Harrell's Concordance Index Research Articles

A nomogram for individualized prediction of mild cognitive impairment in patients with subjective cognitive decline during physical examinations: a cross-sectional study.

To develop a nomogram for mild cognitive impairment (MCI) in patients with subjective cognitive decline (SCD) undergoing physical examinations in China. We enrolled 370 patients undergoing physical examinations at the Medical Center of the First Hospital of Jilin University, Jilin Province, China, from October 2022 to March 2023. Of the participants, 256 were placed in the SCD group, and 74 were placed in the MCI group. The population was randomly divided into a training set and a validation set at a 7:3 ratio. A least absolute shrinkage and selection operator (LASSO) regression model was applied to optimize feature selection for the model. Multivariable logistic regression analysis was applied to construct a predictive model. The performance and clinical utility of the nomogram were determined using Harrell's concordance index, calibration curves, and decision curve analysis (DCA). Cognitive reserve (CR), age, and a family history of hypertension were associated with the occurrence of MCI. The predictive nomogram showed satisfactory performance, with a concordance index of 0.755 (95% CI: 0.681-0.830) in internal verification. The Hosmer-Lemeshow test results suggested that the model exhibited good fit (p = 0.824). In addition, DCA demonstrated that the predictive nomogram had a good clinical net benefit. We developed a simple nomogram that could help secondary preventive health care workers to identify elderly individuals with SCD at high risk of MCI during physical examinations to enable early intervention.

A preliminary probabilistic nomogram model for predicting renal arteriolar damage in IgA nephropathy from clinical parameters.

IgA nephropathy (IgAN) is a significant contributor to chronic kidney disease (CKD). Renal arteriolar damage is associated with IgAN prognosis. However, simple tools for predicting arteriolar damage of IgAN remain limited. We aim to develop and validate a nomogram model for predicting renal arteriolar damage in IgAN patients. We retrospectively analyzed 547 cases of biopsy-proven IgAN patients. Least absolute shrinkage and selection operator (LASSO) regression and logistic regression were applied to screen for factors associated with renal arteriolar damage in patients with IgAN. A nomogram was developed to evaluate the renal arteriolar damage in patients with IgAN. The performance of the proposed nomogram was evaluated based on a calibration plot, ROC curve (AUC) and Harrell's concordance index (C-index). In this study, patients in the arteriolar damage group had higher levels of age, mean arterial pressure (MAP), serum creatinine, serum urea nitrogen, serum uric acid, triglycerides, proteinuria, tubular atrophy/interstitial fibrosis (T1-2) and decreased eGFR than those without arteriolar damage. Predictors contained in the prediction nomogram included age, MAP, eGFR and serum uric acid. Then, a nomogram model for predicting renal arteriolar damage was established combining the above indicators. Our model achieved well-fitted calibration curves and the C-indices of this model were 0.722 (95%CI 0.670-0.774) and 0.784 (95%CI 0.716-0.852) in the development and validation groups, respectively. With excellent predictive abilities, the nomogram may be a simple and reliable tool to predict the risk of renal arteriolar damage in patients with IgAN.

Predictive value of delta-radiomic features for prognosis of advanced non-small cell lung cancer patients undergoing immune checkpoint inhibitor therapy.

No robust predictive biomarkers exist to identify non-small cell lung cancer (NSCLC) patients likely to benefit from immune checkpoint inhibitor (ICI) therapies. The aim of this study was to explore the role of delta-radiomics features in predicting the clinical outcomes of patients with advanced NSCLC who received ICI therapy. Data of 179 patients with advanced NSCLC (stages IIIB-IV) from two institutions (Database 1 =133; Database 2 =46) were retrospectively analyzed. Patients in the Database 1 were randomly assigned into training and validation dataset, with a ratio of 8:2. Patients in Database 2 were allocated into testing dataset. Features were selected from computed tomography (CT) images before and 6-8 weeks after ICI therapy. For each lesion, a total of 1,037 radiomic features were extracted. Lowly reliable [intraclass correlation coefficient (ICC) <0.8] and redundant (r>0.8) features were excluded. The delta-radiomics features were defined as the relative net change of radiomics features between two time points. Prognostic models for progression-free survival (PFS) and overall survival (OS) were established using the multivariate Cox regression based on selected delta-radiomics features. A clinical model and a pre-treatment radiomics model were established as well. The median PFS (after therapy) was 7.0 [interquartile range (IQR): 3.4, 9.1] (range, 1.4-13.2) months. To predict PFS, the model established based on the five most contributing delta-radiomics features yielded Harrell's concordance index (C-index) values of 0.708, 0.688, and 0.603 in the training, validation, and testing databases, respectively. The median survival time was 12 (IQR: 8.7, 15.8) (range, 2.9-23.3) months. To predict OS, a promising prognostic performance was confirmed with the corresponding C-index values of 0.810, 0.762, and 0.697 in the three datasets based on the seven most contributing delta-radiomics features, respectively. Furthermore, compared with clinical and pre-treatment radiomics models, the delta-radiomics model had the highest area under the curve (AUC) value and the best patients' stratification ability. The delta-radiomics model showed a good performance in predicting therapeutic outcomes in advanced NSCLC patients undergoing ICI therapy. It provides a higher predictive value than clinical and the pre-treatment radiomics models.

