SCEPTRE improves calibration and sensitivity in single-cell CRISPR screen analysis

Single-cell CRISPR screens are a promising biotechnology for mapping regulatory elements to target genes at genome-wide scale. However, technical factors like sequencing depth impact not only expression measurement but also perturbation detection, creating a confounding effect. We demonstrate on two single-cell CRISPR screens how these challenges cause calibration issues. We propose SCEPTRE: analysis of single-cell perturbation screens via conditional resampling, which infers associations between perturbations and expression by resampling the former according to a working model for perturbation detection probability in each cell. SCEPTRE demonstrates very good calibration and sensitivity on CRISPR screen data, yielding hundreds of new regulatory relationships supported by orthogonal biological evidence.

Cardiovascular Events in People With Type 2 Diabetes: Performance of Framingham, UKPDS, and ADVANCE Risk Equations

The aim of this study was to assess the performance of the Framingham, UK Prospective Diabetes Study (UKPDS), and the Action in Diabetes and Vascular disease: Preterax and Diamicron-MR Controlled Evaluation (ADVANCE) risk equations in the prediction of 4-year cardiovascular disease CVD) in Iranian people with type 2 diabetes. The 4-year risks of CVD were estimated using the three equations in a community of 557 patients with type 2 diabetes and free of CVD at baseline. A trained physician evaluated all of the participants regarding the occurrence of CVD events during follow-up. CVD was defined as major events including fatal/non-fatal myocardial infarction as well as fatal/non-fatal stroke, minor events including treated coronary heart disease (CHD), and established peripheral arterial disease (PAD). During four years of follow-up, 64 CVD events were observed (66% minor CVD events). Despite having a good calibration (estimated to observed ratio ranging from 91.37 to 98.2 percent, Hosmer–Lemeshow χ2 (HLχ2) values <15), both general (Framingham) and diabetes-specific (UKPDS and ADVANCE) equations did not have adequate discriminative ability (Area Under the Curve (AUC) ranging from 0.48 to 0.56). Framingham, UKPDS, and ADVANCE risk equations, regardless of being general or diabetes-specific, could not precisely predict 4-year risk of CVD in Iranian individuals with type 2 diabetes.

Machine-Learning Based Early Warning System for Prediction for Disseminated Intravascular Coagulation after Allogeneic Hematopoietic Stem Cell Transplantation: A Nationwide Multicenter Study

Introduction Allogeneic haematopoietic stem cell transplantation (allo-HSCT) has been demonstrated to be the most effective therapy for various malignant as well as nonmalignant haematological diseases. The wide use of allo-HSCT has inevitably led to a variety of complications after transplantation, with bleeding complications such as disseminated intravascular coagulation (DIC). DIC accounts for a significant proportion of life-threatening bleeding cases occurring after allo-HSCT. However, information on markers for early identification remains limited, and no predictive tools for DIC after allo-HSCT are available. This research aimed to identify the risk factors for DIC after allo-HSCT and establish prediction models to predict the occurrence of DIC after allo-HSCT. Methods The definition of DIC was based on the International Society of Thrombosis and Hemostasis (ISTH) scoring system. Overall, 197 patients with DIC after allo-HSCT at Peking University People's Hospital and other 7 centers in China from January 2010 to June 2021 were retrospectively identified. Each patient was randomly matched to 3 controls based on the time of allo-HSCT (±3 months) and length of follow-up (±6 months). A lasso regression model was used for data dimension reduction, feature selection, and risk factor building. Multivariable logistic regression analysis was used to develop the prediction model. We incorporated the clinical risk factors, and this was presented with a nomogram. The performance of the nomogram was assessed with respect to its calibration, discrimination, and clinical usefulness. Internal and external validation was assessed. Various machine learning models were further used to perform machine learning modeling by attempting to complete the data sample classification task, including XGBClassifier, LogisticRegression, MLPClassifier, RandomForestClassifier, and AdaBoostClassifier. Results A total of 7280 patients received allo-HSCT from January 2010 to June 2021, and DIC occurred in 197 of these patients (incidence of 2.7%). The derivation cohort included 120 DIC patients received allo-HSCT and 360 patients received allo-HSCT from Peking University People's Hospital, and the validation cohort included the remaining 77 patients received allo-HSCT and 231 patients received allo-HSCT from the other 7 centers. The median time for DIC events was 99.0 (IQR, 46.8-220) days after allo-HSCT. The overall survival of patients with DIC was significantly reduced (P ＜ 0.0001). By Lasso regression, the 10 variables with the highest importance were found to be prothrombin time activity (PTA), shock, C-reactive protein, internationalization normalized ratio, bacterial infection, oxygenation, fibrinogen, blood creatinine, white blood cell count, and acute respiratory distress syndrome (from highest to lowest). In the multivariate analysis, the independent risk factors for DIC included PTA, bacterial infection and shock (P &lt;0.001), and these predictors were included in the clinical prediction nomogram. The model showed good discrimination, with a C-index of 0.975 (95%CI, 0.939 to 0.987 through internal validation) and good calibration. Application of the nomogram in the validation cohort still gave good discrimination (C-index, 0.778 [95% CI, 0.759 to 0.766]) and good calibration. Decision curve analysis demonstrated that the nomogram was clinically useful. The predictive value ROC curves of different machine learning models show that XGBClassifier is the best performing model for this dataset, with an area under the curve of 0.86. Conclusions Risk factors for DIC after allo-HSCT were identified, and a nomogram model and various machine learning models were established to predict the occurrence of DIC after allo-HSCT. Combined, these can help recognize high-risk patients and provide timely treatment. In the future, we will further refine the prognostic model utilizing nationwide multicenter data and conduct prospective clinical trials to reduce the incidence of DIC after allo-HSCT and improve the prognosis. Disclosures No relevant conflicts of interest to declare.

