Dummy molecularly imprinted polymers for class-selective extraction of amphetamine-type stimulants from alcoholic and nonalcoholic beverages.

Molecularly imprinted polymer was constructed for the first time through dummy imprinting strategy with homopiperonylamine as dummy template. The prepared dummy molecularly imprinted polymer (DMIP) showed high class selectivity towards the most popular amphetamine-type stimulants (ATSs) such as methamphetamine, amphetamine, 3,4-methylenedioxymethamphetamine, 3,4-methylenedioxy-amphetamine, and 3,4-methylenedioxy-N-ethylamphetamine with the imprinting factors of 2.280∼3.698 and selectivity factors of 1.654∼3.698. Moreover, ATSs could be rapidly adsorbed from water with the equilibrium time within 5 min. Hydrogen-bonding interaction between the amino groups of ATSs and carboxy on DMIP could be dominated adsorption mechanism. DMIP was employed as solid phase extraction (SPE) sorbents. Under the optimum extraction conditions, the method using DMIP-based SPE and high performance liquid chromatography-tandem mass spectrometry showed good linearity in the range of 0.025∼1.00 µmol L-1, good repeatability (RSD 4.8∼8.6%, n = 5) and low limits of quantification (0.007∼0.200 ng mL-1, S/N = 10). Satisfactory recoveries (72.5∼120%) with low RSD values (<10%) were obtained for all targets viz. spiked coke carbonated drinks, beer and cocktail. Compared with other commercial SPE sorbents, DMIP exhibited lower matrix effect (ME) for coke, beer and cocktail with ME values of 101∼124%, 75.8∼80.2% and 103∼128%, respectively. The obtained results suggested that the developed DMIP materials could be a potential candidate for pretreatment of ATSs in alcoholic and nonalcoholic beverages.

CAR NK-92 cells targeting DLL3 kill effectively small cell lung cancer cells in vitro and in vivo.

Small cell lung cancer (SCLC) is characterized by a high relapse rate, drug tolerance, and limited treatment choices. Chimeric antigen receptor (CAR)-modified NK cells represent a promising immunotherapeutic modality for cancer treatment. However, their potential applications have not been explored in SCLC. Delta-like ligand 3 (DLL3) has been reported to be overexpressed in SCLC and may be a rational target for CAR NK immunotherapy. In this study, we developed DLL3-specific NK-92 cells and explored their potential in the treatment of SCLC. A coculture of DLL3+ SCLC cell lines with DLL3-CAR NK-92 cells exhibited significant in vitro cytotoxicity and cytokine production. DLL3-CAR NK-92 cells induced tumor regression in an H446-derived pulmonary metastasis tumor model under a good safety threshold. The potent antitumor activities of DLL3-CAR NK-92 cells were observed in subcutaneous tumor models of SCLC. Moreover, obvious tumor-infiltrated DLL3-CAR NK-92 cells were detected in DLL3+ SCLC xenografts. These findings indicate that DLL3-CAR NK-92 cells might be a potential strategy for the treatment of SCLC.

Deep Subdomain Adaptation Network for Image Classification.

For a target task where the labeled data are unavailable, domain adaptation can transfer a learner from a different source domain. Previous deep domain adaptation methods mainly learn a global domain shift, i.e., align the global source and target distributions without considering the relationships between two subdomains within the same category of different domains, leading to unsatisfying transfer learning performance without capturing the fine-grained information. Recently, more and more researchers pay attention to subdomain adaptation that focuses on accurately aligning the distributions of the relevant subdomains. However, most of them are adversarial methods that contain several loss functions and converge slowly. Based on this, we present a deep subdomain adaptation network (DSAN) that learns a transfer network by aligning the relevant subdomain distributions of domain-specific layer activations across different domains based on a local maximum mean discrepancy (LMMD). Our DSAN is very simple but effective, which does not need adversarial training and converges fast. The adaptation can be achieved easily with most feedforward network models by extending them with LMMD loss, which can be trained efficiently via backpropagation. Experiments demonstrate that DSAN can achieve remarkable results on both object recognition tasks and digit classification tasks. Our code will be available at https://github.com/easezyc/deep-transfer-learning.

Common and specific altered amplitude of low-frequency fluctuations in Parkinson's disease patients with and without freezing of gait in different frequency bands.

Freezing of gait (FOG), a disabling symptom of Parkinson's disease (PD), severely affects PD patients' life quality. Previous studies found neuropathologies in functional connectivity related to FOG, but few studies detected abnormal regional activities related to FOG in PD patients. In the present study, we analyzed the amplitude of low-frequency fluctuations (ALFF) to detect brain regions showing abnormal activity in PD patients with FOG (PD-with-FOG) and without FOG (PD-without-FOG). As different frequencies of neural oscillations in brain may reflect distinct brain functional and physiological properties, we conducted this study in three frequency bands, slow-5 (0.01-0.027 Hz), slow-4 (0.027-0.073 Hz), and classical frequency band (0.01-0.08 Hz). We acquired rs-fMRI data from 18 PD-with-FOG patients, 18 PD-without-FOG patients, and 17 healthy controls, then calculated voxel-wise ALFF across the whole brain and compared ALFF among the three groups in each frequency band. We found: (1) in slow-5, both PD-with-FOG and PD-without-FOG patients showed lower ALFF in the bilateral putamen compared to healthy controls, (2) in slow-4, PD-with-FOG patients showed higher ALFF in left inferior temporal gyrus (ITG) and lower ALFF in right middle frontal gyrus (MFG) compared to either PD-without-FOG patients or healthy controls, (3) in classical frequency band, PD-with-FOG patients also showed higher ALFF in ITG compared to either PD-without-FOG patients or healthy controls. Furthermore, we found that ALFF in MFG and ITG in slow-4 provided the highest classification accuracy (96.7%) in distinguishing PD-with-FOG from PD-without-FOG patients by using a stepwise multivariate pattern analysis. Our findings indicated frequency-specific regional spontaneous neural activity related to FOG, which may help to elucidate the pathogenesis of FOG.

