Net Reclassification Index Research Articles

Assessing the transportability of radiomic models for lung cancer diagnosis: commercial vs. open-source feature extractors.

Radiomics has shown promise in improving malignancy risk stratification of indeterminate pulmonary nodules (IPNs) with many platforms available, but with no head-to-head comparisons. This study aimed to evaluate transportability of radiomic models across platforms by comparing performances of a commercial radiomic feature extractor (HealthMyne) with an open-source extractor (PyRadiomics) on diagnosis of lung cancer in IPNs. A commercial radiomic feature extractor was used to segment IPNs from computed tomography (CT) scans, and a previously validated radiomic model based on commercial features was used as baseline (ComRad). Using same segmentation masks, PyRadiomics, an open-source feature extractor was used to build three open-source radiomic models (OpenRad) using different methods: de novo open-source model derived using least absolute shrinkage and selection operator (LASSO) for feature selection, selecting open-source features matched to ComRad features based upon Imaging Biomarker Standardization Initiative (IBSI) nomenclature, and selecting open-source features most highly correlated to ComRad features. Radiomic models were trained on an internal cohort (n=161) and externally validated on 3 cohorts (n=278). We added Mayo clinical risk score to OpenRad and ComRad models, creating integrated clinical radiomic (ClinRad) models. All models were compared using area under the curve (AUC) and evaluated for clinical improvement using bias-corrected clinical net reclassification indices (cNRIs). ComRad AUC was 0.76 [95% confidence interval (CI): 0.71-0.82], and OpenRad AUC was 0.75 (95% CI: 0.69-0.81) for LASSO model, 0.74 (95% CI: 0.68-0.79) for Spearman's correlation, and 0.71 (95% CI: 0.65-0.77) for IBSI. Mayo scores were added to OpenRad LASSO model, which performed best, forming open-source ClinRad model with AUC of 0.80 (95% CI: 0.74-0.86), identical to commercial ClinRad's AUC. Both ClinRad models showed clinical improvement compared to Mayo alone, with commercial ClinRad achieving cNRI of 0.09 (95% CI: 0.02-0.15) for benign and 0.07 (95% CI: 0.00-0.13) for malignant, and open-source ClinRad achieving cNRI of 0.09 (95% CI: 0.02-0.15) for benign and 0.06 (95% CI: 0.00-0.12) for malignant. Transportability of radiomic models across platforms directly does not conserve performance, but radiomic platforms can provide equivalent results when building de novo models allowing for flexibility in feature selection to maximize prediction accuracy.

Predictive value of NT pro BNP for new-onset atrial fibrillation in heart failure and preserved ejection fraction.

The prognostic significance of N-terminal pro B-type natriuretic peptide (NT-proBNP) in heart failure with preserved ejection fraction (HFpEF) has been well established. HFpEF and atrial fibrillation (AF) commonly coexist, and each contributes to poor outcomes independently. Nevertheless, the ability of NT-proBNP to predict AF in HFpEF patients remains uncertain. A total of 367 HFpEF patients without baseline AF from the Treatment of Preserved Cardiac Function Heart Failure with an Aldosterone Antagonist (TOPCAT) trial were included. The Cox proportional hazard model was used to assess the association of NT-proBNP with the risk of AF. The C-statistic, categorical net reclassification index (NRI), and integrated discrimination improvement (IDI) were used to evaluate the ability of NT-proBNP in new-onset AF prediction. During a median follow-up of 2.91years, 17 (4.63%) new-onset AF cases occurred. Every 1000pg/mL increase in NT-proBNP was associated with a 16% increase in the risk of AF occurrence after adjustments (hazard ratio, 1.16 [95% CI, 1.02-1.32]). NT-proBNP showed a moderate performance for new-onset AF at 3years (C-statistic, 0.67). Adding NT-proBNP to CHADS2/R2CHADS2/CHA2DS2-VASc/C2HSET scores improved their predictive performance for AF risk (CHADS2: C-statistic, 0.63, CHADS2+NT: C-statistic, 0.69, NRI, 47.46%, IDI, 1.18%; R2CHADS2: C-statistic, 0.65, R2CHADS2+NT: C-statistic, 0.70, NRI, 48.03%, IDI, 0.51%; CHA2DS2-VASc: C-statistic, 0.67, CHA2DS2-VASc+NT: C-statistic, 0.72, NRI, 49.41%, IDI, 0.86%; C2HSET: C-statistic, 0.77, C2HSET+NT: C-statistic, 0.80, NRI, 50.32%, IDI, 1.58%). Among patients with HFpEF, the NT-proBNP level was positively associated with the incidence of new-onset AF and may be a promising predictor.

