determined.Results: Among the 281 PTCs patients with SMI, 152 (51.1%) did not clinically present with lymph node metastasis. Microscopic CLNM was identified in 77 (50.7%) cN0 PTC patients with SMI. According to the univariate and multivariate analyses, male patients and those aged 40 years (odds ratio [OR] = 6.22 [95% confidence interval (CI), 1.43–27.10], p = 0.02 vs. OR = 9.94 [95% CI, 2.79–35.44], p = 0.00). The CLNM positive rate of male patients aged

Cisplatin‑induced cytotoxicity, such as nephrotoxicity, neurotoxicity and ototoxicity, restricts the clinical application of this compound. Panax notoginseng Saponins (PNS) exhibit potent free radical scavenging and antioxidant activity. PNS have been demonstrated to reduce cisplatin‑induced nephrotoxicity and neurotoxicity. The present study investigated the ability of PNS to protect the auditory HEI‑OC1 cell line against ototoxicity induced by cisplatin. PNS induced activation of the AKT/nuclear factor erythroid 2‑related factor 2 (Nrf2) signaling pathway. Following pretreatment with PNS, HEI‑OC1 cells were treated with cisplatin and cultured for 24 h. The viability of HEI‑OC1 cells was examined using a Cell Counting Kit‑8 assay. Double staining analysis was used to measure cell apoptosis. The ability of PNS to reduce reactive oxygen species (ROS) levels was assessed by flow cytometry. The levels of phosphorylated (p)‑AKT, heme oxygenase 1 (HO‑1), NAD(P)H quinone dehydrogenase 1 (NQO1), glutamate‑cysteine ligase catalytic (GCLC) and Nrf2 were measured by western blotting. HEI‑OC1 cells that were pretreated with PNS exhibited significantly increased cell viability compared with that noted in cells treated only with cisplatin. In addition, PNS suppressed the induction of apoptosis and ROS production following cisplatin treatment. The upregulation of NQO1, HO‑1 and GCLC expression in PNS‑pretreated cells was associated with p‑AKT levels and the activation of Nrf2. These findings suggested that PNS protected auditory cells against ototoxicity induced by cisplatin by activating AKT/Nrf2 signaling. PNS may serve as a potential candidate in regulating cisplatin‑induced cytotoxicity.

OBJECTIVE: To study the clinical features of coronavirus disease 2019 (COVID-19) in children aged < 18 years. METHODS: A retrospective analysis was performed from the medical data of 23 children, aged from 3 months to 17 years and 8 months, who were diagnosed with COVID-19 in Jiangxi, China from January 21 to February 29, 2020. RESULTS: Of the 23 children with COVID-19, 17 had family aggregation. Three children (13%) had asymptomatic infection, 6 (26%) had mild type, and 14 (61%) had common type. Among these 23 children, 16 (70%) had fever, 11 (48%) had cough, 8 (35%) had fever and cough, and 8 (35%) had wet rales in the lungs. The period from disease onset or the first nucleic acid-positive detection of SARS-CoV-2 to the virus nucleic acid negative conversion was 6-24 days (median 12 days). Of the 23 children, 3 had a reduction in total leukocyte count, 2 had a reduction in lymphocytes, 2 had an increase in C-reactive protein, and 2 had an increase in D-dimer. Abnormal pulmonary CT findings were observed in 12 children, among whom 9 had patchy ground-glass opacities in both lungs. All 23 children received antiviral therapy and were recovered. CONCLUSIONS: COVID-19 in children aged < 18 years often occurs with family aggregation, with no specific clinical manifestation and laboratory examination results. Most of these children have mild symptoms and a good prognosis. Epidemiological history is of particular importance in the diagnosis of COVID-19 in children aged < 18 years.

