Search

Objectives: Development and validation of risk prediction models at mid-pregnancy and delivery to predict admission to the neonatal care unit. Methods: We used data from all singleton deliveries at Cork University Maternity Hospital (CUMH), Ireland during 2019. Admission to the neonatal care unit was assumed if length of stay in the unit was > 24 h. Multivariable logistic regression with backward stepwise selection was used to develop the models. Discrimination was assessed using the ROC curve C-statistic, and internal validation was assessed using bootstrapping techniques. We conducted temporal external validation using data from all singleton deliveries at CUMH during 2020. Results: Out of 6,077 women, 5,809 (95.6%) with complete data were included in the analyses. A total of 612 infants (10.54%) were admitted to the neonatal care unit for > 24 hours. Six variables were informative at mid-pregnancy: male infants, maternal smoking, advancing maternal age, maternal overweight/obesity, nulliparity and history of gestational diabetes (C-statistic: 0.600, 95% CI: 0.567, 0.614). Seven variables were informative at delivery: male infants, nulliparity, public antenatal care, gestational age < 39 weeks', non-spontaneous vaginal delivery, premature rupture of membranes and time of birth between 17:01–07.59 h (C-statistic: 0.738, 95% CI: 0.715, 0.760). Using these predictors, we developed nomograms to calculate individualised risk of neonatal care unit admission. Bootstrapping indicated good internal performance and external validation suggested good reproducibility. Discussion: Our nomograms allow the user to quickly estimate individualised risk of neonatal care unit admission. Future research should aim to improve accuracy in early pregnancy to better assist counselling of parents. Significance: What is already known on this subject?: Previous prediction models for admission to neonatal care units are often limited to specific populations, lack validation or use composite outcome criteria combining neonatal unit admission with adverse neonatal outcomes. What this study adds?: Using data from a general obstetric population, we developed and validated (internally and externally) two risk prediction models to predict admission to the neonatal care unit (one to be applied at mid-pregnancy, and a second to be applied at delivery). In addition, we developed nomograms which allow the user to quickly estimate individualised risk of neonatal care unit admission. [ABSTRACT FROM AUTHOR]

Objective: The objective was to establish a machine learning-based model for predicting the occurrence and mortality of nonpulmonary sepsis-associated ARDS. Methods: 80% of sepsis patients selected randomly from the MIMIC-IV database, without prior pulmonary conditions and with nonpulmonary infection sites, were used to construct prediction models through machine learning techniques (including K-nearest neighbour, extreme gradient boosting, support vector machine, deep neural network, and decision tree methods). The remaining 20% of patients were utilized to validate the model's accuracy. Additionally, local data were employed for further model validation. Results: A total of 11,409 patients were included, with the most common type of infection being bloodstream infection. A total of 7,632 (66.9%) patients developed nonpulmonary sepsis-associated ARDS (NPS-ARDS). Patients with NPS-ARDS had significantly longer ICU stays (6.2 ± 5.2 days vs. 4.4 ± 3.7 days, p < 0.01) and higher 28-day mortality rates (19.5% vs. 14.9%, p < 0.01). Both internal and external validation demonstrated that the model constructed with the extreme gradient boosting method had high accuracy. In the internal validation, the model predicted NPS-ARDS and mortality in such patients with accuracies of 77.5% and 71.8%, respectively. In the external validation, the model predicted NPS-ARDS and mortality in these patients with accuracies of 78.0% and 81.4%, respectively. Conclusion: The model established via the extreme gradient boosting method can predict the occurrence and mortality of nonpulmonary sepsis-associated ARDS to a certain extent. [ABSTRACT FROM AUTHOR]

Background: Gestational diabetes mellitus (GDM) is a common complication of mid–to-late pregnancy. Here, we constructed a predictive model for GDM based on a combination of clinical characteristics and relevant serum markers. Methods: Data from full-term singleton vaginal deliveries from January 2022 to January 2023 were retrospectively collected from the obstetrics department. The data collected were segregated and assigned to training, validation, and external test sets. Maternal demographic characteristics, living and working habits, and haematological indicators, such as liver function and lipids were collected using a questionnaire designed for the study. The "rms" package in R was used to explore GDM-associated factors through stepwise regression at P < 0.05. A predictive model was developed based on the results of multifactorial logistic regression analysis. We then evaluated the differentiation of the column-line graphical model and performed internal and external validation. To assess the accuracy of the bar graphical model, we plotted calibration and decision curves. Results: Data from 265 pregnant women were included in the training and internal validation sets, and data from 113 pregnant women were included in the external validation set. The logistic regression algorithm screened 8 indicators as predictors. A prediction model was constructed with ALT, TBA, TC, and TG levels while considering whether GDM affects appetite, the husband– wife relationship, family history, and parental relationships as predictors. The Hosmer–Lemeshow goodness-of-fit test revealed that the chi-square values for the modelling, internal validation, and external validation groups (χ2 = 5.964, 3.249, and 12.182, respectively) were all P > 0.05. The ROC curve AUCs for the three groups were 0.93 (95% CI: 0.89–0.97), 0.72 (95% CI: 0.62–0.81), and 0.68 (95% CI: 0.53–0.83), respectively. Conclusion: In this study, a GDM prediction model was constructed to achieve high performance in GDM risk prediction based on routine obstetric tests and information. [ABSTRACT FROM AUTHOR]

STUDY QUESTION Can a simplified ovarian hyperstimulation syndrome (OHSS) risk assessment index be developed and validated with sufficient discrimination of moderate/severe OHSS from those without OHSS? SUMMARY ANSWER This easy-to-use OHSS risk assessment index shows good discriminative power and high calibration accuracy in internal and external validation cohorts. WHAT IS KNOWN ALREADY An early alert and risk stratification is critical to prevent the occurrence of OHSS. We have previously developed a multi-stage smartphone app-based prediction model to evaluate the risk of OHSS, but app use might not be so convenient in many primary institutions. A simplified OHSS risk assessment index has been required. STUDY DESIGN, SIZE, DURATION This training and internal validation of an OHSS risk assessment index used retrospective cohort data from January 2016 to December 2020. External validation was performed with a prospective cohort database from January 2021 to May 2022. There were 15 066 cycles in the training cohort, 6502 cycles in the internal validation cohort, and 8097 cycles in the external validation cohort. PARTICIPANTS/MATERIALS, SETTING, METHODS This study was performed in the reproductive medicine center of a tertiary hospital. Infertile women who underwent ovarian stimulation were included. Data were extracted from the local database with detailed medical records. A multi-stage risk assessment index was constructed at multiple stages. The first stage was before the initiation of ovarian stimulation, the second was before the ovulation trigger, the third was after oocyte retrieval, and the last stage was on the embryo transfer day if fresh embryo transfer was scheduled. MAIN RESULTS AND THE ROLE OF CHANCE We established a simplified multi-stage risk assessment index for moderate/severe OHSS, the performance of which was further evaluated with discrimination and calibration abilities in training and internal and external validation cohorts. The discrimination abilities of the OHSS risk assessment index were determined with C-statistics. C-statistics in training (Stages 1–4: 0.631, 0.692, 0.751, 0.788, respectively) and internal (Stages 1–4: 0.626, 0.642, 0.755, 0.771, respectively) and external validation (Stages 1–4: 0.668, 0.670, 0.754, 0.773, respectively) cohorts were all increased from Stage 1 to 3 with similar trends, and were comparable between Stages 3 and 4. Calibration plots showed high agreement between observed and predicted cases in all three cohorts. Incidences of OHSS based on diverse risk stratification (negligible risk, low risk, medium risk, and high risk) were 0%, 0.6%, 2.7%, and 8.3% in the training cohort, 0%, 0.6%, 3.3%, and 8.5% in the internal validation cohort, and 0.1%, 1.1%, 4.1%, and 7.2% in the external validation cohort. LIMITATIONS, REASONS FOR CAUTION The influence from clinical interventions including cryopreservation of all embryos cannot be eliminated and thus certain risk factors like estrogen level on trigger day might be assigned with a lower risk score. Another weakness of the study is that several preventive treatments, for instance oral aspirin and letrozole, were not recorded and evaluated in the model. Despite the robust reliability of OHSS assessment index, this tool cannot be used directly for clinical decision-making or as a diagnostic tool. Its value lies in its capacity to evaluate the prognosis of various interventions and to facilitate clinician–patient communication. The combination of this tool and further symptoms and examinations should be all taken into consideration for accurate and personalized management of OHSS. WIDER IMPLICATIONS OF THE FINDINGS The OHSS risk assessment index can be implemented to facilitate personalized counseling and management of OHSS. STUDY FUNDING/COMPETING INTEREST(S) This study was supported by National Key R&D Program of China (2022YFC2702504), Medical Research Fund Guangdong Provincial (A2024003), and Xinjiang Support Rural Science and Technology (Special Correspondent) Program in Guangdong Province (KTPYJ 2023014). All authors had nothing to disclose. TRIAL REGISTRATION NUMBER N/A. [ABSTRACT FROM AUTHOR]

