Development and validation of a patient-level model to predict dementia across a network of observational databases.

BACKGROUND: A prediction model can be a useful tool to quantify the risk of a patient developing dementia in the next years and take risk-factor-targeted intervention. Numerous dementia prediction models have been developed, but few have been externally validated, likely limiting their clinical uptake. In our previous work, we had limited success in externally validating some of these existing models due to inadequate reporting. As a result, we are compelled to develop and externally validate novel models to predict dementia in the general population across a network of observational databases. We assess regularization methods to obtain parsimonious models that are of lower complexity and easier to implement. METHODS: Logistic regression models were developed across a network of five observational databases with electronic health records (EHRs) and claims data to predict 5-year dementia risk in persons aged 55-84. The regularization methods L1 and Broken Adaptive Ridge (BAR) as well as three candidate predictor sets to optimize prediction performance were assessed. The predictor sets include a baseline set using only age and sex, a full set including all available candidate predictors, and a phenotype set which includes a limited number of clinically relevant predictors. RESULTS: BAR can be used for variable selection, outperforming L1 when a parsimonious model is desired. Adding candidate predictors for disease diagnosis and drug exposure generally improves the performance of baseline models using only age and sex. While a model trained on German EHR data saw an increase in AUROC from 0.74 to 0.83 with additional predictors, a model trained on US EHR data showed only minimal improvement from 0.79 to 0.81 AUROC. Nevertheless, the latter model developed using BAR regularization on the clinically relevant predictor set was ultimately chosen as best performing model as it demonstrated more consistent external validation performance and improved calibration. CONCLUSIONS: We developed and externally validated patient-level models to predict dementia. Our results show that although dementia prediction is highly driven by demographic age, adding predictors based on condition diagnoses and drug exposures further improves prediction performance. BAR regularization outperforms L1 regularization to yield the most parsimonious yet still well-performing prediction model for dementia.

Establishment and validation of a prediction model for gestational diabetes.

AIM: To develop a visual prediction model for gestational diabetes (GD) in pregnant women and to establish an effective and practical tool for clinical application. METHODS: To establish a prediction model, the modelling set included 1756 women enrolled in the Zunyi birth cohort, the internal validation set included 1234 enrolled women, and pregnant women in the Wuhan cohort were included in the external validation set. We established a demographic-lifestyle factor model (DLFM) and a demographic-lifestyle-environmental pollution factor model (DLEFM) based on whether the women were exposed to environmental pollutants. The least absolute shrinkage and selection lasso-logistic regression analyses were used to identify the independent predictors of GD and construct a nomogram for predicting its occurrence. RESULTS: The DLEFM regression analysis showed that a family history of diabetes (odd ratio [OR] 2.28; 95% confidence interval [CI] 1.05-4.71), a history of GD in pregnant women (OR 4.22; 95% CI 1.89-9.41), being overweight or obese before pregnancy (OR 1.71; 95% CI 1.27-2.29), a history of hypertension (OR 2.61; 95% CI 1.41-4.72), sedentary time (h/day) (OR 1.16; 95% CI 1.08-1.24), monobenzyl phthalate (OR 1.95; 95% CI 1.45-2.67) and Q4 mono-ethyl phthalate concentration (OR 1.85; 95% CI 1.26-2.73) were independent predictors. The area under the receiver operating curves for the internal validation of the DLEFM and the DLFM constructed using these seven factors was 0.827 and 0.783, respectively. The calibration curve of the DLEFM was close to the diagonal line. The DLEFM was thus the more optimal model, and the one which we chose. CONCLUSIONS: A nomogram based on preconception factors was constructed to predict the occurrence of GD in the second and third trimesters. It provided an effective tool for the early prediction and timely management of GD.

Safety signal detection with control of latent factors.