Predicting the nature of pleural effusion in patients with lung adenocarcinoma based on 18F-FDG PET/CT

Background This study aims to establish a predictive model on the basis of 18F-FDG PET/CT for diagnosing the nature of pleural effusion (PE) in patients with lung adenocarcinoma. Methods Lung adenocarcinoma patients with PE who underwent 18F-FDG PET/CT were collected and divided into training and test cohorts. PET/CT parameters and clinical information in the training cohort were collected to estimate the independent predictive factors of malignant pleural effusion (MPE) and to establish a predictive model. This model was then applied to the test cohort to evaluate the diagnostic efficacy. Results A total of 413 lung adenocarcinoma patients with PE were enrolled in this study, including 245 patients with MPE and 168 patients with benign PE (BPE). The patients were divided into training (289 patients) and test (124 patients) cohorts. CEA, SUVmax of tumor and attachment to the pleura, obstructive atelectasis or pneumonia, SUVmax of pleura, and SUVmax of PE were identified as independent significant factors of MPE and were used to construct a predictive model, which was graphically represented as a nomogram. This predictive model showed good discrimination with the area under the curve (AUC) of 0.970 (95% CI 0.954–0.986) and good calibration. Application of the nomogram in the test cohort still gave good discrimination with AUC of 0.979 (95% CI 0.961–0.998) and good calibration. Decision curve analysis demonstrated that this nomogram was clinically useful. Conclusions Our predictive model based on 18F-FDG PET/CT showed good diagnostic performance for PE, which was helpful to differentiate MPE from BPE in patients with lung adenocarcinoma.

Nomogram Combining Radiomics With the American College of Radiology Thyroid Imaging Reporting and Data System Can Improve Predictive Performance for Malignant Thyroid Nodules

PurposeTo develop and validate a nomogram combining radiomics of B-mode ultrasound (BMUS) images and the American College of Radiology (ACR) Thyroid Imaging Reporting and Data System (TI-RADS) for predicting malignant thyroid nodules and improving the performance of the guideline.MethodA total of 451 thyroid nodules referred for surgery and proven pathologically at an academic referral center from January 2019 to September 2020 were retrospectively collected and randomly assigned to training and validation cohorts (7:3 ratio). A nomogram was developed through combining the BMUS radiomics score (Rad-Score) with ACR TI-RADS score (ACR-Score) in the training cohort; the performance of the nomogram was assessed with respect to discrimination, calibration, and clinical application in the validation and entire cohorts.ResultsThe ACR-Rad nomogram showed good calibration and yielded an AUC of 0.877 (95% CI 0.836–0.919) in the training cohort and 0.864 (95% CI 0.799–0.931) in the validation cohort, which were significantly better than the ACR-Score model (p &lt; 0.001 and 0.031, respectively). The significantly improved AUC, net reclassification index (NRI), and integrated discriminatory improvement (IDI) of the nomogram were found for both senior and junior radiologists (all p &lt; 0.001). Decision curve analysis indicated that the nomogram was clinically useful. When cutoff values for 50% predicted malignancy risk (ACR-Rad_50%) were applied, the nomogram showed increased specificity, accuracy and positive predictive value (PPV), and decreased unnecessary fine-needle aspiration (FNA) rates in comparison to ACR TI-RADS.ConclusionThe ACR-Rad nomogram has favorable value in predicting malignant thyroid nodules and improving performance of the ACR TI-RADS for senior and junior radiologists.