Aims: Patients with heart valvular regurgitation is increasing; early screening of potential patients developing heart failure (HF) is crucial. Methods: From 1 November 2019 to 31 October 2023, a total of 509 patients with heart valvular regurgitation hospitalized in the Department of Cardiovascular Disease of the First Affiliated Hospital of Guangzhou University of Traditional Medicine were enrolled. Three hundred fifty‐six cases were selected as the training set for modelling, and 153 cases were selected as the validation set for the internal validation of the model. Results: A predictive model of heart failure with the following nine risk factors was developed: atrial fibrillation (AF), pulmonary infection (PI), coronary artery disease (CAD), creatinine (CREA), low‐density lipoprotein cholesterol (LDL‐C), d‐dimer (DDi), left ventricular end‐diastolic diameter (LVEDd), mitral regurgitation (MR) and aortic regurgitation (AR). The model was evaluated by the C‐index [the training set: area under curve (AUC) 0.937, 95% confidence interval (CI) 0.911–0.963; the validation set: AUC 0.928, 95% CI 0.890–0.967]. Hosmer–Lemeshow test (the training set: χ2 10.908, P = 0.207; the validation set: χ2 4.896, P = 0.769) revealed that both the training and validation sets performed well in terms of model differentiation and calibration. Decision curve analysis showed that both the training and validation sets have higher net benefits, indicating that the model has good utility. Ten‐fold cross‐validation showed that the training set has high similarities with the validation set, which means that the model has good stability. Conclusions: The occurrence of heart failure in patients with valvular regurgitation has a significant correlation with AF, PI, CAD, CREA, LDL‐C, DDi, LVEDd, MR and AR. Based on these risk factors, a prediction model for heart failure was developed and validated, which showed good differentiation and utility, high accuracy and stability, providing a method for predicting heart failure. [ABSTRACT FROM AUTHOR]

Abstract Objective: The objective was to establish a machine learning-based model for predicting the occurrence and mortality of nonpulmonary sepsis-associated ARDS. Methods: 80% of sepsis patients selected randomly from the MIMIC-IV database, without prior pulmonary conditions and with nonpulmonary infection sites, were used to construct prediction models through machine learning techniques (including K-nearest neighbour, extreme gradient boosting, support vector machine, deep neural network, and decision tree methods). The remaining 20% of patients were utilized to validate the model’s accuracy. Additionally, local data were employed for further model validation. Results: A total of 11,409 patients were included, with the most common type of infection being bloodstream infection. A total of 7,632 (66.9%) patients developed nonpulmonary sepsis-associated ARDS (NPS-ARDS). Patients with NPS-ARDS had significantly longer ICU stays (6.2 ± 5.2 days vs. 4.4 ± 3.7 days, p

Abstract Background Gestational diabetes mellitus (GDM) is a common complication of mid–to-late pregnancy. Here, we constructed a predictive model for GDM based on a combination of clinical characteristics and relevant serum markers. Methods Data from full-term singleton vaginal deliveries from January 2022 to January 2023 were retrospectively collected from the obstetrics department. The data collected were segregated and assigned to training, validation, and external test sets. Maternal demographic characteristics, living and working habits, and haematological indicators, such as liver function and lipids were collected using a questionnaire designed for the study. The “rms” package in R was used to explore GDM-associated factors through stepwise regression at P 0.05. The ROC curve AUCs for the three groups were 0.93 (95% CI: 0.89–0.97), 0.72 (95% CI: 0.62–0.81), and 0.68 (95% CI: 0.53–0.83), respectively. Conclusion In this study, a GDM prediction model was constructed to achieve high performance in GDM risk prediction based on routine obstetric tests and information.

Abstract Aims Patients with heart valvular regurgitation is increasing; early screening of potential patients developing heart failure (HF) is crucial. Methods From 1 November 2019 to 31 October 2023, a total of 509 patients with heart valvular regurgitation hospitalized in the Department of Cardiovascular Disease of the First Affiliated Hospital of Guangzhou University of Traditional Medicine were enrolled. Three hundred fifty‐six cases were selected as the training set for modelling, and 153 cases were selected as the validation set for the internal validation of the model. Results A predictive model of heart failure with the following nine risk factors was developed: atrial fibrillation (AF), pulmonary infection (PI), coronary artery disease (CAD), creatinine (CREA), low‐density lipoprotein cholesterol (LDL‐C), d‐dimer (DDi), left ventricular end‐diastolic diameter (LVEDd), mitral regurgitation (MR) and aortic regurgitation (AR). The model was evaluated by the C‐index [the training set: area under curve (AUC) 0.937, 95% confidence interval (CI) 0.911–0.963; the validation set: AUC 0.928, 95% CI 0.890–0.967]. Hosmer–Lemeshow test (the training set: χ2 10.908, P = 0.207; the validation set: χ2 4.896, P = 0.769) revealed that both the training and validation sets performed well in terms of model differentiation and calibration. Decision curve analysis showed that both the training and validation sets have higher net benefits, indicating that the model has good utility. Ten‐fold cross‐validation showed that the training set has high similarities with the validation set, which means that the model has good stability. Conclusions The occurrence of heart failure in patients with valvular regurgitation has a significant correlation with AF, PI, CAD, CREA, LDL‐C, DDi, LVEDd, MR and AR. Based on these risk factors, a prediction model for heart failure was developed and validated, which showed good differentiation and utility, high accuracy and stability, providing a method for predicting heart failure.

Objective: The unique metastatic pattern of skip lateral lymph node metastasis (SLLNM) in PTC patients may lead to missed diagnosis of lateral cervical metastatic lymph nodes. Therefore, many different SLLNM prediction models were constructed. In this study, partially eligible models (Hu 2020, Wang 2020, and Zhao 2023 nomograms) were selected for external validation, and then new variables were incorporated for model reconstruction to extend clinical applicability. Methods: 576 PTC patients from our center were selected to evaluate the performance of the three nomograms using the receiver operating characteristic curve (ROC), calibration curves, and decision curve analyses (DCA). Three new variables were added to calibrate the model, including assessment of LN status on ultrasound (US-SLLNM), the distance from the tumor to the capsule (Capsular distance), and the number of central lymph node dissections (CLND number). Univariate and multivariate logistic regression analyses were used to screen independent predictors to reconstruct the model, and 1000 Bootstrap internal validations were performed. Results: SLLNM were present in 69/576 patients (12.0%). In external validation, the area under the ROC curves (AUCs) for Hu 2020, Wang 2020, and Zhao 2023 nomograms were 0.695 (95% CI:0.633-0.766), 0.792 (95% CI=0.73-0.845), and 0.769 (95% CI:0.713-0.824), respectively. The calibration curves for the three models were overall poorly fitted; DCA showed some net clinical benefit. Model differentiation and net clinical benefit improved by adding three new variables. Based on multivariate analysis, female, age, and maximum tumor diameter = 10 mm, located at the upper pole, Capsular distance < 0mm, US-SLLNM, CLND number = 5 were identified as independent predictors of SLLNM and were used to construct the new model. After 1000 Bootstrap internal validations, the mean AUC of the model was 0.870 (95% CI:0.839-0.901), the calibration curve was close to the ideal curve, and the net clinical benefit was significant. Conclusion: Overall, these nomograms were well differentiated and provided some net clinical benefit, but with varying degrees of underestimation or overestimation of the actual risk and high false-negative rates. New dynamic nomogram was constructed based on the addition of new variables and larger samples, showing better performance. [ABSTRACT FROM AUTHOR]

Objective To construct a postpartum depression risk prediction model for multi - ethnic population in Yunnan Province of China, and identify predictive factors. Methods Women who were 42 days and within 1 year after childbirth were screened, and the Edinburgh Postnatal Depression Scale (EPDS M 9) was used for postpartum depression. 52 influencing factors from economics, social psychology, obstetrics, neonatology, spouse and family dynamics and other characteristics were included in the survey. A random forest algorithm was employed to construct a predictive model for postnatal depression risk in the multi 一 ethnic population of Yunnan Province. The model was evaluated on test sets with accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and the area under the receiver operating characteristic curve (Area Under Curve, AUC) to assess its performance. Results A total of 459 women were analyzed, with a postpartum depression detection rate of 11. 55%. Among them, the detection rates for Han, Zhuang and other ethnic minorities were 7. 56%, 13. 94% and 13.92%, respectively. The top 14 variables in terms of importance scores were: anxiety, history of previous negative emotions, marital relationship, family support level, physical and mental exhaustion in caring for newborns, pregnancy risk classification, mother - infant rooming - in, feeding mode, education level, spouse's education level, frequency of nighttime newborn care, ethnicity, parity and age. The accuracy was 92.74%, specificity was 95. 50%, sensitivity was 69.23%, positive predictive value was 64. 29%, negative predictive value was 96. 36%, and the AUC value was 0. 925, using Han, Zhuang, and other ethnic minorities as validation sets respectively. The model also demonstrated good stability. Conclusion The random forest algorithm 一 based postpartum depression risk prediction model for the multi 一 ethnic population in Yunnan performed well, which can be utilized to predict risk factors for postpartum depression among women in minority ethnic areas, thereby facilitating targeted intervention measures. [ABSTRACT FROM AUTHOR]