Postmarket drug safety database like vaccine adverse event reporting system (VAERS) collect thousands of spontaneous reports annually, with each report recording occurrences of any adverse events (AEs) and use of vaccines. We hope to identify signal vaccine-AE pairs, for which certain vaccines are statistically associated with certain adverse events (AE), using such data. Thus, the outcomes of interest are multiple AEs, which are binary outcomes and could be correlated because they might share certain latent factors; and the primary covariates are vaccines. Appropriately accounting for the complex correlation among AEs could improve the sensitivity and specificity of identifying signal vaccine-AE pairs. We propose a two-step approach in which we first estimate the shared latent factors among AEs using a working multivariate logistic regression model, and then use univariate logistic regression model to examine the vaccine-AE associations after controlling for the latent factors. Our simulation studies show that this approach outperforms current approaches in terms of sensitivity and specificity. We apply our approach in analyzing VAERS data and report our findings.

Spatial-temporal distribution of preterm birth in China, 1990-2020: A systematic review and modelling analysis.

BACKGROUND: Little is known about the long-term trends of preterm birth rates in China and their geographic variation by province. OBJECTIVES: To estimate the annual spatial-temporal distribution of preterm birth rates in China by province from 1990 to 2020. DATA SOURCES: We searched PubMed, EMBASE, Web of Science, CNKI, WANFANG and VIP from January 1990 to September 2023. STUDY SELECTION AND DATA EXTRACTION: Studies that provided data on preterm births in China after 1990 were included. Data were extracted following the Guidelines for Accurate and Transparent Health Estimates Reporting. SYNTHESIS: We assessed the quality of each survey using a 9-point checklist. We estimated the annual preterm birth risk by province using Bayesian multilevel logistic regression models considering potential socioeconomic, environmental, and sanitary predictors. RESULTS: Based on 634 survey data from 343 included studies, we found a gradual increase in the preterm birth risk in most provinces in China since 1990, with an average annual increase of 0.7% nationally. However, the preterm birth rates in Inner Mongolia, Hubei, and Fujian Province showed a decline, while those in Sichuan were quite stable since 1990. In 2020, the estimates of preterm birth rates ranged from 2.9% (95% Bayesian credible interval [BCI] 2.1, 3.8) in Inner Mongolia to 8.5% (95% BCI 6.6, 10.9) in Jiangxi, with the national estimate of 5.9% (95% BCI 4.3, 8.1). Specifically, some provinces were identified as high-risk provinces for either consistently high preterm birth rates (e.g. Jiangxi) or relatively large increases (e.g. Shanxi) since 1990. CONCLUSIONS: This study provides annual information on the preterm birth risk in China since 1990 and identifies high-risk provinces to assist in targeted control and intervention for this health issue.

Frailty and the Correlation Between Total Testosterone Levels and Urinary Incontinence Among Elderly Women.

INTRODUCTION AND HYPOTHESIS: The objective was to explore the correlation between total testosterone levels and stress urinary incontinence (SUI) and urgency urinary incontinence (UUI) in older patients, emphasizing frailty. METHODS: This prospective cross-sectional study included 1,328 women over 60 years of age at an incontinence specialty clinic. Participants were assessed for UI, frailty, using the Japanese Frailty Scale, and total testosterone levels. Analysis of a logistic regression model was employed for age, body mass index (BMI), and vaginal deliveries adjustment, with association and multivariate analyses to evaluate the associations with SUI and UUI. RESULTS: The frailty and nonfrailty groups each consisted of 664 individuals. After age, BMI, and the number of vaginal deliveries adjustment, the analysis showed a negative association between total testosterone levels and both SUI (p < 0.001) and UUI (p < 0.001) in the frailty group. Multivariate analysis revealed that, in the nonfrailty group, factors such as low total testosterone levels (p = 0.0145), diabetes (p = 0.0052), and cerebral infarction (p = 0.0254) were related to SUI, whereas no significant factors were associated with UUI. In the frailty group, factors associated with SUI included low total testosterone levels (p < 0.0001), the number of vaginal deliveries (p < 0.0001), smoking (p = 0.0240), chronic lung disease (p < 0.0248), and hypertension (p < 0.0265). Factors associated with UUI were age (p < 0.0001), low total testosterone levels (p = 0.0025), diabetes (p < 0.0001), and the number of vaginal deliveries (p = 0.0152). CONCLUSIONS: The study highlights the significance of incorporating the assessment of frailty and testosterone levels in addressing UI among older women, particularly in the aged population, underscoring the need for tailored approaches in this demographic.