Multi-representation adaptation network for cross-domain image classification.

In image classification, it is often expensive and time-consuming to acquire sufficient labels. To solve this problem, domain adaptation often provides an attractive option given a large amount of labeled data from a similar nature but different domains. Existing approaches mainly align the distributions of representations extracted by a single structure and the representations may only contain partial information, e.g., only contain part of the saturation, brightness, and hue information. Along this line, we propose Multi-Representation Adaptation which can dramatically improve the classification accuracy for cross-domain image classification and specially aims to align the distributions of multiple representations extracted by a hybrid structure named Inception Adaptation Module (IAM). Based on this, we present Multi-Representation Adaptation Network (MRAN) to accomplish the cross-domain image classification task via multi-representation alignment which can capture the information from different aspects. In addition, we extend Maximum Mean Discrepancy (MMD) to compute the adaptation loss. Our approach can be easily implemented by extending most feed-forward models with IAM, and the network can be trained efficiently via back-propagation. Experiments conducted on three benchmark image datasets demonstrate the effectiveness of MRAN.

Transfer collaborative filtering from multiple sources via consensus regularization.

Collaborative filtering is one of the most successful approaches to build recommendation systems. Recently, transfer learning has been applied to recommendation systems for incorporating information from external sources. However, most existing transfer collaborative filtering algorithms tend to transfer knowledge from one single source domain. Rich information is available in many source domains, which can better complement the data in the target domain than that from a single source. However, it is common to get inconsistent information from different sources. To this end, we proposed a TRA nsfer collaborative filtering framework from multiple sources via C onsE nsus R egularization, called TRACER for short. The TRACER framework handles the information inconsistency with a consensus regularization, which enforces the outputs from multiple sources to converge. In addition, our algorithm is to learn and transfer knowledge at the same time while most of the traditional transfer learning algorithms are to learn knowledge first and then transfer it. Experiments conducted on two real-world data sets validate the effectiveness of the proposed algorithm.

Environmental risk factors for congenital heart disease in the Shandong Peninsula, China: a hospital-based case-control study.

BACKGROUND: In China, and in Shandong province, the proportionate contribution of birth defects to infant mortality has increased, and congenital heart disease (CHD) is now the most common cause of birth defects. The cause of approximately 90% of cases of congenital heart disease is multifactorial. Little is known about modifiable environmental risk factors or regional differences. We investigated putative environmental risk factors for congenital heart disease in the Shandong province of China in order to improve prevention of CHD. METHODS: We conducted a hospital-based 1:2 matched case-control study of 164 patients with congenital heart diseases and 328 controls, all of whom were retrospectively interviewed. Univariate and multivariate analyses were conducted to identify environmental risk factors for CHD. RESULTS: The environmental risk factors associated with CHD were mother's education level (odds ratio [OR], 0.31; 95% confidence interval [CI], 0.15-0.67), neonatal asphyxia or hypoxia (OR, 3.74; 95% CI, 1.25-11.18), number of previous pregnancies (OR, 2.68; 95% CI, 1.44-4.97), maternal upper respiratory tract infection (OR, 4.12; 95% CI, 1.56-10.85), maternal infection (OR, 7.98; 95% CI, 2.14-29.72), maternal B-mode ultrasound examination (OR, 4.05; 95% CI, 1.48-11.08), and maternal mental stress (OR, 3.93; 95% CI, 1.94-7.94) during early pregnancy. No significant interactions were observed among these factors. CONCLUSIONS: Augmenting maternal mental healthcare, obtaining regular health counseling and testing during pregnancy, preventing upper respiratory tract infections, limiting medication during early pregnancy, offering health promotion and health education to women of childbearing age (especially those with less formal education), and improving obstetric procedures and techniques may lower the occurrence of congenital heart disease.

[Study on the status and quality of sleep-related influencing factors in medical college students].

OBJECTIVE: To study the status and influencing factors on sleep quality in some medical college students. METHODS: Stratified sampling, pittsburgh sleep quality index (PSQI), self-evaluation depression scale (SDS), self-evaluation anxiety scale (SAS) and self-developed questionnaire of influencing factors on the quality of sleep in medical college students were used. Cumulative odds logistic model was performed to analyze the related factors on the quality of sleep. RESULTS: 19.17 percent of the medical college students showed poor quality of sleep and the difference between genders was not statistically significant (P > 0.05). Statistically significant (P < 0.05) difference was seen among different years of students and correlation was found between sleep quality and depression or anxiety (P < 0.0001). Factors influencing on the quality of sleep in medical college students would include: worry of sleep, irregular work/rest, worry on examination, stress, relationship with classmates, self-evaluated health condition, environments of the dormitory and late to bed. CONCLUSION: Influencing factors were identified and comprehensive measures should be taken to improve the quality of sleep.