Abstract Background People with opioid use disorder have substantially higher standardised mortality rates compared to the general population; however, lack of clear individual prognostic information presents challenges to prioritise or target interventions within drug treatment services. Previous prognostic models have been developed to estimate the risk of developing opioid use disorder and opioid-related overdose in people routinely prescribed opioids but, to our knowledge, none have been developed to estimate mortality risk in people accessing drug services with opioid use disorder. Initial presentation to drug services is a pragmatic time to evaluate mortality risk given the contemporaneous routine collection of prognostic indicators and as a decision point for appropriate service prioritisation and targeted intervention delivery. This study aims to develop and internally validate a model to estimate 6-month mortality risk for people with opioid use disorder from prognostic indicators recorded at initial assessment in drug services in England. Methods An English national dataset containing records from individuals presenting to drug services between 1 April 2013 and 1 April 2023 (n > 800,000) (the National Drug Treatment Monitoring System (NDTMS)) linked to their lifetime hospitalisation and death records (Hospital Episode Statistics-Office of National Statistics (HES-ONS)). Twelve candidate prognostic indicator variables were identified based on literature review of demographic and clinical features associated with increased mortality for people in treatment for opioid use disorder. Variables will be extracted at initial presentation to drug services with mortality measured at 6 months. Two multivariable Cox regression models will be developed one for 6-month all-cause mortality and one for 6-month drug-related mortality using backward elimination with a fractional polynomial approach for continuous variables. Internal validation will be undertaken using bootstrapping methods. Discrimination of both models will be reported using Harrel’s c and d-statistics. Calibration curves and slopes will be presented comparing expected and observed event rates. Discussion The models developed and internally validated in this study aim to improve clinical assessment of mortality risk for people with opioid use disorder presenting to drug services in England. External validation in different populations will be required to develop the model into a tool to assist future clinical decision-making.

Study designA retrospective study at a single academic institution.ObjectiveThe purpose of this study is to utilize machine learning to predict hospital length of stay (LOS) and discharge disposition following adult elective spine surgery, and to compare performance metrics of machine learning models to the American College of Surgeon's National Surgical Quality Improvement Program's (ACS NSQIP) prediction calculator.Summary of background dataA total of 3678 adult patients undergoing elective spine surgery between 2014 and 2019, acquired from the electronic health record.MethodsPatients were divided into three stratified cohorts: cervical degenerative, lumbar degenerative, and adult spinal deformity groups. Predictive variables included demographics, body mass index, surgical region, surgical invasiveness, surgical approach, and comorbidities. Regression, classification trees, and least absolute shrinkage and selection operator (LASSO) were used to build predictive models. Validation of the models was conducted on 16% of patients (N=587), using area under the receiver operator curve (AUROC), sensitivity, specificity, and correlation. Patient data were manually entered into the ACS NSQIP online risk calculator to compare performance. Outcome variables were discharge disposition (home vs. rehabilitation) and LOS (days).ResultsOf 3678 patients analyzed, 51.4% were male (n=1890) and 48.6% were female (n=1788). The average LOS was 3.66 days. In all, 78% were discharged home and 22% discharged to rehabilitation. Compared with NSQIP (Pearson R2 =0.16), the predictions of poisson regression ( R2 =0.29) and LASSO ( R2 =0.29) models were significantly more correlated with observed LOS ( P =0.025 and 0.004, respectively). Of the models generated to predict discharge location, logistic regression yielded an AUROC of 0.79, which was statistically equivalent to the AUROC of 0.75 for NSQIP ( P =0.135).ConclusionThe predictive models developed in this study can enable accurate preoperative estimation of LOS and risk of rehabilitation discharge for adult patients undergoing elective spine surgery. The demonstrated models exhibited better performance than NSQIP for prediction of LOS and equivalent performance to NSQIP for prediction of discharge location.

Background: Most surveillance systems for catheter-related bloodstream infections (CRBSI) and central line-associated bloodstream infections (CLABSI) are based on manual chart review. Our objective was to validate a fully automated algorithm for CRBSI and CLABSI surveillance in intensive care units (ICU). Methods: We developed a fully automated algorithm to detect CRBSI, CLABSI and ICU-onset bloodstream infections (ICU-BSI) in patients admitted to the ICU of a tertiary care hospital in Switzerland. The parameters included in the algorithm were based on a recently performed systematic review. Structured data on demographics, administrative data, central vascular catheter and microbiological results (blood cultures and other clinical cultures) obtained from the hospital's data warehouse were processed by the algorithm. Validation for CRBSI was performed by comparing results with prospective manual BSI surveillance data over a 6-year period. CLABSI were retrospectively assessed over a 2-year period. Results: From January 2016 to December 2021, 854 positive blood cultures were identified in 346 ICU patients. The median age was 61.7 years [IQR 50–70]; 205 (24%) positive samples were collected from female patients. The algorithm detected 5 CRBSI, 109 CLABSI and 280 ICU-BSI. The overall CRBSI and CLABSI incidence rates determined by automated surveillance for the period 2016 to 2021 were 0.18/1000 catheter-days (95% CI 0.06–0.41) and 3.86/1000 catheter days (95% CI: 3.17–4.65). The sensitivity, specificity, positive predictive and negative predictive values of the algorithm for CRBSI, were 83% (95% CI 43.7–96.9), 100% (95% CI 99.5–100), 100% (95% CI 56.5–100), and 99.9% (95% CI 99.2–100), respectively. One CRBSI was misclassified as an ICU-BSI by the algorithm because the same bacterium was identified in the blood culture and in a lower respiratory tract specimen. Manual review of CLABSI from January 2020 to December 2021 (n = 51) did not identify any errors in the algorithm. Conclusions: A fully automated algorithm for CRBSI and CLABSI detection in critically-ill patients using only structured data provided valid results. The next step will be to assess the feasibility and external validity of implementing it in several hospitals with different electronic health record systems. [ABSTRACT FROM AUTHOR]

Background: The coronavirus disease 2019 (COVID-19) pandemic has found the whole world unprepared for its correct management. Italy was the first European country to experience the spread of the SARS-CoV-2 virus at the end of February 2020. As a result of hospital overcrowding, the quality of care delivered was not always optimal. A substantial number of patients admitted to non-ICU units could have been treated at home. It would have been extremely useful to have a score that, based on personal and clinical characteristics and simple blood tests, could have predicted with sufficient reliability the probability that a patient had or did not have a disease that could have led to their death. This study aims to develop a scoring system to identify which patients with COVID-19 are at high mortality risk upon hospital admission, to expedite and enhance clinical decision making. Methods: A retrospective analysis was performed to develop a multivariable prognostic prediction model. Results: Derivation and external validation cohorts were obtained from two Italian University Hospital databases, including 388 (10.31% deceased) and 1357 (7.68% deceased) patients with confirmed COVID-19, respectively. A multivariable logistic model was used to select seven variables associated with in-hospital death (age, baseline oxygen saturation, hemoglobin value, white blood cell count, percentage of neutrophils, platelet count, and creatinine value). Calibration and discrimination were satisfactory with a cumulative AUC for prediction mortality of 0.924 (95% CI: 0.893–0.944) in derivation cohorts and 0.808 (95% CI: 0.886–0.828) in external validation cohorts. The risk score obtained was compared with the ISARIC 4C Mortality Score, and with all the other most important scores considered so far, to evaluate the risk of death of patients with COVID-19. It performed better than all the above scores to evaluate the predictability of dying. Its sensitivity, specificity, and AUC were higher than the other COVID-19 scoring systems when the latter were calculated for the 388 patients in our derivation cohort. Conclusions: In conclusion, the CZ-COVID-19 Score may help all physicians by identifying those COVID-19 patients who require more attention to provide better therapeutic regimens or, on the contrary, by identifying those patients for whom hospitalization is not necessary and who could therefore be sent home without overcrowding healthcare facilities. We developed and validated a new risk score based on seven variables for upon-hospital admission of COVID-19 patients. It is very simple to calculate and performs better than all the other similar scores to evaluate the predictability of dying. [ABSTRACT FROM AUTHOR]