Ischemic Stroke in Patients With Hypertrophic Cardiomyopathy According to Presence or Absence of Atrial Fibrillation

Background and Purpose: Patients with hypertrophic cardiomyopathy (HCM) have high risk of ischemic stroke (IS), especially if atrial fibrillation (AF) is present. Improvements in risk stratification are needed to help identify those patients with HCM at higher risk of stroke, whether AF is present or not. Methods: This French longitudinal cohort study from the database covering hospital care from 2010 to 2019 analyzed adults hospitalized with isolated HCM. A logistic regression model was used to construct a French HCM score, which was compared with the HCM Risk-CVA and CHA 2 DS 2 -VASc scores using c-indexes and calibration analysis. Results: In 32 206 patients with isolated HCM, 12 498 (38.8%) had AF, and 2489 (7.7%) sustained an IS during follow-up. AF in patients with HCM was independently associated with a higher risk for death (hazard ratio, 1.129 [95% CI, 1.088–1.172]), cardiovascular death (hazard ratio, 1.254 [95% CI, 1.177–1.337]), IS (hazard ratio, 1.210 [95% CI, 1.111–1.317]), and other major cardiovascular events. Independent predictors of IS in HCM were older age, heart failure, AF, prior IS, smoking and poor nutrition (all P <0.05). For the HCM Risk-CVA score, CHA 2 DS 2 -VASc score and a French HCM score, all c-indexes were 0.65 to 0.70, with good calibration. Among patients with AF, the CHA 2 DS 2 -VASc score had marginal improvement over the HCM Risk-CVA score but was less predictive compared with the French HCM score ( P =0.001). In patients without AF, both HCM Risk-CVA score and the French HCM score had significantly better prediction compared with CHA 2 DS 2 -VASc (both P <0.0001). Decision curve analysis demonstrated that the French HCM score had the best clinical usefulness of the 3 tested risk scores. Conclusions: Patients with HCM have a high prevalence of AF and a significant risk of IS, and the presence of AF in patients with HCM was independently associated with worse outcomes. A simple French HCM score shows good prediction of IS in patients with HCM and clinical usefulness, with good calibration.

Recalibration Methods for Improved Clinical Utility of Risk Scores

Background An established risk model may demonstrate miscalibration, meaning predicted risks do not accurately capture event rates. In some instances, investigators can identify and address the cause of miscalibration. In other circumstances, it may be appropriate to recalibrate the risk model. Existing recalibration methods do not account for settings in which the risk score will be used for risk-based clinical decision making. Methods We propose 2 new methods for risk model recalibration when the intended purpose of the risk model is to prescribe an intervention to high-risk individuals. Our measure of risk model clinical utility is standardized net benefit. The first method is a weighted strategy that prioritizes good calibration at or around the critical risk threshold. The second method uses constrained optimization to produce a recalibrated risk model with maximum possible net benefit, thereby prioritizing good calibration around the critical risk threshold. We also propose a graphical tool for assessing the potential for recalibration to improve the net benefit of a risk model. We illustrate these methods by recalibrating the American College of Cardiology (ACC)–American Heart Association (AHA) atherosclerotic cardiovascular disease (ASCVD) risk score within the Multi-Ethnic Study of Atherosclerosis (MESA) cohort. Results New methods are implemented in the R package ClinicalUtilityRecal. Recalibrating the ACC-AHA-ASCVD risk score for a MESA subcohort results in higher estimated net benefit using the proposed methods compared with existing methods, with improved calibration in the most clinically impactful regions of risk. Conclusion The proposed methods target good calibration for critical risks and can improve the net benefit of a risk model. We recommend constrained optimization when the risk model net benefit is paramount. The weighted approach can be considered when good calibration over an interval of risks is important.

