OBJECTIVES: To evaluate the performance of a multi-constraint representation learning classification model for identifying ovarian cancer with missing laboratory indicators. METHODS: Tabular data with missing laboratory indicators were collected from 393 patients with ovarian cancer and 1951 control patients. The missing ovarian cancer laboratory indicator features were projected to the latent space to obtain a classification model using the representational learning classification model based on discriminative learning and mutual information coupled with feature projection significance score consistency and missing location estimation. The proposed constraint term was ablated experimentally to assess the feasibility and validity of the constraint term by accuracy, area under the ROC curve (AUC), sensitivity, and specificity. Cross-validation methods and accuracy, AUC, sensitivity and specificity were also used to evaluate the discriminative performance of this classification model in comparison with other interpolation methods for processing of the missing data. RESULTS: The results of the ablation experiments showed good compatibility among the constraints, and each constraint had good robustness. The cross-validation experiment showed that for identification of ovarian cancer with missing laboratory indicators, the AUC, accuracy, sensitivity and specificity of the proposed multi-constraints representation-based learning classification model was 0.915, 0.888, 0.774, and 0.910, respectively, and its AUC and sensitivity were superior to those of other interpolation methods. CONCLUSIONS The proposed model has excellent discriminatory ability with better performance than other missing data interpolation methods for identification of ovarian cancer with missing laboratory indicators.
Female ; Humans ; Ovarian Neoplasms/diagnosis* ; Machine Learning ; ROC Curve

【Objective】 To evaluate the effectiveness of Roche Cobas s 201 in detecting HBV by analyzing its blood nucleic acid testing (NAT) results. 【Methods】 The results were grouped according to the enzyme-linked immunosorbent assay (ELISA) and NAT minipool test (MP), NAT individual test (ID) and repeated NAT ID test (rID), and categorized into 4 groups as ELISA+ /NAT(ID)+ , ELISA+ /NAT(rID)+ , ELISA-/NAT(ID)+ and ELISA-/NAT(rID)+ . The data were statistically analyzed to explore whether there was a difference in the detection of reactive results by repeated NAT, and the correlation between cycle threshold (Ct) and nucleic acid detection rate for NAT-reactive samples with different ELISA results. The true infection status of blood donors was further analyzed by supplementary tests, including NAT systems and chemiluminescence serological marker assays using other methodologies. 【Results】 A total of 1 691 groups of 766 293 blood donor samples were HBV NAT(MP)+ , of which 1 418 groups(83.86%) were detected with reactive results (1 418 HBV NAT+ , 7 090 NAT-), and there were still 273 groups (16.14%) that remained undetected after repeated testing[a total of 1 638 NAT-, Ct(MP): 39.49±3.62]. Of the HBV NAT+ , 881(62.13%) were ELISA+ /NAT(ID)+ , 19(1.34%) were ELISA+ /NAT(rID)+ , 451(31.81%) were ELISA-/NAT(ID)+ , and 67(4.72%) were ELISA-/NAT(rID)+ . For samples with different ELISA results, difference was found in the detection of HBV by repeated NAT (P<0.05). There was no difference in Ct(ID) values between groups ELISA+ /NAT(rID)+ and ELISA-/ NAT(ID)+ , and groups ELISA+ /NAT(rID)+ and ELISA-/ NAT(rID)+ (P>0.05), but there were significant differences between other groups compared pairwise (P<0.05). Supplementary tests were performed on 228 ELISA-/ NAT(MP)+ (ID)- samples, 56 (24.56%) were reactive by chemiluminescent detection of HBsAg+ and 7 (3.07%) by other NAT systems. Among the remaining 221 NAT- samples/donors (96.93%), 53 (23.98%) HBsAg+ donors were likely to have chronic infection, 40 (18.10%) anti-HBe+ and/or anti-HBc+ donors might have previous infections, and the remaining 128 (57.92%) donors who were non-reactive were NAT (MP) pseudo-reactive, with significant differences in anti-HBs levels \'between groups (P<0.05). 【Conclusion】 Repeated NAT has differential detection of donor samples with different reactivity categories or different serologic results, especially within a certain interval, and repeated NAT for ELISA- samples can significantly improve the detection rate. Ct values can assist in assessing the stability and accuracy of the NAT system. For ELISA-/NAT(MP)+ (ID)- donors, the combination of other highly sensitive assays can reduce the risk of viral residuals and safeguard clinical blood safety.