With an increasing number of web-based calculators designed to provide the probabilities of an individual achieving improvement after lumbar spine surgery, there is a need to determine the accuracy of these models. To perform an internal and external validation study of the reduced Quality Outcomes Database web-based Calculator (QOD-Calc). Observational longitudinal cohort. Patients enrolled study-wide in Quality Outcomes Database (QOD) and patients enrolled in DaneSpine at a single institution who had elective lumbar spine surgery with baseline data to complete QOD-Calc and 12-month postoperative data. Oswestry Disability Index (ODI), Numeric Rating Scales (NRS) for back and leg pain, EuroQOL-5D (EQ-5D). Baseline data elements were entered into QOD-Calc to determine the probability for each patient having Any Improvement and 30% Improvement in NRS leg pain, back pain, EQ-5D and ODI. These probabilities were compared with the actual 12-month postop data for each of the QOD and DaneSpine cases. Receiver-operating characteristics analyses were performed and calibration plots created to assess model performance. 24,755 QOD cases and 8,105 DaneSpine lumbar cases were included in the analysis. QOD-Calc had acceptable to outstanding ability (AUC: 0.694–0.874) to predict Any Improvement in the QOD cohort and moderate to acceptable ability (AUC: 0.658–0.747) to predict 30% Improvement. QOD-Calc had acceptable to exceptional ability (AUC: 0.669–0.734) to predict Any improvement and moderate to exceptional ability (AUC: 0.619–0.862) to predict 30% Improvement in the DaneSpine cohort. AUCs for the DaneSpine cohort was consistently lower that the AUCs for the QOD validation cohort. QOD-Calc performs well in predicting outcomes in a patient population that is similar to the patients that was used to develop it. Although still acceptable, model performance was slightly worse in a distinct population, despite the fact that the sample was more homogenous. Model performance may also be attributed to the low discrimination threshold, with close to 90% of cases reporting Any Improvement in outcome. Prediction models may need to be developed that are highly specific to the characteristics of the population. [ABSTRACT FROM AUTHOR]

Background and purpose: The study’s purpose was to develop and internally validate a prognostic survival model exploring baseline variables for adolescent idiopathic scoliosis curve progression. Methods: A longitudinal prognostic cohort analysis was performed on trial data (n = 135) including girls and boys, Cobb angle 25–40°, aged 9–17 years, remaining growth > 1 year, and previously untreated. Prognostic outcome was defined as curve progression of Cobb angle of > 6° prior to skeletal maturity. 34 candidate prognostic variables were tested. Time-to-event was measured with 6-month intervals. Cox proportional hazards regression survival model (CoxPH) was used for model development and validation in comparison with machine learning models (66.6/33.3 train/test data set). The models were adjusted for treatment exposure. Results: The final primary prognostic model included 127 patients, predicting progress with acceptable discriminative ability (concordance = 0.79, 95% confidence interval [CI] 0.72–0.86). Significant prognostic risk factors were Risser stage of 0 (HR 4.6, CI 2.1–10.1, P < 0.001), larger major curve Cobb angle (HRstandardized 1.5, CI 1.1–2.0, P = 0.005), and higher score on patient-reported pictorial Spinal Appearance Questionnaire (pSAQ) (HRstandardized 1.4, CI 1.0–1.9, P = 0.04). Treatment exposure, entered as a covariate adjustment, contributed significantly to the final model (HR 3.1, CI 1.5–6.0, P = 0.001). Sensitivity analysis displayed that CoxPH maintained acceptable discriminative ability (AUC 0.79, CI 0.65–0.93) in comparison with machine learning algorithms. Conclusion: The prognostic model (Risser stage, Cobb angle, pSAQ, and menarche) predicted curve progression of > 6° Cobb angle with acceptable discriminative ability. Adding patient report of the pSAQ may be of clinical importance for the prognosis of curve progression.

Yan Fang,1,2 Huali Liu,1,2 Yuan Li,1,2 Ji Cheng,1 Xia Wang,1 Bing Shen,3 Hongbo Chen,1,2 Qunhua Wang4 1Department of Obstetrics and Gynaecology, Maternal and Child Health Hospital Affiliated to Anhui Medical University, Hefei, People’s Republic of China; 2The Fifth Clinical College of Anhui Medical University, Hefei, People’s Republic of China; 3School of Basic Medicine, Anhui Medical University, Hefei, People’s Republic of China; 4Department of Obstetrics and Gynaecology, the First Affiliated Hospital of USTC, Hefei, People’s Republic of ChinaCorrespondence: Hongbo Chen, Maternal and Child Health Hospital Affiliated to Anhui Medical University, No. 15 Yimin Street, Luyang District, Hefei City, Anhui Province, People’s Republic of China, Email chenhongbo@ahmu.edu.cn Qunhua Wang, Department of Obstetrics and Gynaecology, the First Affiliated Hospital of USTC, Hefei, People’s Republic of China, Email wqh971234@163.comPurpose: To investigate the risk factors associated with preeclampsia in hyperglycemic pregnancies and develop a predictive model based on routine pregnancy care.Patients and Methods: The retrospective collection of clinical data was performed on 951 pregnant women with hyperglycemia, including those diagnosed with diabetes in pregnancy (DIP) and gestational diabetes mellitus (GDM), who delivered after 34 weeks of gestation at the Maternal and Child Health Hospital Affiliated to Anhui Medical University between January 2017 and December 2019. Observation indicators included liver and kidney function factors testing at 24– 29+6 weeks gestation, maternal age, and basal blood pressure. The indicators were screened univariately, and the “rms” package in R language was applied to explore the factors associated with PE in HIP pregnancy by stepwise regression. Multivariable logistic regression analysis was used to develop the prediction model. Based on the above results, a nomogram was constructed to predict the risk of PE occurrence in pregnant women with HIP. Then, the model was evaluated from three aspects: discrimination, calibration, and clinical utility. The internal validation was performed using the bootstrap procedure.Results: Multivariate logistic regression analysis showed that cystatin C, uric acid, glutamyl aminotransferase, blood urea nitrogen, and basal systolic blood pressure as predictors of PE in pregnancy with HIP. The predictive model yielded an area under curve (AUC) value of 0.8031 (95% CI: 0.7383– 0.8679), with an optimal threshold of 0.0805, at which point the sensitivity was 0.8307 and specificity of 0.6604. Hosmer–Lemeshow test values were P = 0.3736, Brier score value was 0.0461. After 1000 Bootstrap re-samplings for internal validation, the AUC was 0.7886, the Brier score was 0.0478 and the predicted probability of the calibration curve was similar to the actual probability. A nomogram was constructed based on the above to visualize the model.Conclusion: This study developed a model for predicting PE in pregnant women with HIP, achieving high predictive performance of PE risk through the information of routine pregnancy care.Keywords: hyperglycemia in pregnancy, preeclampsia, prediction model, nomogram, internal validation

The GA118-24B Genetic Analyzer (hereafter, "GA118-24B") is an independently developed capillary electrophoresis instrument. In the present research, we designed a series of validation experiments to test its performance at detecting DNA fragments compared to the Applied Biosystems 3500 Genetic Analyzer (hereafter, "3500"). Three commercially available autosomal short tandem repeat multiplex kits were used in this validation. The results showed that GA118-24B had acceptable spectral calibration for three kits. The results of accuracy and concordance studies were also satisfactory. GA118-24B showed excellent precision, with a standard deviation of less than 0.1 bp. Sensitivity and mixture studies indicated that GA118-24B could detect low-template DNA and complex mixtures as well as the results generated by 3500 in parallel experiments. Based on the experimental results, we set specific analytical and stochastic thresholds. Besides, GA118-24B showed superiority than 3500 within certain size ranges in the resolution study. Instead of conventional commercial multiplex kits, GA118-24B performed stably on a self-developed eight-dye multiplex system, which were not performed on 3500 Genetic Analyzer. We compared our validation results with those of previous research and found our results to be convincing. Overall, we conclude that GA118-24B is a stable and reliable genetic analyzer for forensic DNA identification. [ABSTRACT FROM AUTHOR]

Abstract Background Topic models are a class of unsupervised machine learning models, which facilitate summarization, browsing and retrieval from large unstructured document collections. This study reviews several methods for assessing the quality of unsupervised topic models estimated using non-negative matrix factorization. Techniques for topic model validation have been developed across disparate fields. We synthesize this literature, discuss the advantages and disadvantages of different techniques for topic model validation, and illustrate their usefulness for guiding model selection on a large clinical text corpus. Design, setting and data Using a retrospective cohort design, we curated a text corpus containing 382,666 clinical notes collected between 01/01/2017 through 12/31/2020 from primary care electronic medical records in Toronto Canada. Methods Several topic model quality metrics have been proposed to assess different aspects of model fit. We explored the following metrics: reconstruction error, topic coherence, rank biased overlap, Kendall’s weighted tau, partition coefficient, partition entropy and the Xie-Beni statistic. Depending on context, cross-validation and/or bootstrap stability analysis were used to estimate these metrics on our corpus. Results Cross-validated reconstruction error favored large topic models (K ≥ 100 topics) on our corpus. Stability analysis using topic coherence and the Xie-Beni statistic also favored large models (K = 100 topics). Rank biased overlap and Kendall’s weighted tau favored small models (K = 5 topics). Few model evaluation metrics suggested mid-sized topic models (25 ≤ K ≤ 75) as being optimal. However, human judgement suggested that mid-sized topic models produced expressive low-dimensional summarizations of the corpus. Conclusions Topic model quality indices are transparent quantitative tools for guiding model selection and evaluation. Our empirical illustration demonstrated that different topic model quality indices favor models of different complexity; and may not select models aligning with human judgment. This suggests that different metrics capture different aspects of model goodness of fit. A combination of topic model quality indices, coupled with human validation, may be useful in appraising unsupervised topic models.