Association between the pan-immune-inflammation value and coronary collateral circulation in chronic total coronary occlusive patients

BackgroundInflammation and immunity play important roles in the formation of coronary collateral circulation (CCC). The pan-immune-inflammation value (PIV) is a novel marker for evaluating systemic inflammation and immunity. The study aimed to investigate the association between the PIV and CCC formation in patients with chronic total occlusion (CTO).MethodsThis retrospective study enrolled 1150 patients who were diagnosed with CTO through coronary angiographic (CAG) examinations from January 2013 to December 2021 in China. The Cohen-Rentrop criteria were used to catagorize CCC formation: good CCC formation (Rentrop grade 2–3) and poor CCC formation group (Rentrop grade 0–1). Based on the tertiles of the PIV, all patients were classified into three groups as follows: P1 group, PIV ≤ 237.56; P2 group, 237.56< PIV ≤ 575.18; and P3 group, PIV > 575.18.ResultsA significant relationship between the PIV and the formation of CCC was observed in our study. Utilizing multivariate logistic regression and adjusting for confounding factors, the PIV emerged as an independent risk factor for poor CCC formation. Notably, the restricted cubic splines revealed a dose–response relationship between the PIV and risk of poor CCC formation. In terms of predictive accuracy, the area under the ROC curve (AUC) for PIV in anticipating poor CCC formation was 0.618 (95% CI: 0.584–0.651, P < 0.001). Furthermore, the net reclassification index (NRI) and integrated discrimination index (IDI) for PIV, concerning the prediction of poor CCC formation, were found to be 0.272 (95% CI: 0.142–0.352, P < 0.001) and 0.051 (95% CI: 0.037–0.065, P < 0.001), respectively. It’s noteworthy that both the NRI and IDI values were higher for PIV compared to other inflammatory biomarkers, suggesting its superiority in predictive capacity.ConclusionsPIV was associated with the formation of CCC. Notably, PIV exhibited potential as a predictor for poor CCC formation and showcased superior predictive performance compared to other complete blood count-based inflammatory biomarkers.

MicroRNA Profiling as a Predictive Indicator for Time to First Treatment in Chronic Lymphocytic Leukemia: Insights from the O-CLL1 Prospective Study.

A "watch and wait" strategy, delaying treatment until active disease manifests, is adopted for most CLL cases; however, prognostic models incorporating biomarkers have shown to be useful to predict treatment requirement. In our prospective O-CLL1 study including 224 patients, we investigated the predictive role of 513 microRNAs (miRNAs) on time to first treatment (TTFT). In the context of this study, six well-established variables (i.e., Rai stage, beta-2-microglobulin levels, IGVH mutational status, del11q, del17p, and NOTCH1 mutations) maintained significant associations with TTFT in a basic multivariable model, collectively yielding a Harrell's C-index of 75% and explaining 45.4% of the variance in the prediction of TTFT. Concerning miRNAs, 73 out of 513 were significantly associated with TTFT in a univariable model; of these, 16 retained an independent relationship with the outcome in a multivariable analysis. For 8 of these (i.e., miR-582-3p, miR-33a-3p, miR-516a-5p, miR-99a-5p, and miR-296-3p, miR-502-5p, miR-625-5p, and miR-29c-3p), a lower expression correlated with a shorter TTFT, whereas in the remaining eight (i.e., miR-150-5p, miR-148a-3p, miR-28-5p, miR-144-5p, miR-671-5p, miR-1-3p, miR-193a-3p, and miR-124-3p), the higher expression was associated with shorter TTFT. Integrating these miRNAs into the basic model significantly enhanced predictive accuracy, raising the Harrell's C-index to 81.1% and the explained variation in TTFT to 63.3%. Moreover, the inclusion of the miRNA scores enhanced the integrated discrimination improvement (IDI) and the net reclassification index (NRI), underscoring the potential of miRNAs to refine CLL prognostic models and providing insights for clinical decision-making. In silico analyses on the differently expressed miRNAs revealed their potential regulatory functions of several pathways, including those involved in the therapeutic responses. To add a biological context to the clinical evidence, an miRNA-mRNA correlation analysis revealed at least one significant negative correlation between 15 of the identified miRNAs and a set of 50 artificial intelligence (AI)-selected genes, previously identified by us as relevant for TTFT prediction in the same cohort of CLL patients. In conclusion, the identification of specific miRNAs as predictors of TTFT holds promise for enhancing risk stratification in CLL to predict therapeutic needs. However, further validation studies and in-depth functional analyses are required to confirm the robustness of these observations and to facilitate their translation into meaningful clinical utility.

Impact of Operator Experience on Left Atrial Appendage Occlusion Outcomes.

