Search

Background: Breast cancer is the most widespread malignant tumor worldwide. Single-cell sequencing technology offers novel insights and methods to understand the onset, progression, and treatment of tumors. Nevertheless, there is currently an absence of a thorough and unbiased report on the comprehensive research status of single-cell sequencing in breast cancer. This study seeks to summarize and quantify the dynamics and trends of research on breast cancer single-cell sequencing by bibliometric analysis. Methods: Research articles and reviews related to breast cancer single-cell sequencing were selected from the WoSCC database. Visualization of data regarding countries, institutions, authors, references, and keywords was performed by CiteSpace and VOSviewer software. Results: 583 articles and reviews were analyzed in this study. The quantity of publications related to breast cancer single-cell sequencing has been increasing annually. These studies originate from 302 institutions in 46 countries, with YMAX S WICHA producing the most publications and WANG Y being the most cited author. Nature Communications is the most researched journal, while Nature holds the highest number of citations. These journals predominantly cover topics in the molecular/biological/immunological fields. Moreover, an analysis of reference and keyword bursts revealed that current research trends in this area are primarily centered on “clonal evolution,” “tumor microenvironment,” and “immunotherapy.” Conclusion: Breast cancer single-cell sequencing is a rapidly growing area of scientific interest. Future research requires more frequent and in-depth collaborations among countries, institutions, and authors. Furthermore, “clonal evolution,” “tumor microenvironment,” and “immunotherapy” are likely to become major focal points in upcoming research on breast cancer single-cell sequencing.

Abstract Objective To evaluate RSF and Cox models for mortality prediction of hemorrhagic stroke (HS) patients in intensive care unit (ICU). Methods In the training set, the optimal models were selected using five-fold cross-validation and grid search method. In the test set, the bootstrap method was used to validate. The area under the curve(AUC) was used for discrimination, Brier Score (BS) was used for calibration, positive predictive value(PPV), negative predictive value(NPV), and F1 score were combined to compare. Results A total of 2,990 HS patients were included. For predicting the 7-day mortality, the mean AUCs for RSF and Cox regression were 0.875 and 0.761, while the mean BS were 0.083 and 0.108. For predicting the 28-day mortality, the mean AUCs for RSF and Cox regression were 0.794 and 0.649, while the mean BS were 0.129 and 0.174. The mean AUCs of RSF and Cox versus conventional scores for predicting patients’ 7-day mortality were 0.875 (RSF), 0.761 (COX), 0.736 (SAPS II), 0.723 (OASIS), 0.632 (SIRS), and 0.596 (SOFA), respectively. Conclusions RSF provided a better clinical reference than Cox. Creatine, temperature, anion gap and sodium were important variables in both models.

Abstract Background Inferior vena cava tumor thrombus (IVCTT) invading the IVC wall majorly affects the surgical method choice and prognosis in renal tumors. Enhanced multiparameteric MRI plays an important role in preoperative evaluation. In this work, an MRI-based diagnostic model for IVCTT was established so as to guide the preoperative decisions. Methods Preoperative MR images of 165 cases of renal tumors with IVCTT were retrospectively analyzed, and imaging indicators were analyzed, including IVCTT morphology and Mayo grade, IVCTT diameter measurements, bland thrombosis, primary MRI-based diagnosis of renal tumor, and involvement of contralateral renal vein. The indicators were analyzed based on intraoperative performance and resection scope of the IVC wall. Multivariate logistic regression analysis was used to establish the diagnostic model. Results The morphological classification of the IVCTT, primary MRI-based diagnosis of renal tumors, maximum transverse diameter of IVCTT, and length of the bland thrombus were the main indexes predicting IVC wall invasion. The MRI-based diagnostic model established according to these indexes had good diagnostic efficiency. The prediction probability of 0.61 was set as the cutoff value. The area under the curve of the test set was 0.88, sensitivity was 0.79, specificity was 0.85, and prediction accuracy was 0.79 under the optimal cutoff value. Conclusion The preoperative MRI-based diagnostic model could reliably predict IVC wall invasion, which is helpful for better prediction of IVC-associated surgical operations.

