Deep Cascade-Learning Model via Recurrent Attention for Immunofixation Electrophoresis Image Analysis.

Immunofixation Electrophoresis (IFE) analysis has been an indispensable prerequisite for the diagnosis of M-protein, which is an important criterion to recognize diversified plasma cell diseases. Existing intelligent methods of IFE diagnosis commonly employ a single unified classifier to directly classify whether M-protein exists and which isotype of M-protein is. However, this unified classification is not optimal because the two tasks have different characteristics and require different feature extraction techniques. Classifying the M-protein existence depends on the presence or absence of dense bands in IFE data, while classifying the M-protein isotype depends on the location of dense bands. Consequently, a cascading two-classifier framework suitable to the two tasks respectively may achieve better performance. In this paper, we propose a novel deep cascade-learning model, which sequentially integrates a positive-negative classifier based on deep collocative learning and an isotype classifier based on recurrent attention model to address these two tasks respectively. Specifically, the attention mechanism can mimic the visual perception of clinicians, where only the most informative local regions are extracted through sequential partial observations. This not only avoids the interference of redundant regions but also saves computational power. Further, domain knowledge about SP lane and heavy-light-chain lanes is also introduced to assist our attention location. Extensive numerical experiments show that our deep cascade-learning outperforms state-of-the-art methods on recognized evaluation metrics and can effectively capture the co-location of dense bands in different lanes.

Establishing reference intervals for urine and serum iodine levels: A nationwide multicenter study of a euthyroid Chinese population.

CONTEXT: Urinary iodine (UI) is commonly used for evaluating iodine status, whereas serum iodine (SI) is more closely correlated with bioavailable iodine. However, no reliable reference intervals (RIs) for clinical use are available. We aimed to establish RIs for SI, UI, and a ratio of UI to urinary creatinine (U-Cre) applicable to the Chinese population. METHODS: This multicenter cross-sectional study enrolled 930 apparently healthy adults from six representative cities in China (Beijing, Dongying, Guiyang, Urumqi, Shenzhen, and Qiqihar) in 2017. Thyroid ultrasonography and thyroid function tests, including antithyroid antibody tests, were performed to exclude individuals with latent thyroid diseases. An iodine intake-related questionnaire survey was performed. SI and UI were measured using inductively coupled plasma-mass spectrometry. Possible influencing factors of iodine levels were evaluated using multiple regression analysis. RESULTS: Post-exclusion, the final analysis included 894 individuals. Seafood intake frequency was positively correlated with SI (standardized partial regression coefficient = 0.23) but not with UI and UI/U-Cre. SI was positively correlated with serum TT4 (Spearman correlation coefficient: 0.40), TT3 (0.23), and FT4 (0.18). SI and UI showed no age- or sex-specific variations. Significantly higher UI/U-Cre values were observed in Qiqihar than in Beijing, Guizhou, and Shenzhen. Shenzhen showed the lowest UI levels among all evaluated cities. With application of latent abnormal values exclusion procedurere, the RIs for SI, UI, and UI/U-Cre in the population were 36.0-79.3 µg/L, 19-385 µg/L, 22-450 µg/g, respectively. CONCLUSIONS: We established RIs for UI and SI among healthy Chinese individuals with no thyroid nodule or dysfunction.

Nationwide Chinese study for establishing reference intervals for thyroid hormones and related tests.

OBJECTIVE: This nationwide study aimed to establish Chinese specific reference intervals (RIs) for thyroid related tests: thyroid stimulating hormone (TSH), free thyroxine (FT4), free triiodothyronine (FT3), total thyroxine (TT4), total triiodothyronine (TT3), anti-thyroid peroxidase (TPO-Ab) and anti-thyroglobulin antibodies (TG-Ab). METHOD: Apparently healthy individuals (n = 2380) were recruited from Beijing, Guizhou, Urumqi, Dongying, Shenzhen, and Qiqihar in China. All the tests were measured by immunoassay testing. Ultrasonography was performed to exclude thyroid nodules. Multiple regression analysis was performed to identify sources of variation (SVs). Standard deviation ratio (SDR) was calculated by ANOVA for judging the need to partition RI by any given SV. RIs were computed by the parametric method. RESULTS: TPO-Ab and TG-Ab cutoffs were determined as 5 mIU/L and 2 IU/L, respectively using probability plot analysis and extrapolation of the central linear segment. Individuals with thyroid nodule and/or autoantibodies showed altered levels of thyroid hormones, and were thus excluded (n reduced to 1828). Gender difference was observed for FT3 and TT3 with females having lower levels than males. A significant relationship between age and FT3 was observed in males by Spearman correlation analysis (r = -0.231, p < .05). Although the SDR for gender difference in TSH levels was low, the difference in the upper limits of the RI was beyond allowable bias, and thus RIs (mIU/L) were partitioned by sex: females 0.72-5.50, males 0.70-4.59. CONCLUSION: By this nationwide study, RIs for thyroid hormones and cutoff values for anti-thyroid autoantibodies were established as matched to the Chinese population.