Driver’s physiological state is highly correlated to the traffic safety. An affordable and convenient way to monitor driver’s physiological state is remote Photoplethysmography (rPPG). Earlier algorithms achieved high accuracy on measuring rPPG signals in stationary case. But in real cases, such as driving, rPPG signals might be corrupted with interference. To obtain higher Signal-to-Noise-Ratio (SNR) rPPG signals, three algorithms are proposed. The PCA spectral subtraction (PCA-SS) considers the spectrum of the environmental noise and utilizes the energy subtraction to reduce the noise. The machine learning methods, convolution autoencoder (CAE) and multi-channel convolution autoencoder (Multi-CAE), are adopted in order to enhance the rPPG signal. The test data we used are 187 videos recorded in stationary case, passenger case, and real driving situation. In driving situation, the Multi-CAE method, in comparison with the original method provided by W. Wang et al. [1] and G. De Haan et al. [2], achieves 33% & 35% reduction in MAE, RMSE respectively, and 11% improvement in success rate [3].

Facial action unit detection (FAUD) is aimed to detect subtle facial motions, known as action units (AUs), induced by the traction or towing of facial muscles. In recent years, FAUD has become an attractive task since these slight changes on face reveal some cues of emotions which can then be utilized to infer people’s affective. Many prior works are developed based on physical features of AUs, such as regional occurrence and temporal continuity, but few of them concern the negative impact of individual differences, causing an unrobust result in variant subjects and environments. To deal with the problem, a contrastive feature learning method is proposed to make a convolutional neural network (CNN) learn to extract the contrastive feature which is the difference between the features of a neutral face image and an AU-occurred face image. In this way, the individual information will be mitigated so that it becomes easier to detect AUs according to the features. A great disparity between the number of positive and negative samples, named as data imbalance, makes a serious interference in most detection problems, including FAUD. A lower occurrence rate of an AU leads a more severe data imbalance problem, which results in a biased result for the AU detection. Therefore, the class-weighted loss is proposed to change the weight-ratio between the positive and negative samples so as to encourage the learning of the positive samples and ease that of negative samples. Two widely used databases, BP4D and DISFA, are adopted to be the benchmark of the performance testing. In the experiment, it shows that the proposed method performs well both in accuracy and speed comparing with the state-of-the-art approaches.

The structures of phosphinidene germylenoid HP=GeLiF were studied for the first time by using DFT (density functional theory) calculations. The geometries were optimized at the B3LYP/6-311+G (d, p) level at first and then the single-point energies were calculated at QCISD/6-311++G (d, p) level. Theoretical calculations predicted that HP=GeLiF has two equilibrium structures, the p-complex (1) and the three-membered-ring (2) structures, in which structure 1 has the lower energy and is more stable than 2. To exploit the reactivity of HP=GeLiF, the cycloaddition reaction of 1 and ethylene was investigated at the same level of theory. From the potential energy profile, we predicted that the cycloaddition reaction has one dominant reaction pathway. The calculated result shows that the dominant reaction pathway is a [2 + 2] cycloaddition reaction which is the interaction of two π bonds in HP=GeLiF and ethylene molecule, and a four-membered-ring P-heterocyclic germylene is formed. Since sp3 hybridization of Ge atom in this four-membered-ring germylene, it may further react with another ethylene and finally forming a spiro-Ge-heterocyclic compound involving phosphorus. This means that this reaction involves a [2 + 2] cycloaddition as the initial step, and then a [2 + 1] cycloaddition carried out.

In a society with aging population, the demand for electric wheelchairs is growing with the advancement of automation. However, many accidents have occurred due to the misjudgment of the slope angle and wheelchair speed while the wheelchair is traveling on ramps. This research employs the light electronic assistance pal compact motor package to reduce the weight and size of conventional electric wheelchairs. The modular design of proposed uphill controller and ramp detection functions allows users to easily select and incorporate only the functions they need. This paper proposes a ramp detection model implemented using the deep learning algorithm with CNN-4 structure to analyze depth image data. The model's recognition time of each video frame is 11 times faster than that of the AlexNet and GoogleNet. The uphill safety controller is designed as an adaptive network-based fuzzy inference system with Q-learning. The safe speed is automatically calculated according to the angle obtained from slope classification and revised in real-time during the slope driving to prevent the user from moving towards the dangerous ramp or rolling back due to inadequate speed. The accuracy of ramp detection is further increased by 5% to 97.1% due to assistance from the voting system processing and the gyroscope output data. The 5° ramp experiment of our uphill controller with ramp classification takes 20 s to complete the slope driving which is 23% faster than the controller without ramp detection. The energy consumption is also one half less than the experiment without uphill detection.