The relationship between long-term outcomes and operator experience for left atrial appendage occlusion (LAAO) is still unknown. This study sought to explore the association between operator LAAO experience and one-year clinical outcomes. The RECORD study (Registry to Evaluate Chinese Real-World Clinical Outcomes in Patients With AF Using the WATCHMAN Left Atrial Appendage Closure Technology; NCT03917563) was a multicenter, prospective registry that included patients with the WATCHMAN LAAO device (Boston Scientific) in China from April 1, 2019, to October 31, 2020. The current analyses included patients with solely LAAO from the registry; those who had concomitant LAAO and ablation/other procedures were excluded. The primary outcome was a composite endpoint of death, stroke, systemic embolism, and Bleeding Academic Research Consortium (BARC)-defined type 3 or 5 bleeding at 1 year. A total of 1,547 LAAO patients and 111 operators were included. The mean ± SD CHA2DS2-VASc and HAS-BLED scores of patients were 4.0 ± 1.8 and 2.5 ± 1.1, respectively. The mean ± SD age of operators was 47.0 ± 7.2 years, 15(13.5%) were female, and 52 (46.8%) were electrophysiologists. Utilizing maximally selected log-rank statistics, the thresholds to categorize an experienced operator were performing≥32 LAAOs annually or≥134 LAAOs in total. Performing≥32 LAAOs annually is the better criterion than≥134 LAAOs in total (absolute net reclassification index: 25.79%; P< 0.001). Compared with the≥32 LAAO annually group, the<32 group was associated with a 1.8-fold (HRadjusted: 1.79; 95%CI: 1.16-2.78; P=0.009) increase in the risk of the primary endpoint, and such risk in the<32 group can be reduced by ∼12% after performing each additional 5 cases (HRadjusted per 5 cases: 0.88; 95%CI: 0.78-0.99; P=0.033). Performing≥32 LAAOs annually could be a threshold to categorize an experienced operator. Before reaching this threshold, the risk of death, stroke, systemic embolism, and BARC-defined type 3 or 5 bleeding decreased by 12% after every 5 cases performed.

Hypertension phenotypes and adverse pregnancy outcome-related office and ambulatory blood pressure thresholds during pregnancy: a retrospective cohort study.

Blood pressure (BP) phenotypes, as determined by the consistency between office BP (OBP) and ambulatory BP (ABP) measurements, enhance risk assessment during pregnancy. However, diagnostic criteria for hypertension in pregnancy are based on data from non-pregnant populations regarding long-term cardiovascular risks. This study aimed to identify adverse pregnancy outcomes (APOs; including maternal/fetal outcomes)-related BP thresholds to refine risk assessment in pregnant women. We analyzed 967 high-risk pregnant women who underwent simultaneous OBP and ABP measurements at an average gestational age of 29.6 ± 8.0 weeks. All hypertension phenotypes were associated with an increased risk of maternal and fetal outcomes, except white coat hypertension, which showed no association with fetal outcomes. Using an XGBoost algorithm, the receiver operating characteristic (ROC) curve-derived daytime diastolic BP (DBP) thresholds of 81.5 mmHg for maternal and 82.5 mmHg for fetal outcomes were identified as the BP parameters most strongly linked to APOs. Incorporating these thresholds into the BP phenotype-based model improved the area under the curve for APOs and the net reclassification index for maternal and fetal outcomes. Decision curve analysis demonstrated a consistent positive net benefit after incorporating BP thresholds into the phenotype-based model for maternal and composite outcomes. In conclusion, in a Chinese pregnancy cohort, we identified daytime DBP as the most influential parameter for APOs, significantly enhancing the predictive performance of BP phenotype-based models. This study underscores the importance of ABP monitoring in high-risk pregnancies and the need for further research to establish optimal BP monitoring criteria for pregnancy.

Predictive value of neutrophil-to-lymphocyte ratio on admission for intrapartum maternal fever in parturients undergoing epidural analgesia: A retrospective cohort study using propensity score-matched analysis.

To identify the predictive value of the neutrophil-to-lymphocyte ratio (NLR) on admission for intrapartum maternal fever in parturients undergoing epidural analgesia (EA). In this retrospective cohort study, propensity score matching (PSM) was applied to address covariates. Univariate and multivariate regression analyses were implemented in sequence to find out the factors influencing intrapartum fever. The receiver operating characteristics curve was applied to determine the area under the curve (AUC) of NLR for intrapartum fever. NLR and duration of EA were independent risk factors for intrapartum fever. The AUC of the combined indicator (NLR + duration of EA) was higher than that of NLR (AUC = 0.583, 95% confidence interval [CI] 0.53-0.64) and duration of EA (AUC = 0.702, 95% CI 0.66-0.75), reaching 0.715 (95% CI 0.67-0.76; p < 0.001). NLR increased predictive performance for intrapartum fever when added to the duration of EA (net reclassification index 0.076, p = 0.022; integrated discrimination improvement 0.020, p = 0.002). NLR has limited predictive power for intrapartum fever. The combination of NLR and duration of epidural analgesia may be considered a promising predictor for intrapartum maternal fever in parturients undergoing epidural analgesia. The neutrophil-to-lymphocyte ratio is an accessible predictor for the early identification of parturients at risk of intrapartum fever.