Leveraging pharmacokinetic parameters as covariate in Bayesian logistic regression model to optimize dose selection in early phase oncology trial.

Dose selection and optimization in early phase of oncology drug development serves as the foundation for the success of late phases drug development. Bivariate Bayesian logistic regression model (BLRM) is a widely utilized model-based algorithm that has been shown to improve the accuracy for identifying recommended phase 2 dose (RP2D) based on dose-limiting-toxicity (DLT) over traditional method such as 3 + 3. However, it remains a challenge to optimize dose selection that strikes a proper balance between safety and efficacy in escalation and expansion phase of phase I trials. In this paper, we first use a phase I clinical trial to demonstrate how the variability of drug exposure related to pharmacokinetic (PK) parameters among trial participants may add to the difficulties of identifying optimal dose. We use simulation to show that concurrently or retrospectively fitting BLRM model for dose/toxicity data from escalation phase with dose-independent PK parameters as covariate lead to improved accuracy of identifying dose level at which DLT rate is within a prespecified toxicity interval. Furthermore, we proposed both model- and rule-based methods to modify dose at patient level in expansion cohorts based on their PK/exposure parameters. Simulation studies show this approach leads to higher likelihood for a dose level with a manageable toxicity and desirable efficacy margin to be advanced to late phase pipeline after being screened at expansion phase of phase I trial.

Prevalence of childhood hypertension and associated factors in Zhejiang Province: a cross-sectional analysis based on random forest model and logistic regression.

With childhood hypertension emerging as a global public health concern, understanding its associated factors is crucial. This study investigated the prevalence and associated factors of hypertension among Chinese children. This cross-sectional investigation was conducted in Pinghu, Zhejiang province, involving 2,373 children aged 8-14 years from 12 schools. Anthropometric measurements were taken by trained staff. Blood pressure (BP) was measured in three separate occasions, with an interval of at least two weeks. Childhood hypertension was defined as systolic blood pressure (SBP) and/or diastolic blood pressure (DBP) ≥ age-, sex-, and height-specific 95th percentile, across all three visits. A self-administered questionnaire was utilized to collect demographic, socioeconomic, health behavioral, and parental information at the first visit of BP measurement. Random forest (RF) and multivariable logistic regression model were used collectively to identify associated factors. Additionally, population attributable fractions (PAFs) were calculated. The prevalence of childhood hypertension was 5.0% (95% confidence interval [CI]: 4.1-5.9%). Children with body mass index (BMI) ≥ 85th percentile were grouped into abnormal weight, and those with waist circumference (WC) > 90th percentile were sorted into central obesity. Normal weight with central obesity (NWCO, adjusted odds ratio [aOR] = 5.04, 95% CI: 1.96-12.98), abnormal weight with no central obesity (AWNCO, aOR = 4.60, 95% CI: 2.57-8.21), and abnormal weight with central obesity (AWCO, aOR = 9.94, 95% CI: 6.06-16.32) were associated with an increased risk of childhood hypertension. Childhood hypertension was attributable to AWCO mostly (PAF: 0.64, 95% CI: 0.50-0.75), followed by AWNCO (PAF: 0.34, 95% CI: 0.19-0.51), and NWCO (PAF: 0.13, 95% CI: 0.03-0.30). Our results indicated that obesity phenotype is associated with childhood hypertension, and the role of weight management could serve as potential target for intervention.

Determinants and spatial factors of anemia in women of reproductive age in Democratic Republic of Congo (drc): a Bayesian multilevel ordinal logistic regression model approach.