CT-based Radiomic Nomogram for Predicting the Severity of Patients With COVID-19

Background: The coronavirus disease 2019 (COVID-19) is a pandemic now, and the severe COVID-19 determines the management and treatment, even prognosis. We aim to develop and validate a radiomics nomogram for identifying severe patients with COVID-19. To develop and validate a radiomics nomogram for identifying severe patients with COVID-19.Methods: There were 156 and 104 patients with COVID-19 enrolled in primary and validation cohorts respectively. Radiomics features were extracted from chest CT images. Least absolute shrinkage and selection operator (LASSO) method was used for feature selection and radiomics signature building. Multivariable logistic regression analysis was used to develop a predictive model, and the radiomics signature, abnormal WBC counts, and comorbidity were incorporated and presented as a radiomics nomogram. The performance of the nomogram was assessed through its calibration, discrimination, and clinical usefulness.Results: The radiomics signature consisting of 4 selected features was significantly associated with clinical condition of patients with COVID-19 in the primary and validation cohorts (P<0.001). The radiomics nomogram including radiomics signature, comorbidity and abnormal WBC counts, showed good discrimination of severe COVID-19, with an AUC of 0.972, and good calibration in the primary cohort. Application of the nomogram in the validation cohort still gave good discrimination with an AUC of 0.978 and good calibration. Decision curve analysis demonstrated that the radiomics nomogram was clinically useful to identify the severe COVID-19.Conclusion: We present an easy-to-use radiomics nomogram to identify the severe patients with COVID-19 for better guiding a prompt management and treatment.

Construction of a survival prediction model for high-and low -grade UTUC after tumor resection based on “SEER database”: a multicenter study

Background There are differences in survival between high-and low-grade Upper Tract Urothelial Carcinoma (UTUC). Our study aimed to develop a nomogram to predict overall survival (OS) of patients with high- and low-grade UTUC after tumor resection, and to explore the difference between high- and low-grade patients. Methods Patients confirmed to have UTUC between 2004 and 2015 were selected from the Surveillance, Epidemiology and End Results (SEER) database. The UTUCs were identified and classified as high- and low-grade, and 1-, 3- and 5-year nomograms were established. The nomogram was then validated using the Chinese multicenter dataset (patients diagnosed in Shandong, China between January 2010 and October 2020). Results In the high-grade UTUC patients, nine important factors related to survival after tumor resection were identified to construct nomogram. The C index of training dataset was 0.740 (95% confidence interval [CI]: 0.727–0.754), showing good calibration. The C index of internal validation dataset was 0.729(95% CI:0.707–0.750). On the other hand, Two independent predictors were identified to construct nomogram of low-grade UTUC. The C index was 0.714 (95% CI: 0.671–0.758) for the training set,0.731(95% CI:0.670–0.791) for the internal validation dataset. Encouragingly, the nomogram was clinically useful and had a good discriminative ability to identify patients at high risk. Conclusion We constructed a nomogram and a corresponding risk classification system predicting the OS of patients with an initial diagnosis of high-and low-grade UTUC.

Nomogram for Predicting Bone Development State of Female Children and Adolescents–A Fast Screening Approach Based on Pubes Stages for Growth and Development

Objective: To develop a nomogram for predicting bone development state (BDS) of female children and adolescents in a large scale.Methods: Four hundred forty-seven female students were designated as the training cohort to develop the predictive model, whereas 196 female students were used as the validation cohort to verify the established model. Bone age, height, body mass, body fat percentage, and secondary sexual characteristics were recorded, and BDS was determined with the chronological age and bone age. Multivariate logistic regression was conducted to determine the factors, and nomogram was developed and validated with the training and validation cohorts, respectively.Results: One hundred forty-seven female students were identified as BDS abnormal in the training cohort (32.9%), and 104 were determined in the validation cohort (53.1%). Age, height, weight, and pubes stage were selected for the predictive model. A nomogram was developed and showed a good estimation, with a C-index of 0.78 and a good calibration in the training cohort. Application of the nomogram to the validation cohort showed a similar C-index of 0.75 and a good calibration.Conclusion: A nomogram for predicting bone development was developed, which can provide a relatively good estimation of BDS for female children and adolescents in Chinese metropolis.