Dynamic Changes of Neutrophil-to-Lymphocyte Ratio on Predicting Response of Immune Checkpoint Inhibitors Plus Targeted Therapies for Unresectable Hepatocellular Carcinoma.

Multiple regimens of immune checkpoint inhibitors (ICIs) plus targeted therapies are commonly prescribed as first-line treatments for unresectable hepatocellular carcinoma (uHCC). Here, we aimed to investigate the correlation between dynamic changes of neutrophil-to-lymphocyte ratio (NLR) and tumor response to the combination of ICIs and targeted therapies for uHCC. Sixty-one patients who received ICIs plus targeted therapies for uHCC were enrolled in this retrospective study. The NLR before and at 3-6 weeks after treatments were assessed to calculate the dynamic NLR changes (ΔNLR). Multivariate logistic regression and Cox regression models were used to explore the relationship between dynamic NLR changes and tumor response or progression-free survival (PFS), respectively. Furthermore, we assessed the predictive effect of alpha-fetoprotein (AFP) changes in combination with dynamic NLR changes compared to AFP changes alone. The NLR at 3-6 weeks and ΔNLR after treatments significantly increased in patients who underwent progressive disease (PD), while the baseline NLR showed no significant difference between different tumor responses. Increased NLR and AFP after treatments were both independent predictors of PD (For NLR increase: OR, 2.28; 95% CI, 1.47-3.88, P < 0.001; For AFP increase: OR, 1.46; 95% CI, 1.03-2.17, P = 0.043), and correlated with worse PFS (for NLR increase: HR, 4.08; 95% CI, 1.99-8.36, P < 0.001; for AFP increase: HR, 2.10; 95% CI, 1.04-4.24, P = 0.039). The receiver operating characteristic (ROC) curve and net reclassification index (NRI) showed that the combination of dynamic NLR and AFP changes was better than AFP changes alone on predicting PD (AUC: 0.83 vs 0.68, P = 0.034; NRI: 0.340, P = 0.048) and PFS (AUC: 0.80 vs 0.70, P = 0.166; NRI: 0.431, P = 0.042). Dynamic changes of NLR might be an effective predictor of the therapeutic response to ICIs plus targeted therapies for uHCC.

Left Pulmonary Vein Trunk Length as a Robust Predictor of Long-Term Success of Atrial Fibrillation Catheter Ablation.

Radiofrequency catheter ablation (RFCA) is a commonly used treatment for atrial fibrillation (AF), but the long-term recurrence rate remains relatively high. Given the inconsistent results regarding the role of left pulmonary vein (PV) ostial anatomy in post-ablative recurrence of RFCA in previous studies, we sought to investigate the role of left PV trunk length using an alternative methodology. A total of 369 AF patients undergoing catheter ablation were included. The left/right trunk length (LTL/RTL) of the PV was measured from pre-ablative computed tomography (CT) using three-dimensional reconstruction techniques. We constructed three multivariable Cox models, with the inclusion of the LTL, RTL, and no LTL/RTL, and used the Delong test, integrated discrimination index (IDI), and net reclassification index (NRI) to assess model improvement. We identified optimal cut-off values for LTL with the receiver operating characteristic (ROC) curve, and estimated outcomes using the Kaplan-Meier survival curve. We also used subgroup analysis to evaluate interactions. The results of the Delong test, IDI, and NRI indicated that LTL had a favorable impact on the performance of the multivariate model. Subsequently, the multivariate Cox regression analysis identified LTL as a significant risk factor for post-ablative recurrence of AF (adjusted hazard ratio (HR) = 1.08, 95% CI: 1.05-1.12, p 0.001). According to the ROC curve, the optimal cut-off value for LTL is 11.15 mm, and the Kaplan-Meier estimator revealed different outcomes (p 0.001). We calculated p for interaction between LTL and other factors, and no significant interaction terms were observed. LTL is a robust prognostic indicator for post-ablative outcome in AF patients receiving RFCA, with a longer LTL indicating a higher risk of recurrence.

Advancing precision medicine in immunoglobulin light-chain amyloidosis: a novel prognostic model incorporating multi-organ indicators.