Breast cancer progression and metastasis are governed by a complex interplay within the tumor immune microenvironment (TIME), involving numerous cell types. Lymph node metastasis (LNM) is a key prognostic marker associated with distant organ metastasis and reduced patient survival, but the mechanisms underlying its promotion by breast cancer stem cells (CSCs) remain unclear. Our study sought to unravel how CSCs reprogram TIME to facilitate LNM. Utilizing single-cell RNA sequencing, we profiled TIME in primary cancer and corresponding metastatic lymph node samples from patients at our institution. To verify the derived data, we cultured CSCs and performed validation assays employing flow cytometry and CyTOF. Our analysis revealed distinct differences in cellular infiltration patterns between tumor and LNM samples. Importantly, RAC2 and PTTG1 double-positive CSCs, which exhibit the highest stem-like attributes, were markedly enriched in metastatic lymph nodes. These CSCs are hypothesized to foster metastasis via activation of specific metastasis-related transcription factors and signaling pathways. Additionally, our data suggest that CSCs might modulate adaptive and innate immune cell evolution, thereby further contributing to metastasis. In summary, this study illuminates a critical role of CSCs in modifying TIME to facilitate LNM. The enrichment of highly stem-like CSCs in metastatic lymph nodes offers novel therapeutic targeting opportunities and deepens our understanding of breast cancer metastasis.

BackgroundCarotid plaque can progress into stroke, myocardial infarction, etc, which are major global causes of death. Evidence shows a significant increase in carotid plaque incidence among patients with fatty liver disease. However, unlike the high detection rate of fatty liver disease, screening for carotid plaque in the asymptomatic population is not yet prevalent due to cost-effectiveness reasons, resulting in a large number of patients with undetected carotid plaques, especially among those with fatty liver disease. ObjectiveThis study aimed to combine the advantages of machine learning (ML) and logistic regression to develop a straightforward prediction model among the population with fatty liver disease to identify individuals at risk of carotid plaque. MethodsOur study included 5,420,640 participants with fatty liver from Meinian Health Care Center. We used random forest, elastic net (EN), and extreme gradient boosting ML algorithms to select important features from potential predictors. Features acknowledged by all 3 models were enrolled in logistic regression analysis to develop a carotid plaque prediction model. Model performance was evaluated based on the area under the receiver operating characteristic curve, calibration curve, Brier score, and decision curve analysis both in a randomly split internal validation data set, and an external validation data set comprising 32,682 participants from MJ Health Check-up Center. Risk cutoff points for carotid plaque were determined based on the Youden index, predicted probability distribution, and prevalence rate of the internal validation data set to classify participants into high-, intermediate-, and low-risk groups. This risk classification was further validated in the external validation data set. ResultsAmong the participants, 26.23% (1,421,970/5,420,640) were diagnosed with carotid plaque in the development data set, and 21.64% (7074/32,682) were diagnosed in the external validation data set. A total of 6 features, including age, systolic blood pressure, low-density lipoprotein cholesterol (LDL-C), total cholesterol, fasting blood glucose, and hepatic steatosis index (HSI) were collectively selected by all 3 ML models out of 27 predictors. After eliminating the issue of collinearity between features, the logistic regression model established with the 5 independent predictors reached an area under the curve of 0.831 in the internal validation data set and 0.801 in the external validation data set, and showed good calibration capability graphically. Its predictive performance was comprehensively competitive compared with the single use of either logistic regression or ML algorithms. Optimal predicted probability cutoff points of 25% and 65% were determined for classifying individuals into low-, intermediate-, and high-risk categories for carotid plaque. ConclusionsThe combination of ML and logistic regression yielded a practical carotid plaque prediction model, and was of great public health implications in the early identification and risk assessment of carotid plaque among individuals with fatty liver.