In the future, the population’s aging problem will become increasingly serious. For medical care, action-assisted vehicles are one of the solutions. A wheelchair is a common mobility assistance equipment, and the safety of the user is an important issue. For the ordinary wheelchairs, most of them use a fixed hand brake or disk brake to allow the users to stop the wheelchair. However, improper use of brakes tends to cause slippage, and there is no concept of the anti-lock braking system (ABS) for the wheelchair. Therefore, this paper applies this idea to intelligent wheelchairs with the aim of enhancing the safety of users in wheelchairs. The ABS architecture includes the core algorithms of the adaptive fuzzy-neural inference system and the friction coefficient estimation system. The friction coefficient estimation system uses a particle filter to quickly adapt to a non-linear state and unknown environment. The system provides more accurate braking control for the ABS according to the range of the friction coefficient based on the pavement type. In ABS, it uses the gyroscope to detect the acceleration and wheelchair angle information and then calculates the parameters of the wheelchair. The user clicks the stop command on the wheelchair to activate the brake system to achieve a simple and efficient mode of operation. It can efficiently reduce the braking time and braking distance and enhance the riding safety of wheelchairs.

Automated facial expression recognition can greatly improve the human–machine interface. The machine can provide better and more personalized services when it knows the human’s emotion. This kind of improvement is an important progress in this artificial intelligence era. Many deep learning approaches have been applied in recent years due to their outstanding recognition accuracy after training with large amounts of data. The performance is limited, however, by the specific environmental conditions and variations in different persons involved. Hence, this paper addresses the issue of how to customize the generic model without label information from the testing samples. Weighted Center Regression Adaptive Feature Mapping (W-CR-AFM) is mainly proposed to transform the feature distribution of testing samples into that of trained samples. By means of minimizing the error between each feature of testing sample and the center of the most relevant category, W-CR-AFM can bring the features of testing samples around the decision boundary to the centers of expression categories; therefore, their predicted labels can be corrected. When the model which is tuned by W-CR-AFM is tested on extended Cohn-Kanade (CK+), Radboud Faces database, and Amsterdam dynamic facial expression set, our approach can improve the recognition accuracy by about 3.01%, 0.49%, and 5.33%, respectively. Compared to the competing deep learning architectures with the same training data, our approach shows the better performance.

Medical fields have seen increasing attention being given to image based heart rate measurement in recent years. One of the major limitations is motion artifacts of subject’s head. Although there have been many studies focusing on signal extraction using different parameters and models, the development of frequency domain analysis is emerging slowly and moving in many directions. In the field of contact photoplethysmography (PPG), recent studies employed the acceleration signals to assist their spectral peak tracking algorithms. Inspired by the development of contact PPG, we are proposing a motion resistant spectral peak tracking (MRSPT) framework which eliminates the motion artifacts by integrating facial motion signals. The effectiveness of MRSPT coupled with the optimal image-based PPG (iPPG) signal has been tested against the state-of-the-art spectral peak tracking algorithms, multi-channel spectral matrix decomposition (MC-SMD), and the maximum peak selection coupled with optimal iPPG signal (Optimal MPS). Compared with MC-SMD and Optimal MPS, MRSPT uplifts the success rate-10 (success rate-5), the probability in which the absolute error is below ten (five) beats per mins, from 54.7% (36.3%) with MC-SMD and 73.0% (61.3%) with Optimal MPS to 90.7% (75.7%) with MRSPT in motion scenarios where subject moves arbitrarily with different distance or lighting. MRSPT also enhances the success rate-10 (success rate-5) from 40.7% (26.3%) with MC-SMD and 57.4% (45.7%) with Optimal MPS to 73.4% (58.4%) with MRSPT in all seven motion conditions including driving and running. Averagely, the success rate-five of Optimal MRSPT surpass the success rate-10 of both Optimal MPS and MC-SMD.