Data mining: Biological and temporal factors associated with blood cardiac troponin I concentration in a Chinese population.

BACKGROUND: Cardiac troponin is the cornerstone biomarker for the diagnosis of acute myocardial infarction. The aims of this study were to evaluate the association of biological and temporal factors with plasma cardiac troponin I (cTnI) concentration in a large group of Chinese outpatients and to explore which factor (sex, age, time of blood sampling, and season of the year) had the largest influence on plasma cTnI levels. METHODS: Analytical data with outpatient cTnI results were downloaded from the laboratory information system from January 1, 2012 to September 20, 2018. All cTnI measurements were performed with a Siemens Dimension EXL automatic chemiluminescence immunoassay analyzer. A statistical method was used to strictly exclude outliers. A total of 86,381 outpatients were enrolled in the study. RESULTS: In individuals over 60 years old, cTnI levels gradually increased with age in both males and females. cTnI reached its highest levels in individuals over 80 years old (0.030 µg/L in males and 0.027 µg/L in females). In individuals over 70 years old, cTnI levels were significantly higher in males than in females (P < .05). cTnI concentration varied between individuals with different times of blood sampling. In both men and women, cTnI concentrations reached a maximum at 05:00 (0.030 µg/L and 0.026 µg/L, respectively) and peaked again at 20:00 (0.029 µg/L and 0.023 µg/L, respectively). Additionally, there were significant differences in cTnI levels between the four seasons of the year (P < .05). In winter, cTnI levels were usually higher than in spring. Linear regression analysis showed that the factor "age ≥ 80" had the greatest impact on cTnI levels. CONCLUSION: Plasma cTnI levels were significantly influenced by sex, age, time of blood sampling, and season of the year. Thus, in order to avoid incorrect identification of cTnI values as abnormal, a cTnI reference interval should be established, taking into consideration the sex and age of the individual, the time of day of blood sampling, and the season of the year.

Identification of microRNA-mRNA networks involved in cisplatin-induced renal tubular epithelial cells injury.

Cisplatin is a widely used chemotherapeutic drug that often causes acute kidney injury (AKI) in cancer patients. The contribution of miRNAs to the cisplatin-induced renal tubular epithelial cell injury remains largely unknown. Here we performed an integrative network analysis of miRNA and mRNA expression profiles to shed light into the underlying mechanism of cisplatin-induced renal tubular epithelial cell injury. Microarray analysis identified 47 differentially expressed miRNAs, among them 26 were upregulated and 21 were downregulated. Moreover, integrating dysregulated miRNAs target prediction and altered mRNA expression enabled us to identify 1181 putative target genes for further bioinformatics analysis. Gene ontology (GO) analysis revealed that the putative target genes were involved in apoptosis process and regulation of transcription. Pathway analysis indicated that the top upregulated pathways included MAPK and p53 signaling pathway, while the top downregulated pathways were PI3K-Akt and Wnt signaling pathway. Further network analysis showed that MAPK signaling pathway and apoptosis with the highest degree were identified as core pathways, hsa-miR-9-3p and hsa-miR-371b-5p as the most critical miRNAs, and CASK, ASH1L, CDK6 etc. as hub target genes. In addition, the expression level change of selected five microRNAs (hsa-miR-4299, hsa-miR-297, hsa-miR-3135b, hsa-miR-9-3p, and hsa-miR-371b-5p) and two mRNAs( CASK and CDK6) were validated in cisplatin-induced HK-2 cells. Furthermore, a similar trend of expression level change was observed in NRK-52E cells by cisplatin treatment. Overall, our results provide the molecular basis and potential targets for the treatment of cisplatin-induced renal tubular cell injury.

Sigma metrics for assessing the analytical quality of clinical chemistry assays: a comparison of two approaches: Electronic supplementary material available online for this article.