BACKGROUND: As a global public health problem, anemia affects more than 400 million women of reproductive age worldwide, mostly in Africa and India. In the DRC, the prevalence of anemia has decreased slightly from 52.9% in 2007, to 46.4% in 2012 and 42.4% in 2019. However, there is considerable regional variation in its distribution. The aim of this study is to determine the factors contributing to anemia in women of reproductive age and to explore its spatial distribution in the DRC. METHODS: Based on the Bayesian Multilevel Spatial Ordinal Logistic Regression Model, we used the 2013 Democratic Republic of Congo Demographic and Health Survey (DHS-DRC II) data to investigate individual and environmental characteristics contributing to the development of anemia in women of reproductive age and the mapping of anemia in terms of residual spatial effects. RESULTS: Age, pregnancy status, body mass index, education level, current breastfeeding, current marital status, contraceptive and insecticide-treated net use, source of drinking water supply and toilet/latrine use including the province of residence were the factors contributing to anemia in women of reproductive age in DRC. With Global Moran's I = -0.00279, p-value ≥ 0.05, the spatial distribution of anemia in women of reproductive age in DRC results from random spatial processes. Thus, the observed spatial pattern is completely random. CONCLUSION: The Bayesian Multilevel Spatial Ordinal Logistic Regression statistical model is able to adjust for risk and spatial factors of anemia in women of reproductive age in DRC highlighting the combined role of individual and environmental factors in the development of anemia in DRC.

Utilization of reproductive health services among undergraduate regular class students in Assosa University, BGRS: a comparison among three varieties of multilevel binary logistic regression models.

INTRODUCTION: Reproductive health service (RHS) helps for people to have a delighted and safe sex through their life journey. It enables especially for women to go safely through pregnancy and childbirth and provide couples with the best chance of having a healthy infant. Therefore, this study aimed to identify the significant determinants of RHS utilization among undergraduate regular class students in Assosa University by using advanced methodology. METHODS: We used cross-sectional study design to collect RHS data from 362 students in Assosa University from 5 to 16, may 2021. These students were selected using stratified random sampling technique. We also used cross-tabulation to summarize the extents of RHS utilization across all predictors in terms of percentage and three varieties of multilevel binary logistic regression model to model the determinants of RHS. RESULTS: 42.27% of undergraduate regular class students in Assosa University utilize at least one type of RHS during their time at Assosa University whereas, 57.73% of undergraduate regular class students in this University are not utilized it. Among three varieties of multilevel binary logistic regression models, the random slopes two-level model was selected as a best fitted model for the datasets. At 5% level of significance, awareness about RHS, gender, preference of service fees and student's monthly average income were significant predictor variables in this model. In addition, the covariates; age, gender and preference of service fees have a significant random effects on utilization of RHS across all colleges/school. CONCLUSION: Students who; preferred service fee as usual rate, have awareness about RHS, are females and have high monthly average income were more likely to utilize RHS. RHS utilization among undergraduate regular students in Assosa University is likely to increase more effectively with interventions that address these factors.

Association Between Atopic Dermatitis in Pregnant Women and Preterm Births: The Japan Environment and Children's Study.

OBJECTIVES: This study aimed to evaluate the association between atopic dermatitis in pregnant women and preterm births, accounting for maternal ritodrine hydrochloride administration status. METHODS: Data of 83,796 women with singleton pregnancies at and after 22 weeks of gestation (enrolled between 2011 and 2014) were analyzed. These data were obtained from the Japan Environment and Children's Study. Atopic dermatitis was defined based on self-reported questionnaire responses obtained during the first trimester. The primary outcome measures were preterm births before 37, 32, and 28 weeks of gestation. Using a multivariable logistic regression model, odds ratios for preterm births in pregnant women with atopic dermatitis were calculated, with women without atopic dermatitis included in the reference group. This analysis considered confounding factors and maternal ritodrine hydrochloride administration. RESULTS: Among pregnant women with atopic dermatitis, the adjusted odds ratios (95% confidence intervals) for preterm births before 37, 32, and 28 weeks of gestation were 0.89 (0.81-0.98), 0.98 (0.74-1.30), and 0.88 (0.50-1.55), respectively. This trend remained consistent after excluding participants who received ritodrine hydrochloride. CONCLUSIONS FOR PRACTICE: Atopic dermatitis in pregnant women was significantly associated with a decreased incidence of preterm births before 37 weeks of gestation, even after accounting for the effects of maternal ritodrine hydrochloride administration.