The onus of proof in criminal cases is beyond any reasonable doubt, and the issue on the lack of complete internal validation data can be manipulated when it comes to justifying the validity and reliability of the X‐chromosomal short tandem repeats analysis for court representation. Therefore, this research evaluated the efficiency of the optimized 60% reduced volumes for polymerase chain reaction (PCR) amplification using the Qiagen Investigator® Argus X‐12 QS Kit, as well as the capillary electrophoresis (CE) sample preparation for blood samples on Flinder's Technology Associates (FTA) cards. Good‐quality DNA profile (3000–12,000 RFU) from the purified blood sample on FTA card (1.2 mm) were obtained using the optimized PCR (10.0 μL of PCR reaction volume and 21 cycles) and CE (9.0 μL Hi‐Di™ Formamide and 0.3 μL DNA Size Standard 550 [BTO] and 27 s injection time) conditions. The analytical and stochastic thresholds were 100 and 200 RFU, respectively. Hence, the internal validation data supported the use of the optimized 60% reduced PCR amplification reaction volume of the Qiagen Investigator® Argus X‐12 QS Kit as well as the CE sample preparation for producing reliable DNA profiles that comply with the quality assurance standards for forensic DNA testing laboratories, while optimizing the analytical cost. [ABSTRACT FROM AUTHOR]

Abstract Background Complications from preterm birth (PTB) are the leading cause of death and disability in those under five years. Whilst the role of omega-3 (n-3) supplementation in reducing PTB is well-established, growing evidence suggests supplementation use in those replete may increase the risk of early PTB. Aim: To develop a non-invasive tool to identify individuals with total n-3 serum levels above 4.3% of total fatty acids in early pregnancy. Methods We conducted a prospective observational study recruiting 331 participants from three clinical sites in Newcastle, Australia. Eligible participants (n = 307) had a singleton pregnancy between 8 and 20 weeks’ gestation at recruitment. Data on factors associated with n-3 serum levels were collected using an electronic questionnaire; these included estimated intake of n-3 (including food type, portion size, frequency of consumption), n-3 supplementation, and sociodemographic factors. The optimal cut-point of estimated n-3 intake that predicted mothers with total serum n-3 levels likely above 4.3% was developed using multivariate logistic regression, adjusting for maternal age, body mass index, socioeconomic status, and n-3 supplementation use. Total serum n-3 levels above 4.3% was selected as previous research has demonstrated that mothers with these levels are at increased risk of early PTB if they take additional n-3 supplementation during pregnancy. Models were evaluated using various performance metrics including sensitivity, specificity, area under receiver operator characteristic (AUROC) curve, true positive rate (TPR) at 10% false positive rate (FPR), Youden Index, Closest to (0,1) Criteria, Concordance Probability, and Index of Union. Internal validation was performed using 1000-bootstraps to generate 95% confidence intervals for performance metrics generated. Results Of 307 eligible participants included for analysis, 58.6% had total n-3 serum levels above 4.3%. The optimal model had a moderate discriminative ability (AUROC 0.744, 95% CI 0.742–0.746) with 84.7% sensitivity, 54.7% specificity and 37.6% TPR at 10% FPR. Conclusions Our non-invasive tool was a moderate predictor of pregnant women with total serum n-3 levels above 4.3%; however, its performance is not yet adequate for clinical use. Trial registration This trial was approved by the Hunter New England Human Research Ethics Committee of the Hunter New England Local Health District (Reference 2020/ETH00498 on 07/05/2020 and 2020/ETH02881 on 08/12/2020).

Background: Perineal trauma is a common complication of childbirth and can have serious impacts on long-term health. Few studies have examined the combined effect of multiple risk factors. We developed and internally validated a risk prediction model to predict third and fourth degree perineal tears using data from a general obstetric population. Methods: Risk prediction model using data from all singleton vaginal deliveries at Cork University Maternity Hospital (CUMH), Ireland during 2019 and 2020. Third/fourth degree tears were diagnosed by an obstetrician or midwife at time of birth and defined as tears that extended into the anal sphincter complex or involved both the anal sphincter complex and anorectal mucosa. We used univariable and multivariable logistic regression with backward stepwise selection to develop the models. Candidate predictors included infant sex, maternal age, maternal body mass index, parity, mode of delivery, birthweight, post-term delivery, induction of labour and public/private antenatal care. We used the receiver operating characteristic (ROC) curve C-statistic to assess discrimination, and bootstrapping techniques were used to assess internal validation. Results: Of 8,403 singleton vaginal deliveries, 8,367 (99.54%) had complete data on predictors for model development. A total of 128 women (1.53%) had a third/fourth degree tear. Three variables remained in the final model: nulliparity, mode of delivery (specifically forceps delivery or ventouse delivery) and increasing birthweight (per 100 gram increase) (C-statistic: 0.75, 95% CI: 0.71, 0.79). We developed a nomogram to calculate individualised risk of third/fourth degree tears using these predictors. Bootstrapping indicated good internal performance. Conclusions: Use of our nomogram can provide an individualised risk assessment of third/fourth degree tears and potentially aid counselling of women on their potential risk.

Background: Topic models are a class of unsupervised machine learning models, which facilitate summarization, browsing and retrieval from large unstructured document collections. This study reviews several methods for assessing the quality of unsupervised topic models estimated using non-negative matrix factorization. Techniques for topic model validation have been developed across disparate fields. We synthesize this literature, discuss the advantages and disadvantages of different techniques for topic model validation, and illustrate their usefulness for guiding model selection on a large clinical text corpus. Design, setting and data: Using a retrospective cohort design, we curated a text corpus containing 382,666 clinical notes collected between 01/01/2017 through 12/31/2020 from primary care electronic medical records in Toronto Canada. Methods: Several topic model quality metrics have been proposed to assess different aspects of model fit. We explored the following metrics: reconstruction error, topic coherence, rank biased overlap, Kendall's weighted tau, partition coefficient, partition entropy and the Xie-Beni statistic. Depending on context, cross-validation and/or bootstrap stability analysis were used to estimate these metrics on our corpus. Results: Cross-validated reconstruction error favored large topic models (K ≥ 100 topics) on our corpus. Stability analysis using topic coherence and the Xie-Beni statistic also favored large models (K = 100 topics). Rank biased overlap and Kendall's weighted tau favored small models (K = 5 topics). Few model evaluation metrics suggested mid-sized topic models (25 ≤ K ≤ 75) as being optimal. However, human judgement suggested that mid-sized topic models produced expressive low-dimensional summarizations of the corpus. Conclusions: Topic model quality indices are transparent quantitative tools for guiding model selection and evaluation. Our empirical illustration demonstrated that different topic model quality indices favor models of different complexity; and may not select models aligning with human judgment. This suggests that different metrics capture different aspects of model goodness of fit. A combination of topic model quality indices, coupled with human validation, may be useful in appraising unsupervised topic models. [ABSTRACT FROM AUTHOR]

Abstract Background Clinical prediction models should be validated before implementation in clinical practice. But is favorable performance at internal validation or one external validation sufficient to claim that a prediction model works well in the intended clinical context? Main body We argue to the contrary because (1) patient populations vary, (2) measurement procedures vary, and (3) populations and measurements change over time. Hence, we have to expect heterogeneity in model performance between locations and settings, and across time. It follows that prediction models are never truly validated. This does not imply that validation is not important. Rather, the current focus on developing new models should shift to a focus on more extensive, well-conducted, and well-reported validation studies of promising models. Conclusion Principled validation strategies are needed to understand and quantify heterogeneity, monitor performance over time, and update prediction models when appropriate. Such strategies will help to ensure that prediction models stay up-to-date and safe to support clinical decision-making.