INTRODUCTION: Two approaches were compared for the calculation of coefficient of variation (CV) and bias, and their effect on sigma calculation, when different allowable total error (TEa) values were used to determine the optimal method for Six Sigma quality management in the clinical laboratory. MATERIALS AND METHODS: Sigma metrics for routine clinical chemistry tests using three systems (Beckman AU5800, Roche C8000, Siemens Dimension) were determined in June 2017 in the laboratory of Peking Union Medical College Hospital. Imprecision (CV%) and bias (bias%) were calculated for ten routine clinical chemistry tests using a proficiency testing (PT)- or an internal quality control (IQC)-based approach. Allowable total error from the Clinical Laboratory Improvement Amendments of 1988 and the Chinese Ministry of Health Clinical Laboratory Center Industry Standard (WS/T403-2012) were used with the formula: Sigma = (TEa - bias) / CV to calculate the Sigma metrics (σCLIA, σWS/T) for each assay for comparative analysis. RESULTS: For the PT-based approach, eight assays on the Beckman AU5800 system, seven assays on the Roche C8000 system and six assays on the Siemens Dimension system showed σCLIA > 3. For the IQC-based approach, ten, nine and seven assays, respectively, showed σCLIA > 3. Some differences in σ were therefore observed between the two calculation methods and the different TEa values. CONCLUSIONS: Both methods of calculating σ can be used for Six Sigma quality management. In practice, laboratories should evaluate Sigma multiple times when optimizing a quality control schedule.

Methylation of NBPF1 as a novel marker for the detection of plasma cell-free DNA of breast cancer patients.

BACKGROUND: Recent studies revealed that tumor-specific gene methylation can be detected in the circulating cell-free DNA (cfDNA) of cancer patients; therefore, methylated cfDNA is considered a promising biomarker. Human neuroblastoma breakpoint family member 1 (NBPF1) was originally identified in a neuroblastoma (NB) patient. The present study is the first to evaluate the presence of NBPF1 gene methylation in cfDNA in plasma of breast cancer patients. METHODS: Differentially methylated cfDNA was screened using bisulfite sequencing with a next-generation sequencer (BS-seq) among 25 breast cancer patients, 25 patients with a benign breast disease and 25 healthy female volunteers. Then, five specific methylation sites in the NBPF1 gene were verified in 11 breast cancer samples, and two further sites in the NBPF1 promoter were detected in 52 breast cancer patients (stages I-III), 31 patients with benign breast disease and 30 healthy controls by using methylation-specific PCR (MSP). Furthermore, the association between the methylation statuses of NBPF1 and the clinicopathological characteristics of breast cancer patients was analyzed. RESULTS: BS-seq demonstrated that the NBPF1 methylation levels in breast cancer patients were higher than those in patients with benign breast disease and healthy controls. The MSP results showed that the methylation rates of two sites in the NBPF1 promoter were 67.1% and 61.4% in breast cancer patients, 48.2% and 59.6% in patients with benign breast disease, and 40.9% and 48.1% in healthy controls, respectively. The methylation rates of one site were significantly different among the three groups (p < .05), with the highest rate in breast cancer patients. Moreover, there was no statistically significant correlation between the NBPF1 promoter methylation and the major clinicopathological features of the patients. CONCLUSIONS: These results indicate that hypermethylation of the NBPF1 promoter occurs in a significant proportion of breast tumors and that NBPF1-methylated cfDNA may serve as a potential tumor marker for breast cancer.

Establishing reference intervals for sex hormones and SHBG in apparently healthy Chinese adult men based on a multicenter study.

BACKGROUND: Measuring sex hormones is essential in diagnosing health issues such as testicular dysfunction, male infertility and feminization syndrome. However, there are no reports on reference intervals (RIs) in Chinese men. We conducted a nationwide multicenter study to establish RIs for seven sex hormones (luteinizing hormone [LH], follicle-stimulating hormone [FSH], prolactin [PRL], total testosterone [TT], free testosterone [FT], bioavailable testosterone [BAT] and estrogen [E2]), as well as sex hormone-binding globulin (SHBG). METHODS: In 2013, 1043 apparently healthy adult men from five representative cities in China (Beijing, Hangzhou, Guangzhou, Dalian and Urumqi) were recruited; hormones were measured using an automated immunoassay analyzer. Multiple regression analysis (MRA) was performed to identify sources of variation (SVs) that might influence the hormone serum levels. RIs were computed using the parametric method. RESULTS: Dalian and Hangzhou had significantly higher E2 values than other cities; age was a major source of variation for FSH, LH, PRL, SHBG, FT and BAT. FSH, LH and SHBG increased significantly with age, while PRL, FT and BAT decreased with age. TT showed no significant age-related changes. Median (RIs) derived without partition by age were as follows: FSH, 5.6 (1.9-16.3) IU/L; LH, 4.2 (1.6-10.0) IU/L; PRL, 189 (88-450) mIU/L; E2, 85 (4.7-195) pmol/L; SHBG, 29.4 (11.5-66.3) nmol/L; TT, 15.6 (7.4-24.5) nmol/L; FT, 0.31 (0.16-0.52) nmol/L; and BAT, 8.0 (3.7-13.2) nmol/L. RIs were also derived in accordance with between-city and between-age differences. CONCLUSIONS: RIs were established for sex hormones and SHBG in apparently healthy Chinese men in consideration of age.