To develop a more accurate prognostic model that incorporates indicators of multi-organ involvement for immunoglobulin light-chain (AL) Amyloidosis patients. Biopsy-proven AL amyloidosis patients between January 1, 2012, and February 28, 2023, were enrolled and randomly divided into a training set and a test set at a ratio of 7:3. Prognostic indicators that comprehensively cover cardiac, renal, and hepatic involvement were identified in the training set by random survival forest (RSF). Then, RSF and Cox models were established. The Concordance index (C-index) and integrated brier scores (IBS) were applied to evaluate the models' performance in the test set. Besides, the net reclassification index (NRI) and integrated discrimination improvement (IDI) were calculated. A total of 173 eligible patients were included. After a median follow-up of 25.9 (9.2, 50.3) months, 48 (27.7%) patients died. Creatine kinase-MB, estimated glomerular filtration rate ≤ 50mL/min/1.73m2, interventricular septum ≥ 15mm, ejection fraction, alanine aminotransferase and Live involved were selected to develop prediction models. The RSF model based on the above indicators achieved C-index and IBS values of 0.834 (95% CI 0.725-0.915) and 0.151 (95% CI 0.1402-0.181), respectively. At last, the NRI and IDI of the RSF model were 0.301 (95% CI 0.048-0.546, P = 0.012) and 0.157 (95% CI 0.041-0.269, P < 0.001) at 5-year by comparing the RSF model with the Cox model which is based on the Mayo 2012 staging system. The RSF model that incorporates indicators of multi-organ involvement had a great performance, which may be helpful for physicians' decision-making and more accurate overall survival prediction.

A prediction model based on deep learning and radiomics features of DWI for the assessment of microsatellite instability in endometrial cancer.

To explore the efficacy of a prediction model based on diffusion-weighted imaging (DWI) features extracted from deep learning (DL) and radiomics combined with clinical parameters and apparent diffusion coefficient (ADC) values to identify microsatellite instability (MSI) in endometrial cancer (EC). This study included a cohort of 116 patients with EC, who were subsequently divided into training (n = 81) and test (n = 35) sets. From DWI, conventional radiomics features and convolutional neural network-based DL features were extracted. Random forest (RF) and logistic regression were adopted as classifiers. DL features, radiomics features, clinical variables, ADC values, and their combinations were applied to establish DL, radiomics, clinical, ADC, and combined models, respectively. The predictive performance was evaluated through the area under the receiver operating characteristic curve (AUC), total integrated discrimination index (IDI), net reclassification index (NRI), calibration curves, and decision curve analysis (DCA). The optimal predictive model, based on an RF classifier, comprised four DL features, three radiomics features, two clinical variables, and an ADC value. In the training and test sets, this model exhibited AUC values of 0.989 (95% CI: 0.935-1.000) and 0.885 (95% CI: 0.731-0.967), respectively, demonstrating different degrees of improvement compared with the clinical, DL, radiomics, and ADC models (AUC-training = 0.671, 0.873, 0.833, and 0.814, AUC-test = 0.685, 0.783, 0.708, and 0.713, respectively). The NRI and IDI analyses revealed that the combined model resulted in improved risk reclassification of the MSI status compared to the clinical, radiomics, DL, and ADC models. The calibration curves and DCA indicated good consistency and clinical utility of this model, respectively. The predictive model based on DWI features extracted from DL and radiomics combined with clinical parameters and ADC values could effectively assess the MSI status in EC.

Alteration of Maternal Serum Ferritin in Pregnancy and Maternal-fetal Infections: A retrospective cohort study.

To investigate the association of altered serum ferritin during pregnancy with chorioamnionitis and neonatal sepsis. This retrospective cohort study included 78,521 pregnant women who attended antenatal check-ups at maternal and child health centers in Fujian Province, China. Study lasted from January 2014 to January 2019. A total of 59,812 pregnant women were followed up. Patients with suspected infection before the delivery were selected and divided into the chorioamnionitis and non-chorioamnionitis groups according to placental pathology. Differences in late and early pregnancy serum ferritin between the two groups were compared. Multiple logistics regression was used to adjust for confounding factors and to analyze the association between serum ferritin changes and pregnancy outcomes. Importance of altered serum ferritin during pregnancy was assessed by receiver operating characteristic (ROC) curve and net reclassification index (NRI). Clinical records of 8506 pregnant women were included in the study. there were 1010 (11.9%) cases of confirmed chorioamnionitis and 263 (3.1%) cases of neonatal sepsis. There was a significant difference in maternal serum ferritin changes between the groups with and without chorioamnionitis. No significant difference was detected in cases with or without neonatal sepsis. Multiple logistic regressions, corrected for confounding factors yielded similar conclusions. Maternal serum ferritin difference NRI 12.18% (p = 0.00014) was similar to the ROC results in predicting the occurrence of chorioamnionitis. Differential serum ferritin during pregnancy may predict chorioamnionitis but does not correlate well with neonatal sepsis.