Background: Non-alcoholic fatty liver disease (NAFLD) is one of the most common liver diseases worldwide. Currently, most NAFLD prediction models are diagnostic models based on cross-sectional data, which failed to provide early identification or clarify causal relationships. We aimed to use time-series deep learning models with longitudinal health checkup records to predict the onset of NAFLD in the future, and update the model stepwise by incorporating new checkup records to achieve dynamic prediction. Methods: 10,493 participants with over 6 health checkup records from Beijing MJ Health Screening Center were included to conduct a retrospective cohort study, in which the constantly updated initial 5 checkup data were incorporated stepwise to predict the risk of NAFLD at and after their sixth health checkups. A total of 33 variables were considered, consisting of demographic characteristics, medical history, lifestyle, physical examinations, and laboratory tests. L1-penalized logistic regression (LR) was used for feature selection. The long short-term memory (LSTM) algorithm was introduced for model development, and five-fold cross-validation was conducted to tune and choose optimal hyperparameters. Both internal validation and external validation were conducted, using the 20% randomly divided holdout test dataset and previously unseen data from Shanghai MJ Health Screening Center, respectively, to evaluate model performance. The evaluation metrics included area under the receiver operating characteristic curve (AUROC), sensitivity, specificity, Brier score, and decision curve. Bootstrap sampling was implemented to generate 95% confidence intervals of all the metrics. Finally, the Shapley additive explanations (SHAP) algorithm was applied in the holdout test dataset for model interpretability to obtain time-specific and sample-specific contributions of each feature. Results: Among the 10,493 participants, 1662 (15.84%) were diagnosed with NAFLD at and after their sixth health checkups. The predictive performance of the deep learning model in the internal validation dataset improved over the incorporation of the checkups, with AUROC increasing from 0.729 (95% CI: 0.698,0.760) at baseline to 0.818 (95% CI: 0.798,0.844) when consecutive 5 checkups were included. The external validation dataset, containing 1728 participants, was used to verify the results, in which AUROC increased from 0.700 (95% CI: 0.657,0.740) with only the first checkups to 0.792 (95% CI: 0.758,0.825) with all five. The results of feature significance showed that body fat percentage, alanine transaminase (ALT), and uric acid owned the greatest impact on the outcome, time-specific, individual-specific and dynamic feature contributions were also produced for model interpretability. Conclusion: A dynamic prediction model was successfully established in our study, and the prediction capability kept improving with the renewal of the latest checkup records. In addition, we identified key features associated with the onset of NAFLD, making it possible to optimize the prevention and control strategies of the disease in the general population.

IntroductionPrevious studies based on a single measure of fasting plasma glucose (FPG) showed an inconsistent conclusion about the association between FPG and osteoporosis risk. Not accounting for time-varying and cumulative average of FPG over time could bias the true relation between FPG and osteoporosis. Our study aims to investigate the association between the trajectories of FPG and osteoporosis risk for non-diabetic and diabetic populations.MethodsA total of 18,313 participants who attended physical examinations during 2008–2018 were included. They were free of osteoporosis at their first physical examination and followed until their last physical examination before December 31, 2018. We recorded their incidence of osteoporosis and at least three FPG values during follow-up. Their longitudinal FPG trajectories were identified by the latent class growth analysis model based on the changes in FPG. Multivariable logistic regression models were used to analyze the association between the trajectories of FPG and osteoporosis diagnosed in the follow-up physical examination in both non-diabetics and diabetics.ResultsThere were 752 incident osteoporosis among 16,966 non-diabetic participants, and 57 incident osteoporosis among 1,347 diabetic participants. Among non-diabetics, the elevated-increasing FPG trajectory was negatively associated with osteoporosis risk in women (odds ratio (OR), 0.62; 95% confidence interval (CI), 0.43–0.88). Premenopausal women with elevated-increasing FPG trajectory had lower osteoporosis risk than those women with normal-stable FPG trajectory (OR, 0.41; 95% CI, 0.20–0.88), while this association was insignificant in postmenopausal women. Among diabetics, those whose longitudinal FPG is kept at a very high level had the highest risk of osteoporosis (OR, 3.09; 95% CI, 1.16–8.22), whereas those whose FPG starts with the high level and keeps on increasing did not exhibit a significantly increased risk (OR, 1.75; 95% CI, 0.81–3.76) compared with those who keep stable moderate-high level of FPG, except in men (OR, 2.49; 95% CI, 1.02–6.12).ConclusionDistinct trajectories of FPG are associated with differential risk of osteoporosis in non-diabetic and diabetic populations. Controlling a proper FPG level in different populations is necessary for osteoporosis prevention.