e21081 Background: Immune checkpoint inhibitors (ICIs) have improved clinical outcomes in non-small cell lung cancer (NSCLC), whereas they also can cause various adverse events (AEs). We reported the incidences, spectrum and severity of AEs in real-life practice after the approval of ICIs in Chinese NSCLC patients. Methods: In this single center, retrospective study, anonymized electronic medical record data were collected from the dedicated lung cancer immunotherapy outpatient clinic of Guangdong Provincial People’s Hospital through the LinkDoc database. The patients (≥18 years old) with pathologically diagnosed metastatic NSCLC who receiving ICIs between 1 June 2019 and 31 December 2020 were included. The immune-related adverse events (irAEs) were diagnosed and assessed by oncologists. All AEs were classified and graded according to the National Cancer Institute Common Toxicity Criteria for Adverse Events (CTCAE V5.0). Results: A total of 203 patients with a median age of 60.5 years were included in this study. Most patients were male (81.8%), smokers (49.8%), non-squamous histological type (51.7%). 61.1% of patients visited the outpatient clinic more than once. Pembrolizumab and Nivolumab were given to the majority of patients (80.8%). 114 patients (56.2%) received single-agent and 89 patients (43.8%) received combination therapy. The median duration of ICIs therapy was 271.2 days. In our study, 68.0% of patients experienced any-grade AEs, and 19.7% experienced grade≥3 AEs. The most commonly AEs were abnormal laboratory findings (57.1%), blood and lymphatic system disorders (25.1%), and respiratory disorders (19.7%). All of 93 patients (45.8%) experienced irAEs and 6 patients (8.4%) developed grade≥3 irAEs. The most frequent irAEs included blood and lymphatic system disorders (25.1%), hepatobiliary disorders (17.7%), and cardiac disorders (9.9%). Cough (3.4%), sputum (2.5%), and pain (2.0%) were the most frequent AEs which received treatment. Conclusions: Our results showed 68.0% and 45.8% patients experienced AEs and irAEs, respectively. The AEs and irAEs were primarily low grade. This study also demonstrated the information of AEs could be well collected and managed through our dedicated lung cancer immunotherapy outpatient clinic. [Table: see text]

In present paper, the insertion and H2 elimination reactions of H2GeFMgF germylenoid with RH (R = Cl, SH, PH2) were investigated by means of B3LYP and QCISD calculation methods. One transition state and one intermediate were found along the potential energy surface in each insertion reaction, while for the H2 elimination reactions, only one transition state was found between the reactants and products in each reaction process. Both for the insertion and H2 elimination reactions, RH reactivity increases in the following order: H–Cl > H–SH > H–PH2. The insertion and H2 elimination reactions were compared, and the results demonstrated that the H2 elimination should be more favorable than the corresponding insertion. The solvent effects on these two types of reactions were considered. The calculated results indicated that the solvents could accelerate the reactions by reducing their barrier heights.

OBJECTIVES Erectile dysfunction (ED) is a problem commonly encountered by patients on methadone maintenance therapy (MMT). This study aimed to assess the prevalence of ED among this group of patients along with its risk factors and association with quality of life (QOL). METHODS Male patients on MMT in a tertiary hospital in Malaysia were included in the study. A total of 134 patients with sexual partners were assessed for ED using the International Index of Erectile Function. Patients were assessed for substance use using Opiate Treatment Index (OTI) and depression using the Malay version of the self-rated Montgomery-Asberg Depression Rating Scale (MADRS-BM). QOL was evaluated using World Health Organisation Quality of Life (WHOQOL)-BREF. RESULTS The prevalence of ED among patients on MMT was 67%, with 26.1% having mild ED, 30.4% having mild-to-moderate ED, 7.0% having moderate ED, and 17.2% having severe ED. Patients with depression were 4 times more likely to have ED compared with patients without depression, whereas increasing age significantly correlated with the severity of ED. Having ED predicted a poorer QOL in the social relationships domain. CONCLUSION Depression is highly associated with ED, which negatively influences the social aspect of QOL among patients on methadone maintenance therapy.