Prognostic Value of Coronary Angiography–Derived Index of Microcirculatory Resistance in Non–ST-Segment Elevation Myocardial Infarction Patients

BackgroundThe index of microcirculatory resistance is a reliable measure for evaluating coronary microvasculature, but its prognostic value in patients with non–ST-segment elevation myocardial infarction (NSTEMI) remains unclear. ObjectivesThis study aimed to evaluate the prognostic impact of postpercutaneous coronary intervention (PCI) angiography-derived index of microcirculatory resistance (angio-IMR) in patients with NSTEMI. MethodsThe culprit vessel’s angio-IMR was measured after PCI in 2,212 NSTEMI patients at 3 sites. The primary endpoint was 2-year major adverse cardiac events (MACEs), defined as a composite of cardiac death, readmission for heart failure, myocardial reinfarction, and target vessel revascularization. ResultsThe mean post-PCI angio-IMR was 20.63 ± 4.17 in NSTEMI patients. A total of 206 patients were categorized as the high post-PCI angio-IMR group according to maximally selected log-rank statistics. Patients with angio-IMR >25 showed a higher rate of MACEs than those with angio-IMR ≤25 (32.52% vs 9.37%; P < 0.001). Post-PCI angio-IMR >25 was an independent predictor of MACEs (HR: 4.230; 95% CI: 3.151-5.679; P < 0.001) and showed incremental prognostic value compared with conventional risk factors (AUC: 0.774 vs 0.716; P < 0.001; net reclassification index: 0.317; P < 0.001; integrated discrimination improvement: 0.075; P < 0.001). ConclusionsIn patients undergoing PCI for NSTEMI, an increased post-PCI angio-IMR is associated with a higher risk of MACEs. The addition of post-PCI angio-IMR into conventional risk factors significantly improves the ability to reclassify patients and estimate the risk of MACEs. (Angiograph-Derived Index of Microcirculatory Resistance in Patients With Acute Myocardial Infarction; NCT05696379)

Incremental value of multidomain risk factors for dementia prediction: A machine learning approach

ObjectiveThe current evidence regarding how different predictor domains contribute to predicting incident dementia remains unclear. This study aims to assess the incremental value of five predictor domains when added to a simple dementia risk prediction model (DRPM) for predicting incident dementia in older adults. DesignPopulation-based, prospective cohort study SettingUK Biobank study ParticipantsIndividuals aged 60 or older without dementia. Measurements55 dementia-related predictors were gathered and categorized into clinical and medical history, questionnaire, cognition, polygenetic risk, and neuroimaging domains. Incident dementia (all-cause) and the subtypes, Alzheimer's disease (AD) and vascular dementia (VaD), were determined through hospital and death registries. Ensemble machine learning (ML) DRPMs were employed for prediction. The incremental values of risk predictors were assessed using the percent change in Area Under the Curve (∆AUC%) and the net reclassification index (NRI). ResultsThe simple DRPM which included age, body mass index, sex, education, diabetes, hyperlipidaemia, hypertension, depression, smoking, and alcohol consumption yielded an AUC of 0.711 (± 0.008 SD). The five predictor domains exhibited varying levels of incremental value over the basic model when predicting all-cause dementia and the two subtypes. Neuroimaging markers provided the highest incremental value in predicting all-cause dementia (∆AUC% +9.6%) and AD (∆AUC% +16.5%) while clinical and medical history data performed the best at predicting VaD (∆AUC% +12.2%). Combining clinical and medical history, and questionnaire data synergistically enhanced ML DRPM performance. ConclusionCombining predictors from different domains generally results in better predictive performance. Selecting predictors entails trade-offs, and while neuroimaging markers can significantly enhance predictive accuracy, they may pose challenges in terms of cost or accessibility.

Incremental prognostic value of coronary hyper-intensity plaque on non-contrast cardiac magnetic resonance with global longitudinal strain for major adverse cardiac events in patients with acute coronary syndrome

Rationale and ObjectivesThis study aims to determine the long-term prognostic value of coronary hyper-intensity plaques and left ventricular (LV) myocardial strain for major adverse cardiac events (MACEs). Materials and MethodsThe study prospectively recruited 71 patients with acute coronary syndrome (ACS). All patients underwent CMR before PCI to determine the plaque-to-myocardium signal intensity ratio and LV strains. The MACEs included all-cause death, reinfarction, and new congestive heart failure. Mann–Whitney U test and chi-square test to compare patients with and without MACE, Kaplan–Meier survival analysis, Cox proportional hazards regression and C-statistics to assess prognosis, Receiver operating characteristic (ROC) curve analysis to define the cutoff value. A P value of <0.05 was considered statistically significant. ResultsCox proportional hazard analysis showed that plaque-to-myocardium signal intensity ratio and global longitudinal strain (GLS) were independently associated with MACEs (plaque-to-myocardium signal intensity ratio: hazard ratio (HR)2.80, 95% CI, 1.25–6.26, P=0.01; GLS: HR1.21, 95% CI, 1.07–1.38, P＜0.01). ROC showed that a plaque-to-myocardium signal intensity ratio of 1.65 and a GLS of -10% were the best cutoff values for MACEs. The C-statistic values for plaque-to-myocardium signal intensity ratio, GLS, and plaque-to-myocardium signal intensity ratio+GLS for MACEs were 0.691, 0.792, and 0.825, respectively. Compared with GLS alone, the addition of plaque-to-myocardium signal intensity ratio to GLS increased the net reclassification index by 0.664 (P=0.017). ConclusionPlaque-to-myocardium signal intensity ratio and GLS were significantly associated with MACEs. Adding plaque-to-myocardium signal intensity ratio to GLS substantially improved the prediction for MACEs. Our findings indicate that plaque-to-myocardium signal intensity ratio combined with GLS provides incremental prognostic value for MACEs.