BackgroundIn-hospital mortality, prolonged length of stay (LOS), and 30-day readmission are common outcomes in the intensive care unit (ICU). Traditional scoring systems and machine learning models for predicting these outcomes usually ignore the characteristics of ICU data, which are time-series forms. We aimed to use time-series deep learning models with the selective combination of three widely used scoring systems to predict these outcomes.Materials and methodsA retrospective cohort study was conducted on 40,083 patients in ICU from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database. Three deep learning models, namely, recurrent neural network (RNN), gated recurrent unit (GRU), and long short-term memory (LSTM) with attention mechanisms, were trained for the prediction of in-hospital mortality, prolonged LOS, and 30-day readmission with variables collected during the initial 24 h after ICU admission or the last 24 h before discharge. The inclusion of variables was based on three widely used scoring systems, namely, APACHE II, SOFA, and SAPS II, and the predictors consisted of time-series vital signs, laboratory tests, medication, and procedures. The patients were randomly divided into a training set (80%) and a test set (20%), which were used for model development and model evaluation, respectively. The area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and Brier scores were used to evaluate model performance. Variable significance was identified through attention mechanisms.ResultsA total of 33 variables for 40,083 patients were enrolled for mortality and prolonged LOS prediction and 36,180 for readmission prediction. The rates of occurrence of the three outcomes were 9.74%, 27.54%, and 11.79%, respectively. In each of the three outcomes, the performance of RNN, GRU, and LSTM did not differ greatly. Mortality prediction models, prolonged LOS prediction models, and readmission prediction models achieved AUCs of 0.870 ± 0.001, 0.765 ± 0.003, and 0.635 ± 0.018, respectively. The top significant variables co-selected by the three deep learning models were Glasgow Coma Scale (GCS), age, blood urea nitrogen, and norepinephrine for mortality; GCS, invasive ventilation, and blood urea nitrogen for prolonged LOS; and blood urea nitrogen, GCS, and ethnicity for readmission.ConclusionThe prognostic prediction models established in our study achieved good performance in predicting common outcomes of patients in ICU, especially in mortality prediction. In addition, GCS and blood urea nitrogen were identified as the most important factors strongly associated with adverse ICU events.

Background and Objectives: Triple-negative breast cancer (TNBC), a highly aggressive and heterogeneous subtype of breast cancer, accounts for ap-proximately 10–15% of all breast cancer cases. Currently, there is no effective therapeutic target for TNBC. Tu-mor-associated macrophages (TAMs), which can be phenotypically classified into M1 and M2 subtypes, have been shown to influence the prognosis of various cancers, including ovarian cancer. This study aimed to investigate the role of M1/M2 macrophages in the TNBC tumor microenvironment (TME), with a focus on identifying prognostic genes and predicting immunotherapy response. Materials and Methods: The study employed the CIBERSORT algorithm to analyze immune cell expression in the TME. Genes associated with the M1/M2 macrophage ratio were identified using Pearson correlation analysis and used to classify patients into dis-tinct clusters. Dimensionality reduction techniques, including univariate Cox regression and Lasso, were applied to these genes. The expression of prognostic genes was validated through immunohistochemistry. Results: The study found a high prevalence of TAMs in the TME. Among the patient clusters, 109 differentially expressed genes (DEGs) were identified. Three significant DEGs (LAMP3, GZMB, and CXCL13) were used to construct the riskScores. The riskScore model effectively stratified patients based on mortality risk. Gene Set Enrichment Analysis (GSEA) associated the riskScore with several significant pathways, including mismatch repair, JAK/STAT3 signaling, VEGF signaling, antigen processing presentation, ERBB signaling, and P53 signaling. The study also predicted patient sensitivity to im-munotherapy using the riskScores. The expression of the three significant DEGs was validated through immunohisto-chemistry. Conclusions: The study concluded that the riskScore model, based on the M1/M2 macrophage ratio, is a valid prognostic tool for TNBC. The findings underscore the importance of the TME in TNBC progression and prognosis and highlight the po-tential of the riskScore model in predicting immunotherapy response in TNBC patients.