Comparison of long-term survival of neoadjuvant therapy plus surgery versus upfront surgery and the role of adjuvant therapy for T1b-2N0-1 esophageal cancer: a population study of the SEER database and Chinese cohort.

For stage T1b-2N0-1 esophageal cancer, the impact of neoadjuvant therapy plus surgery (NS), surgery alone (SA), and surgery plus adjuvant therapy (ST) on cancer-specific survival (CSS) and overall survival (OS) is uncertain. Stage T1b-2N0-1 esophageal cancer patients from the SEER database and two Chinese cancer centers were included in this study. The Kaplan-Meier method was used to plot survival curves, which were compared using the log-rank test. Propensity score matching was used to equalize differences between the groups. Cox proportional hazards regression models were used to analyze prognostic factors. A nomogram for OS was developed after screening the variables using the Cox proportional hazards regression model and the least absolute shrinkage and selection operator. The performance of the nomogram was assessed by the Harrell concordance index (C-index), the area under the curve (AUC) of the receiver operating characteristic curve, calibration plots, and decision curve analysis. After propensity score matching analysis, the 3-year CSS and OS rates in the NS group compared to the SA group were 80.3% versus 62.1% (P=0.016) and 75.8% versus 55.5% (P=0.006), the 3-year CSS and OS rates in the NS group compared to the ST group were 71.3% versus 68.3% (P=0.560) and 69.8% versus 62.9% (P=0.330), the 3-year CSS and OS rates in the SA group compared to the ST group were 54.6% versus 66.7% (P=0.220) and 50.2% versus 57.9% (P=0.290), respectively. The predictive nomogram for OS in T1b-2N0-1 patients ultimately incorporated five clinicopathological variables: T stage, N stage, age, examined lymph nodes , and therapy modality. The nomogram C-index for predicting OS was 0.648, 0.663, and 0.666 in the training group, external validation group-1, and external validation group-2, respectively. The 1-, 3-, and 5-year predicted AUC values of the OS prediction model were 0.659, 0.639, and 0.612 for the training group, and 0.786, 0.758, and 0.692 for validation group-1, and 0.805, 0.760, and 0.693 for validation group-2, respectively. For patients with stage T1b-2N0-1 esophageal cancer, neoadjuvant therapy significantly improves prognosis compared to surgery alone, those presenting with positive lymph nodes after upfront surgery can achieve survival benefits from adjuvant therapy.

Triglyceride-glucose index improves risk prediction beyond traditional risk factors and hypertension mediated organ damage in healthy adults

Background and aimsTriglyceride-glucose (TyG) index, a surrogate measure of insulin resistance, is associated with hypertension mediated organ damage (HMOD) and cardiovascular disease. This study investigated the association between TyG index and major adverse cardiovascular events (MACE) and its interaction with traditional risk factors and HMOD. Methods and resultsHealthy subjects recruited from the general population were thoroughly examined and followed for MACE using nation-wide registries. Cox proportional hazard models were used to calculate the association between TyG index and MACE occurrence. Models were adjusted for Systematic Coronary Risk Evaluation (SCORE) risk factors, pulse wave velocity, left ventricular mass index, carotid atherosclerotic plaque status, and microalbuminuria. Continuous net reclassification and Harrell's Concordance index (C-index) were used to assess the added prognostic value of TyG index.During a follow-up period of mean 15.4 ± 4.7 years, MACE were observed in 332 (17%) of 1970 included participants. TyG index was associated with MACE; HR = 1.44 [95%CI:1.30–1.59] per standard deviation. After adjustment for traditional cardiovascular (CV) risk factors, HR was 1.16 [95%CI:1.03–1.31]. The association between TyG index and MACE remained significant after further adjustment for each HMOD component. However, this finding was evident only in subjects aged 41 or 51 years (HR = 1.39; 95%CI:1.15–1.69). Including TyG index in a risk model based on traditional CV risk factors improved C-index with 0.005 (P = 0.042). ConclusionIn this population-based study of healthy middle-aged subjects, TyG index was associated with MACE independently of traditional CV risk factors and HMOD. TyG index may have a potential role in future risk prediction systems.