Effect of admission time of arrival on quality of in-hospital care in acute ischemic stroke patients in China.

OBJECTIVES: The quality of care for patients may be partly determined by the time they are admitted to the hospital. This study was conducted to explore the effect of admission time and describe the pattern and magnitude of weekly variation in the quality of patient care. STUDY DESIGN: A retrospective observational study. METHODS: Data were collected from the Medical Care Quality Management and Control System for Specific (Single) Diseases in China. A total of 238,122 patients treated for acute ischemic stroke between January 2015 and December 2017 were included. The primary outcomes were completion of the ten process indicators and in-hospital death. RESULTS: The quality of in-hospital care varied according to hospital arrival time. We identified several patterns of variation across the days of the week. In the first pattern, the quality of four indicators, such as stroke physicians within 15 min, was lowest for arrivals between 08:00 and 11:59, increased throughout the day, and peaked for arrivals between 20:00 and 23:59 or 00:00 and 03:59. In the second pattern, the quality of four indicators, such as the application of antiplatelet therapy within 48 h, was not significantly different between days and weeks. There was no difference in in-hospital mortality between the different admission times. CONCLUSIONS: The effect of admission time on the quality of in-hospital care of patients with acute ischemic stroke showed several diurnal patterns. Detecting the times when quality is relatively low may lead to quality improvements in health care. Quality improvement should also focus on reducing diurnal temporal variation.

On the relative conservativeness of Bayesian logistic regression method in oncology dose-finding studies.

The Bayesian logistic regression method (BLRM) is a widely adopted and flexible design for finding the maximum tolerated dose in oncology phase I studies. However, the BLRM design has been criticized in the literature for being overly conservative due to the use of the overdose control rule. Recently, a discussion paper titled "Improving the performance of Bayesian logistic regression model with overall control in oncology dose-finding studies" in Statistics in Medicine has proposed an overall control rule to address the "excessive conservativeness" of the standard BLRM design. In this short communication, we discuss the relative conservativeness of the standard BLRM design and also suggest a dose-switching rule to further enhance its performance.

Diagnostic signature composed of seven genes in HIF-1 signaling pathway for preeclampsia.

PURPOSE: In this study, we explored the relationship of genes in HIF-1 signaling pathway with preeclampsia and establish a logistic regression model for diagnose preeclampsia using bioinformatics analysis. METHOD: Two microarray datasets GSE75010 and GSE35574 were downloaded from the Gene Expression Omnibus database, which was using for differential expression analysis. DEGs were performed the Gene Ontology (GO) analysis, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis and Gene set enrichment analysis (GSEA). Then we performed unsupervised consensus clustering analysis using genes in HIF-1 signaling pathway, and clinical features and immune cell infiltration were compared between these clusters, as well as the least absolute shrinkage and selection operator (LASSO) method to screened out key genes to constructed logistic regression model, and receiver operating characteristic (ROC) curve was plotted to evaluate the accuracy of the model. RESULTS: 57 DEGs were identified, of which GO, KEGG and analysis GSEA showed DEGs were mostly involved in HIF-1 signaling pathway. Two subtypes were identified of preeclampsia and 7 genes in HIF1-signaling pathway were screened out to establish the logistic regression model for discrimination preeclampsia from controls, of which the AUC are 0.923 and 0.845 in training and validation datasets respectively. CONCLUSION: Seven genes (including MKNK1, ARNT, FLT1, SERPINE1, ENO3, LDHA, BCL2) were screen out to build potential diagnostic model of preeclampsia.

Analysis and prediction of older adult sports participation in South Korea using artificial neural networks and logistic regression models.