Remote-photoplethysmography (rPPG) is a contactless method which can measure Blood Volume Pulse (BVP) and several vital signals from human faces. Until now, most existing techniques depend on frequency analysis and regard the signal component with largest energy as Heart Rate (HR). Nonetheless, the property, that frequency analysis could not identify the instantaneous HR changes restricts the applications on Heart Rate Variability (HRV) analysis and Cardiac Arrhythmia Detection. Consequently, this paper proposes a time domain systolic peak selection algorithm. It reaches 89% Suc5 in stationary case and 90% Suc10 in indoor scenario which are neck and neck with frequency domain analysis. At the same time, the comparison method, Rei-En, gets Suc10 lower than 80% in both situations. We investigate the influence of dicrotic notch and utilize a concise method to prevent wrong Inter-beat interval (IBI) detection for HR and HRV estimation. Benefiting from the characteristic of time domain analysis, the first HR output could be estimated within five seconds, which is much shorter than seventeen seconds in frequency domain analysis. The results suggest that the proposed algorithm improves instantaneity and provides faster first output of rPPG in HR, HRV estimation.

This paper specifically focuses on how to build a Chinese facial expression database collecting the facial expressions of college students and describes a strategy to develop an automatically detecting technique for academic emotions to support teachers making better decisions in blended and digital learning environments. There are some famous worldwide databases of facial emotion expressions, e.g., Amsterdam Dynamic Facial Expression Set (ADFES), Montreal set of facial displays of emotion, or Brazillian FEI database. Their major collections are full facial expression of western people with very limited Asian or Chinese faces. Because some emotion facial expressions might be culturally bounded, it arises the necessity to develop a Chinese facial expression database as a critical step to develop an automatically facial emotion expression dictating technique with high accuracy.

To handle the customer distribution in the certain areas, crowd counting is necessary for such applications, which is a labor-intensive work for human. Therefore, an automatic crowd counting system is in great demand, but it is still a challenging problem since the human heads and bodies are usually highly overlapping in crowd images. In this paper, a counting-by-regression framework is employed. The human head is modeled as a Guassian distribution. With a crowd density map estimator, the head count can be obtained by integrating over the density map. Most existing approaches only apply density map regression for training a density map estimator, but it is hard to find a suitable training parameters to train a good one; actually, the head count is overestimated easily. To mitigate this problem, counting regression is combined with density map regression. A deeper and lighter fully convolutional network (FCN) is designed to be a crowd density map estimator. The input and output size of the FCN are the same. After training by the proposed method, our model is more competitive comparing with others. The parameter quantity of the model is the lowest, and it needs the least inference time.

In this paper we investigate the latest vision-based method for systolic blood pressure (SBP) and diastolic blood pressure (DBP) measurement. However, constantly blood pressure supervision needs sufficient medical equipment and may require the potential patients to tie a cuff, which is extremely inconvenient for them. What's more, continuously blood pressure measuring requires the patients to stay in the hospital and professional personnel to stand by. From the research before, we have learned that photoplethysmography (PPG) can be used to measure the blood pressure, which is known as cuffless blood pressure measurement. However, for the neonate and patients with empyrosis, photoplethysmography measuring device is still less practical and restricted in use due to the necessary contact for it to measure the systolic and diastolic blood pressure. Certain level of discomfort is still unavoidable with the use of PPG. We thus focus on remote PPG (rPPG); with green red difference (GRD) and Euler video magnification (EVM) and finite impulse response (FIR) bandpass filters, we are able to recover PPG signals from remote photoplethysmography. We propose a feature extraction measuring methods which yields a root mean square error for SBP as 11.22 mmHg and 7.83 mmHg for pulse pressure (PP) combined with the ANN model. For comparison, we've also used K nearest neighbor (KNN) and deep belief network-deep neural network (DBN-DNN).

A theoretical model for damage evolution and fatigue life prediction of Shape Memory Alloy (SMA) under low cycle fatigue is proposed in this work. Based on the theory of continuous damage mechanics and the obtained constitutive model of the SMAs, a fatigue cumulative damage model for SMAs is established, in which the threshold of damage initiation is defined, and the damage degree of SMA under different cycles is given. By using such model, the stress-strain relationship for the SMAs considering the damage under different cycles is given, and the different failure life of SMAs under different stress amplitudes is finally predicted. The results show that after the damage occurs, the hysteretic curves of the material become smaller, and the fatigue life decreases with the increasing of stress amplitudes.