CCCTC-binding factor inhibits breast cancer cell proliferation and metastasis via inactivation of the nuclear factor-kappaB pathway.

CCCTC-binding factor (CTCF) is an important epigenetic regulator implicated in multiple cellular processes, including growth, proliferation, differentiation, and apoptosis. Although CTCF deletion or mutation has been associated with human breast cancer, the role of CTCF in breast cancer is questionable. We investigated the biological functions of CTCF in breast cancer and the underlying mechanism. The results showed that CTCF expression in human breast cancer cells and tissues was significantly lower than that in normal breast cells and tissues. In addition, CTCF expression correlated significantly with cancer stage (P = 0.043) and pathological differentiation (P = 0.029). Furthermore, CTCF overexpression resulted in the inhibition of proliferation, migration, and invasion, while CTCF knockdown induced these processes in breast cancer cells. Transcriptome analysis and further experimental confirmation in MDA-MD-231 cells revealed that forced overexpression of CTCF might attenuate the DNA-binding ability of nuclear factor-kappaB (NF-κB) p65 subunit and inhibit activation of NF-κB and its target pro-oncogenes (tumor necrosis factor alpha-induced protein 3 [TNFAIP3]) and genes for growth-related proteins (early growth response protein 1 [EGR1] and growth arrest and DNA-damage-inducible alpha [GADD45a]). The present study provides a new insight into the tumor suppressor roles of CTCF in breast cancer development and suggests that the CTCF/NF-κB pathway is a potential target for breast cancer therapy.

Establishing age-specific reference intervals for anti-Müllerian hormone in adult Chinese women based on a multicenter population.

AIM OF THIS STUDY: Anti-Müllerian hormone (AMH) is useful for the assessment of ovarian reserve and treatment of individualized in vitro fertilization (IVF). The aim of this study is to establish AMH reference interval for adult Chinese women on the Beckman Beckman DxI 800 platform. MATERIAL AND METHODS: From May to September 2013, serum from 1169 apparently healthy adult females from five representative cities in China (Beijing, Hangzhou, Guangzhou, Dalian and Urumqi) were collected, and AMH was analyzed on the platform of Beckman DxI 800 automated chemiluminescence immunoassay. Multiple regression analysis was used to investigate the effects of region, sex, age, body mass index (BMI), systolic blood pressure (SBP), exercise on AMH. Age specific reference interval for AMH was established. RESULTS: The main factor affecting AMH levels was age (B=-0.756, P<0.001). The AMH reference intervals for adult Chinese women aged 19-24years, 25-29years, 30-34years, 35-39years, 40-44years, 45-49years and ≥50years were 0.74-16.06, 0.67-11.64, 0.50-9.99, 0.09-8.33, 0.04-4.09, 0.01-1.46 and <0.01-0.18ng/ml, respectively. The linear, quadratic and cubic models could either provide good fit regression model to describe the decline of AMH with age (R2=0.40). CONCLUSION: This study firstly established age-specific reference intervals for AMH in Chinese women based on multicenter population.

Nationwide Multicenter Reference Interval Study for 28 Common Biochemical Analytes in China.

A nationwide multicenter study was conducted in the China to explore sources of variation of reference values and establish reference intervals for 28 common biochemical analytes, as a part of the International Federation of Clinical Chemistry and Laboratory Medicine, Committee on Reference Intervals and Decision Limits (IFCC/C-RIDL) global study on reference values. A total of 3148 apparently healthy volunteers were recruited in 6 cities covering a wide area in China. Blood samples were tested in 2 central laboratories using Beckman Coulter AU5800 chemistry analyzers. Certified reference materials and value-assigned serum panel were used for standardization of test results. Multiple regression analysis was performed to explore sources of variation. Need for partition of reference intervals was evaluated based on 3-level nested ANOVA. After secondary exclusion using the latent abnormal values exclusion method, reference intervals were derived by a parametric method using the modified Box-Cox formula. Test results of 20 analytes were made traceable to reference measurement procedures. By the ANOVA, significant sex-related and age-related differences were observed in 12 and 12 analytes, respectively. A small regional difference was observed in the results for albumin, glucose, and sodium. Multiple regression analysis revealed BMI-related changes in results of 9 analytes for man and 6 for woman. Reference intervals of 28 analytes were computed with 17 analytes partitioned by sex and/or age. In conclusion, reference intervals of 28 common chemistry analytes applicable to Chinese Han population were established by use of the latest methodology. Reference intervals of 20 analytes traceable to reference measurement procedures can be used as common reference intervals, whereas others can be used as the assay system-specific reference intervals in China.