Background: Complications from preterm birth (PTB) are the leading cause of death and disability in those under five years. Whilst the role of omega-3 (n-3) supplementation in reducing PTB is well-established, growing evidence suggests supplementation use in those replete may increase the risk of early PTB. Aim: To develop a non-invasive tool to identify individuals with total n-3 serum levels above 4.3% of total fatty acids in early pregnancy. Methods: We conducted a prospective observational study recruiting 331 participants from three clinical sites in Newcastle, Australia. Eligible participants (n = 307) had a singleton pregnancy between 8 and 20 weeks' gestation at recruitment. Data on factors associated with n-3 serum levels were collected using an electronic questionnaire; these included estimated intake of n-3 (including food type, portion size, frequency of consumption), n-3 supplementation, and sociodemographic factors. The optimal cut-point of estimated n-3 intake that predicted mothers with total serum n-3 levels likely above 4.3% was developed using multivariate logistic regression, adjusting for maternal age, body mass index, socioeconomic status, and n-3 supplementation use. Total serum n-3 levels above 4.3% was selected as previous research has demonstrated that mothers with these levels are at increased risk of early PTB if they take additional n-3 supplementation during pregnancy. Models were evaluated using various performance metrics including sensitivity, specificity, area under receiver operator characteristic (AUROC) curve, true positive rate (TPR) at 10% false positive rate (FPR), Youden Index, Closest to (0,1) Criteria, Concordance Probability, and Index of Union. Internal validation was performed using 1000-bootstraps to generate 95% confidence intervals for performance metrics generated. Results: Of 307 eligible participants included for analysis, 58.6% had total n-3 serum levels above 4.3%. The optimal model had a moderate discriminative ability (AUROC 0.744, 95% CI 0.742–0.746) with 84.7% sensitivity, 54.7% specificity and 37.6% TPR at 10% FPR. Conclusions: Our non-invasive tool was a moderate predictor of pregnant women with total serum n-3 levels above 4.3%; however, its performance is not yet adequate for clinical use. Trial registration: This trial was approved by the Hunter New England Human Research Ethics Committee of the Hunter New England Local Health District (Reference 2020/ETH00498 on 07/05/2020 and 2020/ETH02881 on 08/12/2020). [ABSTRACT FROM AUTHOR]

Clustering approaches are extensively used by many areas such as IR, Data Integration, Document Classification, Web Mining, Query Processing, and many other domains and disciplines. Nowadays, much literature describes clustering algorithms on multivariate data sets. However, there is limited literature that presented them with exhaustive and extensive theoretical analysis as well as experimental comparisons. This experimental survey paper deals with the basic principle, and techniques used, including important characteristics, application areas, run-time performance, internal, external, and stability validity of cluster quality, etc., on five different data sets of eleven clustering algorithms. This paper analyses how these algorithms behave with five different multivariate data sets in data representation. To answer this question, we compared the efficiency of eleven clustering approaches on five different data sets using three validity metrics-internal, external, and stability and found the optimal score to know the feasible solution of each algorithm. In addition, we have also included four popular and modern clustering algorithms with only their theoretical discussion. Our experimental results for only traditional clustering algorithms showed that different algorithms performed different behavior on different data sets in terms of running time (speed), accuracy and, the size of data set. This study emphasized the need for more adaptive algorithms and a deliberate balance between the running time and accuracy with their theoretical as well as implementation aspects. [ABSTRACT FROM AUTHOR]

The purpose of this study was to identify paternity relations and increase the likelihood success in DNA identification of biological samples found in mass disasters and criminal events by using the 6-dye GlobalFilerTM amplification kit with 21 autosomal STR loci in our laboratory for validation purpose and routine use. In this regard, swab samples from different regions of Turkey (n=350) were collected and studied. The validation of the GlobalFilerTM Amplification Kit was carried out in accordance with the guidelines published by the Scientific Working Group on DNA Analysis Methods (SWGDAM). Our study was conducted to show the efficacy and performance of the kit for sensitivity, specificity, limit of detection, dynamic range, limit of quantification, stochastic threshold study, reproducibility and repeatability, mixture and contamination study parameters. Validation parameters and population genetics data were calculated by using Arlequin v 3.5.2.2 version. According to our results, SE33 (PM=0.014) locus showed the greatest power of discrimination and TPOX (PM=0.132) has the least power of discrimination power in Turkish population. Combined power of exclusion and combined power of discrimination were calculated as 99.99999963% and 99.99999999%, respectively. In addition, allelic frequencies from our study were compared to previously published population data of Iraq, Saud Arabia, China (Han), USA (Cauc), Iran, Afghan, Azerbaijan, Romania, and South Africa. The study revealed that there was a significant difference between Turkish and South Africa (0.1198) population and no difference was observed for Azerbaijan (0.0097) population. [ABSTRACT FROM AUTHOR]

Abstract Aim Clinical prediction models need to be validated. In this study, we used simulation data to compare various internal and external validation approaches to validate models. Methods Data of 500 patients were simulated using distributions of metabolic tumor volume, standardized uptake value, the maximal distance between the largest lesion and another lesion, WHO performance status and age of 296 diffuse large B cell lymphoma patients. These data were used to predict progression after 2 years based on an existing logistic regression model. Using the simulated data, we applied cross-validation, bootstrapping and holdout (n = 100). We simulated new external datasets (n = 100, n = 200, n = 500) and simulated stage-specific external datasets (1), varied the cut-off for high-risk patients (2) and the false positive and false negative rates (3) and simulated a dataset with EARL2 characteristics (4). All internal and external simulations were repeated 100 times. Model performance was expressed as the cross-validated area under the curve (CV-AUC ± SD) and calibration slope. Results The cross-validation (0.71 ± 0.06) and holdout (0.70 ± 0.07) resulted in comparable model performances, but the model had a higher uncertainty using a holdout set. Bootstrapping resulted in a CV-AUC of 0.67 ± 0.02. The calibration slope was comparable for these internal validation approaches. Increasing the size of the test set resulted in more precise CV-AUC estimates and smaller SD for the calibration slope. For test datasets with different stages, the CV-AUC increased as Ann Arbor stages increased. As expected, changing the cut-off for high risk and false positive- and negative rates influenced the model performance, which is clearly shown by the low calibration slope. The EARL2 dataset resulted in similar model performance and precision, but calibration slope indicated overfitting. Conclusion In case of small datasets, it is not advisable to use a holdout or a very small external dataset with similar characteristics. A single small testing dataset suffers from a large uncertainty. Therefore, repeated CV using the full training dataset is preferred instead. Our simulations also demonstrated that it is important to consider the impact of differences in patient population between training and test data, which may ask for adjustment or stratification of relevant variables.

Postoperative recovery after adult spinal deformity (ASD) operations is arduous, fraught with complications, and often requires extended hospital stays. A need exists for a method to rapidly predict patients at risk for extended length of stay (eLOS) in the preoperative setting. To develop a machine learning model to preoperatively estimate the likelihood of eLOS following elective multi-level lumbar/thoracolumbar spinal instrumented fusions (≥3 segments) for ASD. Retrospectively from a state-level inpatient database hosted by the Health care cost and Utilization Project. Of 8,866 patients of age ≥50 with ASD undergoing elective lumbar or thoracolumbar multilevel instrumented fusions. The primary outcome was eLOS (>7 days). Predictive variables consisted of demographics, comorbidities, and operative information. Significant variables from univariate and multivariate analyses were used to develop a logistic regression-based predictive model that use six predictors. Model accuracy was assessed through area under the curve (AUC), sensitivity, and specificity. Of 8,866 patients met inclusion criteria. A saturated logistic model with all significant variables from multivariate analysis was developed (AUC=0.77), followed by generation of a simplified logistic model through stepwise logistic regression (AUC=0.76). Peak AUC was reached with inclusion of six selected predictors (combined anterior and posterior approach, surgery to both lumbar and thoracic regions, ≥8 level fusion, malnutrition, congestive heart failure, and academic institution). A cutoff of 0.18 for eLOS yielded a sensitivity of 77% and specificity of 68%. This predictive model can facilitate identification of adults at risk for eLOS following elective multilevel lumbar/thoracolumbar spinal instrumented fusions for ASD. With a fair diagnostic accuracy, the predictive calculator will ideally enable clinicians to improve preoperative planning, guide patient expectations, enable optimization of modifiable risk factors, facilitate appropriate discharge planning, stratify financial risk, and accurately identify patients who may represent high-cost outliers. Future prospective studies that validate this risk assessment tool on external datasets would be valuable. [ABSTRACT FROM AUTHOR]