Ultrafast sequence-based prediction model and nomogram to differentiate additional suspicious lesions on preoperative breast MRI.

To investigate whether ultrafast sequence improves the diagnostic performance of conventional dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) in differentiating additional suspicious lesions (ASLs) on preoperative breast MRI. A retrospective database search identified 668 consecutive patients who underwent preoperative breast DCE-MRI with ultrafast sequence between June 2020 and July 2021. Among these, 107 ASLs from 98 patients with breast cancer (36 multifocal, 42 multicentric, and 29 contralateral) were identified. Clinical, pathological, conventional MRI findings, and ultrafast sequence-derived parameters were collected. A prediction model that adds ultrafast sequence-derived parameters to clinical, pathological, and conventional MRI findings was developed and validated internally. Decision curve analysis and net reclassification index statistics were performed. A nomogram was constructed. The ultrafast model adding time to peak enhancement, time to enhancement, and maximum slope showed a significantly increased area under the receiver operating characteristic curve compared with the conventional model which includes age, human epidermal growth factor receptor 2 expression of index cancer, size of index cancer, lesion type of index cancer, location of ASL, and size of ASL (0.92 vs. 0.82; p = 0.002). The decision curve analysis showed that the ultrafast model had a higher overall net benefit than the conventional model. The net reclassification index of ultrafast model was 23.3% (p = 0.001). A combination of ultrafast sequence-derived parameters with clinical, pathological, and conventional MRI findings can aid in the differentiation of ASL on preoperative breast MRI. Our prediction model and nomogram that was based on ultrafast sequence-derived parameters could help radiologists differentiate ASLs on preoperative breast MRI. Ultrafast MRI can diminish background parenchymal enhancement and possibly improve diagnostic accuracy for additional suspicious lesions (ASLs). Locationof ASL, larger size of ASL, and highermaximum slope were associated with malignant ASL. The ultrafast model and nomogram can help preoperatively differentiate additional malignancies.

A novel real-time model for predicting acute kidney injury in critically ill patients within 12 hours.

A major challenge in prevention and early treatment of acute kidney injury (AKI) is the lack of high-performance predictors in critically ill patients. Therefore, we innovatively constructed U-AKIpredTM for predicting AKI in critically ill patients within 12h of panel measurement. The prospective cohort study included 680 patients in the training set and 249 patients in the validation set. After performing inclusion and exclusion criteria, 417 patients were enrolled in the training set and 164 patients were enrolled in the validation set finally. AKI was diagnosed by Kidney Disease Improving Global Outcomes (KDIGO) criteria. Twelve urinary kidney injury biomarkers (mALB, IgG, TRF, α1MG, NAG, NGAL, KIM-1, L-FABP, TIMP2, IGFBP7, CAF22 and IL-18) exhibited good predictive performance for AKI within 12h in critically ill patients. U-AKIpredTM, combined with three crucial biomarkers (α1MG, L-FABP and IGFBP7) by multivariate logistic regression analysis, exhibited better predictive performance for AKI in critically ill patients within 12h than the other twelve kidney injury biomarkers. The area under the curve (AUC) of the U-AKIpredTM, as a predictor of AKI within 12h, was 0.802 (95% CI: 0.771-0.833, P<0.001) in the training set and 0.844 (95% CI: 0.792-0.896, P<0.001) in validation cohort. A nomogram based on the results of the training and validation sets of U-AKIpredTM was developed which showed optimal predictive performance for AKI. The fitting effect and prediction accuracy of U-AKIpredTM was evaluated by multiple statistical indicators. To provide a more flexible predictive tool, the dynamic nomogram (https://www.xsmartanalysis.com/model/U-AKIpredTM) was constructed using a web-calculator. Decision curve analysis (DCA) and a clinical impact curve were used to reveal that U-AKIpredTM with the three crucial biomarkers had a higher net benefit than these twelve kidney injury biomarkers respectively. The net reclassification index (NRI) and integrated discrimination index (IDI) were used to improve the significant risk reclassification of AKI compared with the 12 kidney injury biomarkers. The predictive efficiency of U-AKIpredTM was better than the NephroCheck® when testing for AKI and severe AKI. U-AKIpredTM is an excellent predictive model of AKI in critically ill patients within 12h and would assist clinicians in identifying those at high risk of AKI.

Bleeding risk prediction after acute myocardial infarction-integrating cancer data: the updated PRECISE-DAPT cancer score.