Objective:To analyze the situation of hepatitis B enzyme immunoassay-positive but nucleic acid negative infection (HBsAg ELISA(+ )/HBV-DNA(-)) among unpaid blood donors in Zhejiang Blood Center, and to explore the causes of inconsistency between enzyme immunoassay and nucleic acid result.Methods:A single nucleic acid test was performed on blood donors whose routine blood screening result were HBsAg-ELISA(+ )/HBV-DNA(-), and the test result of such blood donors were analyzed.Results:A total of 205 HBsAg-ELISA(+ )/HBV-DNA(-) samples were screened from 114017 blood donors from May to November, 2022. The proportion of male blood donors (0.14%) were significantly lower than that of the female blood donors (0.24%)( χ2= 14.761, P<0.005); the proportion of the first blood donor (0.32%) was significantly higher than that of the second blood donor (0.09%) ( χ2 = 78.781, P<0.005); the difference between different education levels is statistically significant ( χ2 =47.753, P<0.005). After single-person nucleic acid re-detection, the re-detection rate of nucleic acid in ELISA double-reagent positive samples was higher than that in single-reagent positive samples ( χ2=94.378, P<0.005); there was no significant difference between ELISA reagent 1 and reagent 2 in the detection rate of nucleic acid ( χ2 =0.163, P>0.005). There was no significant difference in the positive rate of secondary nucleic acid detection between the two nucleic acid detection systems ( χ2=0.626, P>0.005). Serological supplementary test showed that 11 HBV-DNA(+ ) samples showed two serological combination patterns after chemiluminescence detection, namely HBsAg(+ )/HBeAb(+ ) and HBeAb(+ ), most of which were HBsAg(+ )/HBeAb (+ ), a total of 10 cases, accounting for 90.91%, and only one case was HBeAb (+ ), accounting for 9.09%. The quantitative result of HBsAg showed that most of them were at low HBsAg level. Conclusions:After re-detection by single nucleic acid detection method, HBsAg-ELISA(+ )/HBV-DNA(-) samples of blood donors do have a certain proportion of HBV-DNA(+ ), but most of the samples were still HBV-DNA (-), additional experiments on HBV serological markers and HBV-DNA are needed to determine their true infection status and clarify the reasons for the inconsistency between enzyme immunoassay and nucleic acid test result. In addition, nucleic acid and HBsAg detection reagents with high sensitivity and specificity should be selected as far as possible in blood donor screening to ensure the accuracy of result.

Objective:To explore the characteristics of theory of mind(TOM) ability of adolescent depressive disorder patients with non-suicidal self-injury (NSSI) behavior, and analyze the relationship between NSSI behavior and TOM, so as to identify patients with NSSI behavioral risk early and give timely intervention.Methods:A total of 54 adolescent patients with depressive disorder who were hospitalized for the first time in Shenzhen Kangning Hospital from May to October 2021 were enrolled.They were divided into NSSI group( n=31)and non-NSSI group( n=23)according to DSM-5.Hamilton anxiety rating scale-14 item (HAMA-14) and Hamilton depression rating scale-17 item(HAMD-17)were used to evaluate the severity of anxiety and depression.Hinting task, Yoni task and reading the mind in the eyes test(RMET) were used to test TOM of the two groups.Binary Logistic regression analysis in SPSS 22.0 was used to explore the influence of TOM ability on NSSI behavior of adolescents with depressive disordor. Results:Compared with patients in non-NSSI group, patients in NSSI had significantly higher scores of HAMA-14(21.48±4.92 vs 16.35±5.61, t=3.57, P<0.05) and HAMD-17(25.61±4.08 vs 21.43±4.64)( t=3.51, P<0.05). The patients in NSSI group had lower scores on the hinting task(16.10±2.84 vs 18.17±1.15, t=-3.68, P<0.05), RMET task (21.61±2.58 vs 23.61±3.07)( t=-2.59, P<0.05), and second-order cognitive TOM((18.90±3.70) vs (20.96±2.72), t=-2.27, P<0.05) and second-order affective TOM(28.84±3.93 vs 31.04±3.04)( t=-2.24, P<0.05) in Yoni task than those in non-NSSI group.Binary Logistic regression analysis showed effective TOM ( β=-0.306, OR=0.736, 95% CI= 0.552-0.982, P=0.037)and cognitive TOM ( β=-0.485, OR=0.616, 95% CI=0.396-0.957, P=0.031) were associated with non-suicidal self-injury in patients with depressive disorder. Conclusion:The increased impairment of the emotional component (decoding component) and cognitive component (reasoning component) of TOM may be risk factors for the occurrence of NSSI behavior in adolescent depressive disorder patients.

ObjectiveTo explore the characteristics of impulsivity in adolescent depressive disorder patients with non-suicidal self-injury （NSSI） behavior， analyze the relationship between NSSI behavior and impulsivity， so as to identify patients with NSSI behavior and provide targeted intervention at early stages. MethodsA total of 53 adolescent patients with depressive disorder who were hospitalized for the first hospitalization in Shenzhen Kangning Hospital were enrolled， diagnosed using Diagnostic and Statistical Manual of Mental Disorders， fifth edition （DSM-5）. Thereafter， the adolescents were divided into NSSI group （n=30） and non-NSSI group （n=23）. Hamilton Depression Scale-17 item （HAMD-17）， Adolescents Self-Harm Scale and Barratt Impulsivity Scale （BIS-11） were used to evaluate the severity of depression， NSSI and impulsivity. ResultsCompared with the non-NSSI group， the NSSI group scored higher on HAMD-17 ［（25.50±4.10） vs. （21.43±4.64）， t=3.379， P<0.01］， motor impulsiveness of BIS-11 ［（51.67±15.95） vs. （38.70±14.90）， t=3.018， P<0.01］， cognitive impulsiveness ［（52.75±13.22） vs. （43.37±18.40）， t=2.161， P<0.05］， non-planning impulsiveness ［（68.00±15.32） vs. （50.76±21.35）， t=3.424， P<0.01］ and BIS-11 ［（57.42±11.08） vs. （44.27±14.83）， t=3.695， P<0.01］. Within NSSI group， the score of Adolescents Self-Harm Scale was positively correlated with the score of motor impulsiveness in BIS-11 （r=0.691， P<0.01）. Binary Logistic regression analysis showed that HAMD-17 score （β=0.172， OR=1.187， 95% CI： 1.007~1.400） and non-planning impulsiveness of BIS-11 （β=0.044， OR=1.045， 95% CI： 1.002~1.091） were associated with NSSI in adolescent patients with depressive disorder （P<0.05）. ConclusionThe severity of depressive symptoms and non-planning impulsiveness may be risk factors for NSSI behavior in adolescent patients with depressive disorder.