Risk stratification and prognosis prediction are critical for appropriate management of anti-neutrophil cytoplasmic antibody (ANCA) associated vasculitis (AAV). Herein, we aim to develop and internally validate a prediction model specifically for long-term survival of patients with AAV. We thoroughly reviewed the medical charts of patients with AAV admitted to Peking Union Medical College Hospital from January 1999 to July 2019. The Least Absolute Shrinkage and Selection Operator method and the COX proportional hazard regression was used to develop the prediction model. The Harrell's concordance index (C-index), calibration curves and Brier scores were calculated to evaluate the model performance. The model was internally validated by bootstrap resampling methods. A total of 653 patients were included in the study, including 303 patients with microscopic polyangiitis, 245 patients with granulomatosis with polyangiitis and 105 patients with eosinophilic granulomatosis with polyangiitis, respectively. During a median follow-up of 33 months (interquartile range 15–60 months), 120 deaths occurred. Age at admission, chest and cardiovascular involvement, serum creatinine grade, hemoglobin levels at baseline and AAV sub-types were selected as predictive parameters in the final model. The optimism-corrected C-index and integrated Brier score of our prediction model were 0.728 and 0.109. The calibration plots showed fine agreement between observed and predicted probability of all-cause death. The decision curve analysis (DCA) showed that in a wide range of threshold probabilities, our prediction model had higher net benefits compared with the revised five factor score (rFFSand) and the birmingham vasculitis activity score (BVAS) system. Our model performs well in predicting outcomes of AAV patients. Patients with moderate-to-high probability of death should be followed closely and personalized monitoring plan should be scheduled. [ABSTRACT FROM AUTHOR]

Background: Clinical prediction models should be validated before implementation in clinical practice. But is favorable performance at internal validation or one external validation sufficient to claim that a prediction model works well in the intended clinical context? Main body: We argue to the contrary because (1) patient populations vary, (2) measurement procedures vary, and (3) populations and measurements change over time. Hence, we have to expect heterogeneity in model performance between locations and settings, and across time. It follows that prediction models are never truly validated. This does not imply that validation is not important. Rather, the current focus on developing new models should shift to a focus on more extensive, well-conducted, and well-reported validation studies of promising models. Conclusion: Principled validation strategies are needed to understand and quantify heterogeneity, monitor performance over time, and update prediction models when appropriate. Such strategies will help to ensure that prediction models stay up-to-date and safe to support clinical decision-making. [ABSTRACT FROM AUTHOR]

Acute kidney injury (AKI) is a common complication after cardiac surgery. This study aims to develop and validate a risk model for predicting AKI after cardiac valve replacement surgery. Data from patients undergoing surgical valve replacement between January 2015 and December 2018 in our hospital were retrospectively analyzed. The subjects were randomly divided into a derivation cohort and a validation cohort at a ratio of 7:3. The primary outcome was defined as AKI within 7 days after surgery. Logistic regression analysis was conducted to select risk predictors for developing the prediction model. Receiver operator characteristic curve (ROC), calibration plot and clinical decision curve analysis (DCA) will be used to evaluate the discrimination, precision and clinical benefit of the prediction model. A total of 1159 patients were involved in this study. The prevalence of AKI following surgery was 37.0% (429/1159). Logistic regression analysis showed that age, hemoglobin, fibrinogen, serum uric acid, cystatin C, bicarbonate, and cardiopulmonary bypass time were independent risk factors associated with AKI after surgical valve replacement (all P < 0.05). The areas under the ROC curves (AUCs) in the derivation cohort and the validation cohort were 0.777 (95% CI 0.744–0.810) and 0.760 (95% CI 0.706–0.813), respectively. The calibration plots indicated excellent consistency between the prediction probability and actual probability. DCA demonstrated great clinical benefit of the prediction model. We developed a prediction model for predicting AKI after cardiac valve replacement surgery that was internally validated to have good discrimination, calibration, and clinical practicability. • Few studies have focused on the prediction of AKI following surgical valve replacement. • This study aimed to develop and validate a risk model for the early prediction of AKI after cardiac valve replacement surgery. • Age, hemoglobin, fibrinogen, serum uric acid, cystatin C, bicarbonate, and cardiopulmonary bypass time are independent risk factors associated with AKI after surgical valve replacement. • We developed a prediction model for predicting AKI after cardiac valve replacement surgery based on 7 risk factors that were internally validated to have good discrimination, calibration, and clinical practicability. [ABSTRACT FROM AUTHOR]

Abstract Background Teachers with a teacher-centred perspective have difficulties applying student-centred approaches in Problem Based Learning (PBL) because they are inclined to show teacher-centred behaviours. The six aspects explained in Korthagen’s Onion Model (environment, behaviour, competencies, beliefs, identity, and mission) are assumed to contribute to teachers’ perspectives, showing that both the environment and personal characteristics influence behaviours. For teachers to function properly in PBL, those six aspects should reflect a student-centred perspective. Previous instruments to measure teaching perspectives focused on only a few of these relevant aspects. Therefore, we developed the Student-Centred Perspective of Teachers (SCPT) questionnaire with subscales for each aspect in the Onion Model. This study aimed to provide evidence for its internal and external validity. Methods The SCPT was distributed in a survey to 795 teachers from 20 medical schools. For the internal validation, Confirmatory Factor Analysis was performed to analyse theoretical fit model validation, convergent validation, and discriminant validation. For the external validation, teachers’ perspective scores were compared among three groups of amount of PBL training using Analysis of Variance (ANOVA) and post-hoc Least Significant Difference (LSD) tests. The p-value for all tests was set at .05. Results A total of 543 out of 795 teachers (68.3%) participated. Confirmatory Factor Analysis showed the evidence of the SCPT’s internal validation with acceptable fit for the six subscales measured by 19 items and the following Composite Reliability scores: environment (.72), behaviour (.74), competencies (.63), beliefs (.55), identity (.76), and mission (.60). All items’ factors loadings reached a good standard (.5 or greater). Only the environment subscale had the Average Variance Extracted (AVE) score higher than .5 and the Maximum Shared Variance score lower than the AVE score. ANOVA and Post-hoc LSD tests showed that teachers who participated in more PBL training showed significantly higher student-centred perspectives, providing evidence for external validity. Conclusion The SCPT is a reliable and valid instrument to measure teaching perspectives. Identifying aspects that do not represent the adoption of a student-centred perspective may provide valuable input for faculty development in the context of PBL.

Ting Yin,1 Shi Shi,1 Xu Zhu,1 Iokfai Cheang,1 Xinyi Lu,1 Rongrong Gao,1 Haifeng Zhang,1,2 Wenming Yao,1 Yanli Zhou,1 Xinli Li1 1Department of Cardiology, The First Affiliated Hospital of Nanjing Medical University, Jiangsu Province Hospital, Nanjing, 210029, People’s Republic of China; 2Department of Cardiology, The Affiliated Suzhou Hospital of Nanjing Medical University, Suzhou Municipal Hospital, Gusu School, Nanjing Medical University, Suzhou, 215002, People’s Republic of ChinaCorrespondence: Xinli Li; Yanli Zhou, Tel +86 136 1157 3111 ; +86 137 7787 9077, Email xinli3267@njmu.edu.cn; zhyl88@qq.comPurpose: The current study aimed to develop a convenient and accurate prognostic dynamic nomogram model for the risk of all-cause death in acute heart failure (AHF) patients that incorporates clinical characteristics including N-terminal pro-brain natriuretic peptide (NT-pro BNP) and growth stimulation expresses gene 2 protein (ST2).Patients and Methods: We prospectively studied 537 consecutive AHF patients and derived a clinical prediction model. The least absolute shrinkage and selection operator regression model combined with clinical characteristics were used for dimensional reduction and feature selection. Multivariate Cox proportional hazard analysis and “Dynnom” package were used to build the dynamic nomogram for prediction of 1-,2-,and 5-year overall survival for AHF. With bootstrap validation, the time-dependent concordance index (C-index) and calibration curves were used to assess predictive discrimination and accuracy. The contributions of NT-pro BNP and ST2 to the nomogram were evaluated using integrated discrimination improvement (IDI) and net reclassification improvement (NRI), while decision curve analysis (DCA) was used to assess clinical value.Results: Patients were randomly divided into derivation (74.9%, n=402) and validation (25.1%, n=135) cohorts. Optimal independent prognostic factors for 1-,2-, and 5-year all-cause mortality were BS-ACMR (B: NT-pro BNP; S: ST2; A: age; C: complete right bundle branch block; M: mean arterial pressure; and R: red cell distribution width > 14.5%); these were incorporated into the dynamic nomogram (https://bs-acmr-nom.shinyapps.io/dynnomapp/) with bootstrap validation. The C-indexes in the derivation (0.793) and validation (0.782) cohorts were consistent with comparable performance parameters. The calibration curve showed good agreement between the nomogram-predicted and actual survival. Adding NT-pro BNP and ST2 provided a significant net benefit and improved performance over other less adequate schemes in terms of DCA of survival probability compared to those neglecting either of these two factors.Conclusion: The study constructed a dynamic BS-ACMR nomogram, which is a convenient, practical and effective clinical decision-making tool for providing accurate prognosis in AHF patients.Keywords: acute heart failure, all-cause mortality, dynamic nomogram, prediction model, internal validation