This study assessed the impact of incorporating cancer as a predictor on performance of the PRECISE-DAPT score. A nationally linked cohort of ST-elevation myocardial infarction patients between 1 January 2005 and 31 March 2019 was derived from the UK Myocardial Ischaemia National Audit Project and the UK Hospital Episode Statistics Admitted Patient Care registries. The primary outcome was major bleeding at 1 year. A new modified score was generated by adding cancer as a binary variable to the PRECISE-DAPT score using a Cox regression model and compared its performance to the original PRECISE-DAPT score. A total of 216 709 ST-elevation myocardial infarction patients were included, of which 4569 had cancer. The original score showed moderate accuracy (C-statistic .60), and the modified score showed modestly higher discrimination (C-statistics .64; hazard ratio 1.03, 95% confidence interval 1.03-1.04) even in patients without cancer (C-statistics .63; hazard ratio 1.03, 95% confidence interval 1.03-1.04). The net reclassification index was .07. The bleeding rates of the modified score risk categories (high, moderate, low, and very low bleeding risk) were 6.3%, 3.8%, 2.9%, and 2.2%, respectively. According to the original score, 65.5% of cancer patients were classified as high bleeding risk (HBR) and 21.6% were low or very low bleeding risk. According to the modified score, 94.0% of cancer patients were HBR, 6.0% were moderate bleeding risk, and no cancer patient was classified as low or very low bleeding risk. Adding cancer to the PRECISE-DAPT score identifies the majority of patients with cancer as HBR and can improve its discrimination ability without undermining its performance in patients without cancer.

CT radiomics combined with clinical and radiological factors predict hematoma expansion in hypertensive intracerebral hemorrhage.

This study aimed to establish a hematoma expansion (HE) prediction model for hypertensive intracerebral hemorrhage (HICH) patients by combining CT radiomics, clinical information, and conventional imaging signs. A retrospective continuous collection of HICH patients from three medical centers was divided into a training set (n = 555), a validation set (n = 239), and a test set (n = 77). Extract radiomics features from baseline CT plain scan images and combine them with clinical information and conventional imaging signs to construct radiomics models, clinical imaging sign models, and hybrid models, respectively. The models will be evaluated using the area under the curve (AUC), clinical decision curve analysis (DCA), net reclassification index (NRI), and integrated discrimination improvement (IDI). In the training, validation, and testing sets, the radiomics model predicts an AUC of HE of 0.885, 0.827, and 0.894, respectively, while the clinical imaging sign model predicts an AUC of HE of 0.759, 0.725, and 0.765, respectively. Glasgow coma scale score at admission, first CT hematoma volume, irregular hematoma shape, and radiomics score were used to construct a hybrid model, with AUCs of 0.901, 0.838, and 0.917, respectively. The DCA shows that the hybrid model had the highest net profit rate. Compared with the radiomics model and the clinical imaging sign model, the hybrid model showed an increase in NRI and IDI. The hybrid model based on CT radiomics combined with clinical and radiological factors can effectively individualize the evaluation of the risk of HE in patients with HICH. CT radiomics combined with clinical information and conventional imaging signs can identify HICH patients with a high risk of HE and provide a basis for clinical-targeted treatment. HE is an important prognostic factor in patients with HICH. The hybrid model predicted HE with training, validation, and test AUCs of 0.901, 0.838, and 0.917, respectively. This model provides a tool for a personalized clinical assessment of early HE risk.

Development and validation of a novel nomogram model predicting the unfavorable outcome based on NAR and collaterals status for patients with AIS

IntroductionStroke is a leading cause of disability and mortality globally. This study aimed to develop a prognostic nomogram based on neutrophil-to-albumin ratio (NAR) and collateral status in acute ischemic stroke (AIS) patients with anterior large vessel occlusion (LVO). Material & method590 AIS patients with LVO assessed for regional leptomeningeal collateral (rLMC) were retrospectively enrolled, and randomly divided into a training set (n = 414) and a testing set (n = 176). Unfavorable functional outcome was defined as a modified Rankin scale (mRS) score of 3 to 6 at 3 months. We assessed the accuracy and clinical utility of the nomogram using calibration plots, area under the curve (AUC), decision curve analysis (DCA), net reclassification index (NRI), and integrated discrimination improvement (IDI). ResultsBoth NAR and rLMC were independently associated with unfavorable outcome at 3 months (OR=8.96, p=0.0341; OR=0.89, p=0.0002, respectively). The developed nomogram (akaike information criterion (AIC)=398.77), which included NAR, rLMC and other factors, showed good performance (the AUC for the development and validation cohorts was 0.848 and 0.840 respectively) and improved the predictive value compared to a model without NAR and rLMC, according to an overall NRI of 3.27% (p=0.2401), overall IDI of 3.27% (p=0.2414), and a higher AUC (0.848 vs 0.831). ConclusionsNAR can serve as an independent predictor in AIS patients with anterior LVO, and the nomogram incorporating NAR and rLMC is reliable in predicting unfavorable outcome. Further studies with larger sample sizes are needed to validate and extend these findings.