Breast cancer (BC) is the most prevalent cancer in women worldwide. A systematic approach to BC treatment, comprising adjuvant and neoadjuvant chemotherapy (NAC), as well as hormone therapy, forms the foundation of the disease’s therapeutic strategy. The extracellular matrix (ECM) is a dynamic network that exerts a robust biological effect on the tumor microenvironment (TME), and it is highly regulated by several immunological components, such as chemokines and cytokines. It has been established that the ECM promotes the development of an immunosuppressive TME. Therefore, while analyzing the ECM of BC, immune-related genes must be considered. In this study, we used bioinformatic approaches to identify the most valuable ECM-related immune genes. We used weighted gene co-expression network analysis to identify the immune-related genes that potentially regulate the ECM and then combined them with the original ECM-related gene set for further analysis. Least absolute shrinkage and selection operator (LASSO) regression and SurvivalRandomForest were used to narrow our ECM-related gene list and establish an ECM index (ECMI) to better delineate the ECM signature. We stratified BC patients into ECMI high and low groups and evaluated their clinical, biological, and genomic characteristics. We found that the ECMI is highly correlated with long-term BC survival. In terms of the biological process, this index is positively associated with the cell cycle, DNA damage repair, and homologous recombination but negatively with processes involved in angiogenesis and epithelial–mesenchymal transition. Furthermore, the tumor mutational burden, copy number variation, and DNA methylation levels were found to be related to the ECMI. In the Metabric cohort, we demonstrated that hormone therapy is more effective in patients with a low ECMI. Additionally, differentially expressed genes from the ECM-related gene list were extracted from patients with a pathologic complete response (pCR) to NAC and with residual disease (RD) to construct a neural network model for predicting the chance of achieving pCR individually. Finally, we performed qRT-PCR to validate our findings and demonstrate the important role of the gene OGN in predicting the pCR rate. In conclusion, delineation of the ECM signature with immune-related genes is anticipated to aid in the prediction of the prognosis of patients with BC and the benefits of hormone therapy and NAC in BC patients.

The tumor microenvironment (TME) is a complex ecosystem comprised of cancer cells, stromal cells, and immune cells. Analysis of the composition of TME is essential to assess the prognosis of patients with breast cancer (BC) and the efficacy of different regimes. Treg plays a crucial role in the microenvironment of breast cancer subtypes, and its function contributes to the development and progression of BC by suppressing anti-tumor immunity directly or indirectly through multiple mechanisms. In addition, conventional treatments, such as anthracycline-based neoadjuvant chemotherapy, and neo-therapies, such as immune-checkpoint blockades, have a significant impact on the absence of Tregs in BC TME, thus gaining additional anti-tumor effect to some extent. Strikingly, Treg in BC TME revealed the predicted efficacy of some therapeutic strategies. All these results suggest that we can manipulate the abundance of Treg to achieve the ultimate effect of both conventional and novel treatments. In this review, we discuss new insights into the characteristics of Treg in BC TME, the impact of different regiments on Treg, and the possibilities of Treg as a predictive marker of efficacy for certain treatments.

Background: Acute high-altitude (HA) exposure results in blood pressure (BP) and cardiac function variations in most subjects, some of whom suffer from acute mountain sickness (AMS). Several previous studies have found that cardiovascular function indicators are potentially correlated with AMS.Objectives: This study aims to examine HA-induced cardiovascular adaptations in AMS patients and compare them with healthy subjects. It also aims to investigate the relationship between cardiovascular function indicators and AMS, as well as to provide some insightful information about the prevention and treatment of AMS.Methods: Seventy-two subjects were enrolled in this cohort study. All the subjects ascended Litang (4,100 m above sea level). They were monitored by a 24-h ambulatory blood pressure (ABP) device and underwent echocardiography examination within 24 h of altitude exposure. The 2018 Lake Louise questionnaire was used to evaluate AMS.Results: Acute mountain sickness group consisted of more women (17 [60.7%] vs. 10 [22.7%], p = 0.001) and fewer smokers (5 [17.9%] vs. 23 [52.3%], p = 0.003). Compared with subjects without AMS, subjects with AMS had lower pulse pressure (PP) (daytime PP, 45.23 ± 7.88 vs. 52.14 ± 4.75, p < 0.001; nighttime PP, 42.81 ± 5.92 vs. 49.39 ± 7.67, p < 0.001) and lower effective arterial elastance (Ea) (1.53 ± 0.24 vs. 1.73 ± 0.39, p = 0.023). Multivariate regression indicated that female sex (OR = 0.23, p = 0.024), lower daytime PP (OR = 0.86, p = 0.004), and lower Ea (OR = 0.03, p = 0.015) at low altitude (LA) were independent risk factors for AMS. Combined daytime PP and Ea at LA had a high predictive value for AMS (AUC = 0.873; 95% CI: 0.789–0.956). Correlation analysis showed that AMS-induced headache correlated with daytime PP (R = −0.401, p < 0.001) and nighttime PP at LA (R = −0.401, p < 0.001).Conclusion: Our study demonstrated that AMS patients had a lower PP and Ea at LA. These baseline indicators of vasodilation at LA were closely associated with AMS, which may explain the higher headache severity in subjects with higher PP at LA.