BACKGROUND: Korea's aging population and the lack of older adult participation in sports are increasing medical expenses. AIMS: This study aimed to segment older adult sports participants based on their demographic characteristics and exercise practice behavior and applied artificial neural network and logistic regression models to these segments to best predict the effect of medical cost reduction. It presents strategies for older adult sports participation. METHODS: A sample comprising data on 1,770 older adults aged 50 years and above was drawn from the 2019 National Sports Survey. The data were analyzed through frequency analysis, hierarchical and K-means clustering, artificial neural network, logistic regression, cross-tabulation analyses, and one-way ANOVA using SPSS 23 and Modeler 14.2. RESULTS: The participants were divided into five clusters. The artificial neural network and logistic analysis models showed that the cluster comprising married women in their 60s who participated in active exercise had the highest possibility of reducing medical expenses. DISCUSSION: Targeting women in their 60s who actively participate in sports, the government should expand the supply of local gymnasiums, community centers, and sports programs. If local gymnasiums and community centers run sports programs and appoint appropriate sports instructors, the most effective medical cost reduction effect can be obtained. CONCLUSIONS: This study contributes to the field by providing insights into the specific demographic segments to focus on for measures to reduce medical costs through sports participation.

Factors affecting depression among married adults: a gender-based household cross-sectional study.

BACKGROUND: Depression is a common mental health problem all over the world including Bangladesh. World Health Organization included it in the Mental Health Gap Action Programme as one of its priority conditions. Research on this issue is scanty in Bangladesh. Therefore, we designed to a gender-based household study on associated factors of depression among married adults in Rajshahi City of Bangladesh. METHODS: We conducted this household cross-sectional study from August 01 to October 31, 2019. A total of 708 married adults currently living together in Rajshahi City were recruited for this study. We applied a multi-stage random sampling technique for selecting samples and used a semi-structured questionnaire to collect necessary information from them. The Patient Health Questionnaire-9 was used for measuring depression and frequency distribution and binary logistic regression model were used for data analysis. RESULTS: The prevalence of depression (moderate to severe) was 14.4% (95% CI: 11.9-16.9) among married adults, and females (21.2%, 95% CI: 17.2-25.4) suffered more than males (7.6%, 95% CI: 4.8-10.5). A multiple binary logistic regression model established four main factors of depression among married females: (i) multiple marriage [AOR = 19.982; 95% CI: 10.081-39.610; p < 0.01]; (ii) poor relationship with spouse [AOR = 2.175; 95%CI: 1.068-4.428; p < 0.05]; (iii) chronic medical comorbidity [AOR = 1.876; 95%CI: 1.009-2.626; p < 0.05]; and (iv) 7-12 years duration of conjugal life [AOR = 2.091, 1.009-4.334; p < 0.05]. Two main factors of depression among married males were (i) multiple marriage [AOR = 24.605; 95% CI: 20.228-40.402; p < 0.01] and hard work [AOR = 4.358; 95%CI: 1.109-7.132; p < 0.05]. CONCLUSION: The prevalence of depression was significantly high among the study population, and females were the most vulnerable group. The concerned authorities and stakeholders should take appropriate measures to manage the problem with special focus on the risk factors and the vulnerable groups.

A random forest algorithm-based prediction model for moderate to severe acute postoperative pain after orthopedic surgery under general anesthesia.

BACKGROUND: Postoperative pain is one of the most common complications after surgery. In order to detect early and intervene in time for moderate to severe postoperative pain, it is necessary to identify risk factors and construct clinical prediction models. This study aimed to identify significant risk factors and establish a better-performing model to predict moderate to severe acute postoperative pain after orthopedic surgery under general anesthesia. METHODS: Patients who underwent orthopedic surgery under general anesthesia were divided into patients with moderate to severe pain group (group P) and patients without moderate to severe pain group (group N) based on VAS scores. The features selected by Lasso regression were processed by the random forest and multivariate logistic regression models to predict pain outcomes. The classification performance of the two models was evaluated through the testing set. The area under the curves (AUC), the accuracy of the classifiers, and the classification error rate for both classifiers were calculated, the better-performing model was used to predict moderate to severe acute postoperative pain after orthopedic surgery under general anesthesia. RESULTS: A total of 327 patients were enrolled in this study (228 in the training set and 99 in the testing set). The incidence of moderate to severe postoperative pain was 41.3%. The random forest model revealed a classification error rate of 25.2% and an AUC of 0.810 in the testing set. The multivariate logistic regression model revealed a classification error rate of 31.3% and an AUC of 0.764 in the testing set. The random forest model was chosen for predicting clinical outcomes in this study. The risk factors with the greatest and second contribution were immobilization and duration of surgery, respectively. CONCLUSIONS: The random forest model can be used to predict moderate to severe acute postoperative pain after orthopedic surgery under general anesthesia, which is of potential clinical application value.