Abstract Background Our aim was to create and validate a nomogram predicting cesarean delivery after induction of labor among nulliparous women at term. Methods Data were obtained from medical records from Nanjing Drum Tower Hospital. Nulliparous women with singleton pregnancies undergoing induction of labor at term were involved. A total of 2950 patients from Jan. 2014 to Dec. 2015 were served as derivation cohort. A nomogram was constructed by multivariate logistic regression using maternal, fetal and pregnancy characteristics. The predictive accuracy and discriminative ability of the nomogram were internal validated by 1000-bootstrap resampling, followed by external validation of a new dataset from Jan. 2016 to Dec. 2016. Results Logistic regression revealed nine predictors of cesarean delivery, including maternal height, age, uterine height, abdominal circumference, estimated fetal weight, indications for induction of labor, initial cervical consistency, cervical effacement and station. Nomogram was well calibrated and had an AUC of 0.73 (95% confidence interval [CI], 0.70-0.75) after bootstrap resampling for internal validation. The AUC in external validation reached 0.67, which was significantly higher than that of three models published previously (P

We investigated the relevance of the validation principles on the Quantitative Structure Activity Relationship models issued by Organization for Economic and Co‐operation and Development. We checked the goodness‐of‐fit, robustness and predictivity categories in linear and nonlinear models using benchmark datasets. Most of our conclusions are drawn using the sample size dependence of the different validation parameters. We found that the goodness‐of‐fit parameters misleadingly overestimate the models on small samples. In the case of neural network and support vector models, the feasibility of the goodness‐of‐fit parameters often might be questioned. We propose to use the simplest y‐scrambling method to estimate chance correlation. We found that the leave‐one‐out and leave‐many‐out cross‐validation parameters can be rescaled to each other in all models and the computationally feasible method should be chosen depending on the model type. We assessed the interdependence of the validation parameters by calculating their rank correlations. Goodness of fit and robustness correlate quite well over a sample size for linear models and one of the approaches might be redundant. In the rank correlation between internal and external validation parameters, we found that the assignment of good and bad modellable data to the training or the test causes negative correlations. [ABSTRACT FROM AUTHOR]

Background: Accurate and interpretable models are essential for clinical decision-making, where predictions can directly impact patient care. Machine learning (ML) survival methods can handle complex multidimensional data and achieve high accuracy but require post-hoc explanations. Traditional models such as the Cox Proportional Hazards Model (Cox-PH) are less flexible, but fast, stable, and intrinsically transparent. Moreover, ML does not always outperform Cox-PH in clinical settings, warranting a diligent model validation. We aimed to develop a set of R functions to help explore the limits of Cox-PH compared to the tree-based and deep learning survival models for clinical prediction modelling, employing ensemble learning and nested cross-validation., Methods: We developed a set of R functions, publicly available as the package "survcompare". It supports Cox-PH and Cox-Lasso, and Survival Random Forest (SRF) and DeepHit are the ML alternatives, along with the ensemble methods integrating Cox-PH with SRF or DeepHit designed to isolate the marginal value of ML. The package performs a repeated nested cross-validation and tests for statistical significance of the ML's superiority using the survival-specific performance metrics, the concordance index, time-dependent AUC-ROC and calibration slope. To get practical insights, we applied this methodology to clinical and simulated datasets with varying complexities and sizes., Results: In simulated data with non-linearities or interactions, ML models outperformed Cox-PH at sample sizes ≥ 500. ML superiority was also observed in imaging and high-dimensional clinical data. However, for tabular clinical data, the performance gains of ML were minimal; in some cases, regularised Cox-Lasso recovered much of the ML's performance advantage with significantly faster computations. Ensemble methods combining Cox-PH and ML predictions were instrumental in quantifying Cox-PH's limits and improving ML calibration. Traditional models like Cox-PH or Cox-Lasso should not be overlooked while developing clinical predictive models from tabular data or data of limited size., Conclusion: Our package offers researchers a framework and practical tool for evaluating the accuracy-interpretability trade-off, helping make informed decisions about model selection., Competing Interests: Declaration of competing interest The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Diana Shamsutdinova reports financial support was provided by National Institute for Health Research Maudsley Biomedical Research Centre. Daniel Stahl reports financial support was provided by National Institute for Health Research Maudsley Biomedical Research Centre. Daniel Stamate reports financial support was provided by Alzheimer’s Research UK. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 The Author(s). Published by Elsevier B.V. All rights reserved.)

We performed an internal laboratory validation of the Precision ID GlobalFiler NGS STR panel v2 kit to assist the introduction of the technology into the routine forensic casework practice. The study was designed and evaluated based not only on the key validation standards like sensitivity, stability, reproducibility, repeatability, mixture, and concordance, but we also tested the effect of reduced input DNA, we measured and applied locus-specific analytical threshold values, tested two different PCR cycle conditions, sequence artifacts and stutters were also analysed. During the study we also tested the new method on real casework samples. The sensitivity study confirmed that adding 500 pg template DNA for library preparation can be optimal at base PCR cycle number (that was 24), because the measured average heterozygote balance was not lower than 0.82, and each allele was detected above the analytical threshold. However, contrary to previous communications, increasing the PCR cycle numbers up to 28 has not resulted the significant elevation of the heterozygote imbalance. According to our results, raised PCR cycle condition (i.e. 28) is appropriate at or below 150 pg total input DNA. For most loci, the calculated AT was lower than the manufacturer's recommended. Applying the newly established ATs with raised PCR cycle conditions the allele detection sensitivity and reliability increased. We observed allele dropouts only at the 15 pg template DNA experiments with 5 % frequency, that is better to previously published studies. This result indicates that this low amount of DNA (i.e. 15 pg) could be a minimum limit of template input for a potentially successful analysis. In the mixture study the minor contributor could be detected up to 1:19 mixture ratio. We detected minor alleles in all measurements and concentrations above the threshold if the template DNA were fixed and only SNP differences were observed between the same alleles of the contributors. To test concordance between the new method and traditional STR genotyping we analysed 58 Hungarian individual samples in parallel. Nearby the detected 248 different length-based alleles on the 31 loci in the sample pool we revealed additional 75 sequence variant alleles, that represent an approximately 23 % increase in the total number of observed alleles. The casework study confirmed that the Precision ID GlobalFiler NGS STR panel v2 kit is effective even in genotyping degraded samples with extremely low levels of DNA, if we apply elevated cycle number for library preparation and use locus-specific analytical thresholds., Competing Interests: Declaration of Competing Interest The authors declare that they have no competing interests., (Copyright © 2024 The Authors. Published by Elsevier B.V. All rights reserved.)

Aim: Clinical prediction models need to be validated. In this study, we used simulation data to compare various internal and external validation approaches to validate models. Methods: Data of 500 patients were simulated using distributions of metabolic tumor volume, standardized uptake value, the maximal distance between the largest lesion and another lesion, WHO performance status and age of 296 diffuse large B cell lymphoma patients. These data were used to predict progression after 2 years based on an existing logistic regression model. Using the simulated data, we applied cross-validation, bootstrapping and holdout (n = 100). We simulated new external datasets (n = 100, n = 200, n = 500) and simulated stage-specific external datasets (1), varied the cut-off for high-risk patients (2) and the false positive and false negative rates (3) and simulated a dataset with EARL2 characteristics (4). All internal and external simulations were repeated 100 times. Model performance was expressed as the cross-validated area under the curve (CV-AUC ± SD) and calibration slope. Results: The cross-validation (0.71 ± 0.06) and holdout (0.70 ± 0.07) resulted in comparable model performances, but the model had a higher uncertainty using a holdout set. Bootstrapping resulted in a CV-AUC of 0.67 ± 0.02. The calibration slope was comparable for these internal validation approaches. Increasing the size of the test set resulted in more precise CV-AUC estimates and smaller SD for the calibration slope. For test datasets with different stages, the CV-AUC increased as Ann Arbor stages increased. As expected, changing the cut-off for high risk and false positive- and negative rates influenced the model performance, which is clearly shown by the low calibration slope. The EARL2 dataset resulted in similar model performance and precision, but calibration slope indicated overfitting. Conclusion: In case of small datasets, it is not advisable to use a holdout or a very small external dataset with similar characteristics. A single small testing dataset suffers from a large uncertainty. Therefore, repeated CV using the full training dataset is preferred instead. Our simulations also demonstrated that it is important to consider the impact of differences in patient population between training and test data, which may ask for adjustment or stratification of relevant variables. [ABSTRACT FROM AUTHOR]