The high prevalence of hypovitaminosis D in China: a multicenter vitamin D status survey.

Vitamin D deficiency, which is usually detected by using immunoassays or the more reliable liquid chromatography tandem mass spectrometry (LC-MS/MS) methods, has recently been considered a public health problem worldwide. However, the vitamin D status in Chinese populations, as measured using the LC-MS/MS method, is not available. The objective of this multicenter study was to determine the vitamin D status and prevalence of vitamin D deficiency by using a reliable method in 5 large cities in China. From May 1 to September 31, 2013, we conducted a multicenter study on 2173 apparently healthy adults who were recruited from 5 Chinese cities. The 25-hydroxyvitamin D 25OHD2 and 25OHD3 levels were measured using the LC-MS/MS method. Intact parathyroid hormone (iPTH), calcium, phosphorus, and alkaline phosphate levels were also measured using an automatic analyzer. The mean 25OHD level of all participants was 19.4 ±â€Š6.4 ng/mL (2.5-97.5%: 7.9-32.6 ng/mL), and only 109 (5.0%) participants had a 25OHD2 level >2.5 ng/mL (maximum, 22.4 ng/mL). In this study, the prevalence of severe vitamin D deficiency (<10 ng/mL), vitamin D deficiency (10-20 ng/mL), vitamin D insufficiency (20-30 ng/mL), and vitamin D sufficiency (>30 ng/mL) was 5.9%, 50.0%, 38.7%, and 5.4%, respectively. Women had a significant higher rate of deficiency than men (66.3% vs 45.3%, P < 0.01). Participants aged 18 to 39 years had a lower 25OHD level than elderly individuals (>59 years). Lifestyle may influence the 25OHD level more than the latitude, with participants in Dalian having the highest 25OHD level and the lowest deficiency rate. The serum iPTH level showed a significant negative correlation with the 25OHD level (r = -0.23, P < 0.01) after correcting for age and sex. In conclusion, the present study evaluated the vitamin D status using a reliable method, and our results indicate that vitamin D deficiency is prevalent among all age groups in China, especially among younger adults. We also observed significant differences in the 25OHD levels according to sex, age, and region among apparently healthy individuals.

Apolipoprotein E gene polymorphisms are associated with primary hyperuricemia in a Chinese population.

OBJECTIVE: Primary hyperuricemia, an excess of uric acid in the blood, is a major public health problem. In addition to the morbidity that is attributable to gout, hyperuricemia is also associated with metabolic syndrome, hypertension, and cardiovascular disease. This study aims to assess the genetic associations between Apolipoprotein E (APOE) polymorphisms and hyperuricemia in a Chinese population. METHODS: A total of 770 subjects (356 hyperuricemic cases and 414 normouricemic controls) were recruited from the Ningxia Hui Autonomous Region, China. A physical examination was performed and fasting blood was collected for biochemical tests, including determination of the levels of serum lipid, creatinine, and uric acid. Multi-ARMS PCR was applied to determine the APOE genotypes, followed by an investigation of the distribution of APOE genotypes and alleles frequencies in the controls and cases. RESULTS: The frequencies of the APOE-ε2ε3 genotype (17.70% vs. 10.39%, P = 0.003) and the APOE-ε2 allele (10.53% vs. 5.80%, P = 0.001) were significantly higher in the hyperuricemic group than in the normouricemic group. Furthermore, male cases were more likely to have the APOE-ε2ε3 genotype and APOE-ε2 allele, compared with male controls. In both Han and Hui subjects, cases were more likely to have the APOE-ε2ε3 genotype and the APOE-ε2 allele compared with controls. Furthermore, multivariate logistic regression showed that carriers of the APOE-ε2ε3 genotype (P = 0.001, OR = 2.194) and the ε2 allele (P = 0.001, OR = 2.099) were significantly more likely to experience hyperuricemia than carriers of the ε3/ε3 genotype and the ε3 allele after adjustment for sex, body mass index (BMI), diastolic blood pressure (DBP), triglyceride (TG), low density lipoprotein cholesterol (LDL-C), creatinine (Cr) and fasting blood glucose (FBG). CONCLUSIONS: The APOE-ε2ε3 genotype and the APOE-ε2 allele are associated with serum uric acid levels in Chinese subjects, indicating that individuals carrying the APOE-ε2 allele have a higher risk of hyperuricemia than non-carriers.