Ultrasound Radiomics-Based Logistic Regression Model to Differentiate Between Benign and Malignant Breast Nodules.

OBJECTIVES: To explore the potential value of ultrasound radiomics in differentiating between benign and malignant breast nodules by extracting the radiomic features of two-dimensional (2D) grayscale ultrasound images and establishing a logistic regression model. METHODS: The clinical and ultrasound data of 1000 female patients (500 pathologically benign patients, 500 pathologically malignant patients) who underwent breast ultrasound examinations at our hospital were retrospectively analyzed. The cases were randomly divided into training and validation sets at a ratio of 7:3. Once the region of interest (ROI) of the lesion was manually contoured, Spearman's rank correlation, least absolute shrinkage and selection operator (LASSO) regression, and the Boruta algorithm were adopted to determine optimal features and establish a logistic regression classification model. The performance of the model was assessed using the area under the receiver operating characteristic curve (AUC), and calibration and decision curves (DCA). RESULTS: Eight ultrasound radiomic features were selected to establish the model. The AUC values of the model were 0.979 and 0.977 in the training and validation sets, respectively (P = .0029), indicating good discriminative ability in both datasets. Additionally, the calibration and DCA suggested that the model's calibration efficiency and clinical application value were both superior. CONCLUSIONS: The proposed logistic regression model based on 2D grayscale ultrasound images could facilitate differential diagnosis of benign and malignant breast nodules. The model, which was constructed using ultrasound radiomic features identified in this study, demonstrated good diagnostic performance and could be useful in helping clinicians formulate individualized treatment plans for patients.

Expression of EMT-related genes in lymph node metastasis in endometrial cancer: a TCGA-based study.

BACKGROUND: Endometrial cancer (EC) with metastasis in pelvic/para-aortic lymph nodes suggests an unsatisfactory prognosis. Nevertheless, there is still rare literature focusing on the role of epithelial-mesenchymal transition (EMT) in lymph node metastasis (LNM) in EC. METHODS: Transcriptional data were derived from the TCGA database. Patients with stage IA-IIIC2 EC were included, constituting the LN-positive and LN-negative groups. To evaluate the extent of EMT, an EMT signature composed of 315 genes was adopted. The EMT-related genes (ERGs) were obtained from the dbEMT2 database, and the differentially expressed ERGs (DEERGs) between these two groups were screened. On the basis of DEERGs, pathway analysis was carried out. We eventually adopted the logistic regression model to build an ERG-based gene signature with predictive value for LNM in EC. RESULTS: A total of 498 patients were included, with 75 in the LN-positive group. Median EMT score of tumor tissues from LN-negative group was - 0.369, while that from the LN-positive group was - 0.296 (P < 0.001), which clearly exhibited a more mesenchymal phenotype for LNM cases on the EMT continuum. By comparing expression profiles, 266 genes were identified as DEERGs, in which 184 were upregulated and 82 were downregulated. In pathway analysis, various EMT-related pathways were enriched. DEERGs shared between molecular subtypes were comparatively few. The ROC curve and logistic regression analysis screened 7 genes with the best performance to distinguish between the LN-positive and LN-negative group, i.e., CIRBP, DDR1, F2RL2, HOXA10, PPARGC1A, SEMA3E, and TGFB1. A logistic regression model including the 7-gene-based risk score, age, grade, myometrial invasion, and histological subtype was built, with an AUC of 0.850 and a favorite calibration (P = 0.074). In the validation dataset composed of 83 EC patients, the model exhibited a satisfactory predictive value and was well-calibrated (P = 0.42). CONCLUSION: The EMT status and expression of ERGs varied in LNM and non-LNM EC tissues, involving multiple EMT-related signaling pathways. Aside from that, the distribution of DEERGs differed among molecular subtypes. An ERG-based gene signature including 7 DEERGs exhibited a desirable predictive value for LNM in EC, which required further validation based upon clinical specimens in the future.