The aim of this study is to evaluate the changes in soy isoflavones and peptides in soy milk after lactic acid bacterial fermentation, and explore the positive effects of fermented soy milk on an oxidative aging mouse model induced with D-galactose. We found that free soybean isoflavones and peptides increased after soy milk was fermented by Lactobacillus fermentum CQPC04. The in vivo results indicated that L. fermentum CQPC04-fermented soy milk enhanced the organ index of the liver and spleen, and improved the pathological morphology of the liver, spleen, and skin. L. fermentum CQPC04-fermented soy milk increased the enzymatic activity of glutathione peroxidase (GSH-Px), total superoxide dismutase (T-SOD), and catalase (CAT), increased glutathione (GSH), but decreased the levels of nitric oxide (NO) and malondialdehyde (MDA) in serum, liver, and brain tissues of oxidative aging mice. The above mentioned fermented soy milk also increased the levels of collagen I (Col I), hyaluronic acid (HA), and collagen III (Col III), and decreased the levels of advanced glycation End products (AGEs) and hydrogen peroxide (H2O2). The RT-qPCR results showed that L. fermentum CQPC04-fermented soy milk upregulated the mRNA expression of nuclear factor erythroid 2?related factor (Nrf2), heme oxygenase-1 (HMOX1), quinone oxido-reductase 1 (Nqo1), neuronal nitric oxide synthase (NOS1), endothelial nitric oxide synthase (NOS3), Cu/Zn–superoxide dismutase (Cu/Zn-SOD), Mn–superoxide dismutase (Mn-SOD), and CAT, but downregulated the expression of inducible nitric oxide synthase (NOS2) and glutamate cysteine ligase modifier subunit (Gclm) in liver and spleen tissues. Lastly, the fermented soy milk also increased the gene expression of Cu/Zn-SOD, Mn-SOD, CAT, GSH-Px, matrix metalloproteinases 1 (TIMP1), and matrix metalloproteinases 2 (TIMP2), and decreased the expression of matrix metalloproteinase 2 (MMP2) and matrix metalloproteinase 9 (MMP9) in skin tissue. In conclusion, L. fermentum CQPC04-fermented soy milk was able to satisfactorily delay oxidative aging effects, and its mechanism may be related to the increase in free soy isoflavones and peptides.

After soymilk fermented by L. plantarum HFY09, the bound soybean isoflavones content was reduced, but free soybean isoflavones and peptides increased. In vivo results show that L. plantarum HFY09-fermented soymilk increased the T-SOD, MDA, GSH-Px, GSH, CAT, and NO in the serum, liver, and brain tissues of mice. The aforementioned soymilk also increased the Col I and hyaluronic acid, and reduced the contents of Col III, H2O2, and AGEs in the skin. L. plantarum HFY09-fermented soymilk upregulated Nrf2, HO-1, NOS1, NOS3, SOD1, SOD2, CAT, Nqol, Gclm expression in the liver and the spleen, reduced the NOS2 expression, and upregulated SOD1 SOD2, CAT, GSH-Px, TIMP1, and TIMP2 expression in skin, downregulated MMP3, and MMP9 expression. These results show that L. plantarum HFY09-fermented soymilk could effectively alleviate aging in D-galactose-induced aging mice, the mechanism may be related to the activation of Nrf2 signaling pathway by soy isoflavone aglycone and small molecule peptides.

Cross-cultural experience plays a very important role to the growth of pre-service teachers. This narrative study explored Canadian pre-service teachers’ Chinese language learning and their cross-cultural experiences in China related to participating in a three-month international program between a university in Canada and a university in China. This study focuses on four participants’ Chinese foreign language learning and how their Chinese learning influenced their cross-cultural learning when being immersed in a Chinese language environment. The findings show that the pre-service teachers not only developed a basic level of oral language proficiency, but also developed a higher language tolerance for the learners who are non-native speakers of English. In addition, through learning Chinese, the pre-service teachers learned more about Chinese culture and developed an appreciation of different cultures which helped them develop better perceptions and attitudes toward multicultural education in Canada.