Intersectional Inequalities in Anthropometric Failure among Indian Children: Evidence from the National Family Health Survey (2015-2016).

Increasing body of health planning and policy research focused upon unravelling the fundamental drivers of population health and nutrition inequities, such as wealth status, educational status, caste/ethnicity, gender, place of residence, and geographical context, that often interact to produce health inequalities. However, very few studies have employed intersectional framework to explicitly demonstrate how intersecting dimensions of privilege, power, and resources form the burden of anthropometric failures of children among low-and-middle income countries including India. Data on 2,15,554 sampled children below 5 years of age from the National Family Health Survey 2015-2016 were analysed. This study employed intersectional approach to examine caste group inequalities in the anthropometric failure (i.e. moderate stunting, severe stunting, moderate underweight, severe underweight, moderate wasting, severe wasting) among children in India. Descriptive statistics and multinomial logistic regression models were fitted to investigate the heterogeneities in the burden of anthropometric failure across demographic, socioeconomic and contextual factors. Interaction effects were estimated to model the joint effects of socioeconomic position (household wealth, maternal education, urban/rural residence and geographical region) and caste groups with the likelihood of anthropometric failure among children.More than half of under-5 children suffered from anthropometric failure in India. Net of the demographic and socioeconomic characteristics, children from the disadvantageous caste groups whose mother were illiterate, belonged to economically poor households, resided in the rural areas, and coming from the central and eastern regions experienced disproportionately higher risk of anthropometric failure than their counterparts in India. Concerted policy processes must recognize the existing heterogeneities between and within population groups to improve the precision targeting of the beneficiary and enhance the efficiency of the nutritional program among under-5 children, particularly for the historically marginalized caste groups in India.

Construction and validation of machine learning models for sepsis prediction in patients with acute pancreatitis.

BACKGROUND: This study aimed to construct predictive models for the risk of sepsis in patients with Acute pancreatitis (AP) using machine learning methods and compared optimal one with the logistic regression (LR) model and scoring systems. METHODS: In this retrospective cohort study, data were collected from the Medical Information Mart for Intensive Care III (MIMIC III) database between 2001 and 2012 and the MIMIC IV database between 2008 and 2019. Patients were randomly divided into training and test sets (8:2). The least absolute shrinkage and selection operator (LASSO) regression plus 5-fold cross-validation were used to screen and confirm the predictive factors. Based on the selected predictive factors, 6 machine learning models were constructed, including support vector machine (SVM), K-nearest neighbour (KNN), multi-layer perceptron (MLP), LR, gradient boosting decision tree (GBDT) and adaptive enhancement algorithm (AdaBoost). The models and scoring systems were evaluated and compared using sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy, and the area under the curve (AUC). RESULTS: A total of 1, 672 patients were eligible for participation. In the training set, 261 AP patients (19.51%) were diagnosed with sepsis. The predictive factors for the risk of sepsis in AP patients included age, insurance, vasopressors, mechanical ventilation, Glasgow Coma Scale (GCS), heart rate, respiratory rate, temperature, SpO2, platelet, red blood cell distribution width (RDW), International Normalized Ratio (INR), and blood urea nitrogen (BUN). The AUC of the GBDT model for sepsis prediction in the AP patients in the testing set was 0.985. The GBDT model showed better performance in sepsis prediction than the LR, systemic inflammatory response syndrome (SIRS) score, bedside index for severity in acute pancreatitis (BISAP) score, sequential organ failure assessment (SOFA) score, quick-SOFA (qSOFA), and simplified acute physiology score II (SAPS II). CONCLUSION: The present findings suggest that compared to the classical LR model and SOFA, qSOFA, SAPS II, SIRS, and BISAP scores, the machine learning model-GBDT model had a better performance in predicting sepsis in the AP patients, which is a useful tool for early identification of high-risk patients and timely clinical interventions.