Radiomics analysis of lesion-specific pericoronary adipose tissue to predict major adverse cardiovascular events in coronary artery disease

ObjectiveTo investigate the prognostic performance of radiomics analysis of lesion-specific pericoronary adipose tissue (PCAT) for major adverse cardiovascular events (MACE) with the guidance of CT derived fractional flow reserve (CT-FFR) in coronary artery disease (CAD).Materials and methodsThe study retrospectively analyzed 608 CAD patients who underwent coronary CT angiography. Lesion-specific PCAT was determined by the lowest CT-FFR value and 1691 radiomic features were extracted. MACE included cardiovascular death, nonfatal myocardial infarction, unplanned revascularization and hospitalization for unstable angina. Four models were generated, incorporating traditional risk factors (clinical model), radiomics score (Rad-score, radiomics model), traditional risk factors and Rad-score (clinical radiomics model) and all together (combined model). The model performances were evaluated and compared with Harrell concordance index (C-index), area under curve (AUC) of the receiver operator characteristic.ResultsLesion-specific Rad-score was associated with MACE (adjusted HR = 1.330, p = 0.009). The combined model yielded the highest C-index of 0.718, which was higher than clinical model (C-index = 0.639), radiomics model (C-index = 0.653) and clinical radiomics model (C-index = 0.698) (all p < 0.05). The clinical radiomics model had significant higher C-index than clinical model (p = 0.030). There were no significant differences in C-index between clinical or clinical radiomics model and radiomics model (p values were 0.796 and 0.147 respectively). The AUC increased from 0.674 for clinical model to 0.721 for radiomics model, 0.759 for clinical radiomics model and 0.773 for combined model.ConclusionRadiomics analysis of lesion-specific PCAT is useful in predicting MACE. Combination of lesion-specific Rad-score and CT-FFR shows incremental value over traditional risk factors.

Development of a novel staging classification for Siewert II adenocarcinoma of the esophagogastric junction after neoadjuvant chemotherapy.

Stage classification for Siewert II adenocarcinoma of the esophagogastric junction (AEG) treated with neoadjuvant chemotherapy (NAC) has not been established. To investigate the optimal stage classification for Siewert II AEG with NAC. A nomogram was established based on Cox regression model that analyzed variables associated with overall survival (OS) and disease-specific survival (DSS). The nomogram performance in terms of discrimination and calibration ability was evaluated using the likelihood-ratio test, Akaike information criterion, Harrell concordance index, time-receiver operating characteristic curve, and decision curve analysis. Data from 725 patients with Siewert type II AEG who underwent neoadjuvant therapy and gastrectomy were obtained from the Surveillance, Epidemiology, and End Results database. Univariate and multivariate analyses revealed that sex, marital status, race, ypT stage, and ypN stage were independent prognostic factors of OS, whereas sex, race, ypT stage, and ypN stage were independent prognostic factors for DSS. These factors were incorporated into the OS and DSS nomograms. Our novel nomogram model performed better in terms of OS and DSS prediction compared to the 8th American Joint Committee of Cancer pathological staging system for esophageal and gastric cancer. Finally, a user-friendly web application was developed for clinical use. The nomogram established specifically for patients with Siewert type II AEG receiving NAC demonstrated good prognostic performance. Validation using external data is warranted before its widespread clinical application.

A nomogram for predicting hemorrhagic shock in pediatric patients with multiple trauma

The timely detection and management of hemorrhagic shock hold paramount importance in clinical practice. This study was designed to establish a nomogram that may facilitate early identification of hemorrhagic shock in pediatric patients with multiple-trauma. A retrospective study was conducted utilizing a cohort comprising 325 pediatric patients diagnosed with multiple-trauma, who received treatment at the Children's Hospital, Zhejiang University School of Medicine, Zhejiang, China. For external validation, an additional cohort of 144 patients from a children's hospital in Taizhou was included. The model's predictor selection was optimized through the application of the Least Absolute Shrinkage and Selection Operator (LASSO) regression. Subsequently, a prediction nomogram was constructed using multivariable logistic regression analysis. The performance and clinical utility of the developed model were comprehensively assessed utilizing various statistical metrics, including Harrell's Concordance Index (C-index), receiver operating characteristic (ROC) curve analysis, calibration curve analysis, and decision curve analysis (DCA). Multivariate logistic regression analysis identified systolic blood pressure (ΔSBP), platelet count, activated partial thromboplastin time (APTT), and injury severity score (ISS) as independent predictors for hemorrhagic shock. The nomogram constructed using these predictors demonstrated robust predictive capabilities, as evidenced by an impressive area under the curve (AUC) value of 0.963. The model's goodness-of-fit was assessed using the Hosmer–Lemeshow test (χ2 = 10.023, P = 0.209). Furthermore, decision curve analysis revealed significantly improved net benefits with the model. External validation further confirmed the reliability of the proposed predictive nomogram. This study successfully developed a nomogram for predicting the occurrence of hemorrhagic shock in pediatric patients with multiple trauma. This nomogram may serve as an accurate and effective tool for timely and efficient management of children with multiple trauma.

Validation of the AJCC 8th Edition Breast Cancer Prognostic Staging System in Legacy Alliance Trials (AFT-01).

The 8th edition American Joint Committee on Cancer staging system combined anatomic stage (AS) with receptor status and grade to create prognostic stage (PS). PS has been validated in single-institution and cancer registry studies; however, missing human epidermal growth factor receptor 2 (HER2) status and variable treatment and follow-up create limitations. Our objective was to compare the relative prognostic ability of PS versus AS to predict survival using breast cancer clinical trial data. Women with non-metastatic breast cancer enrolled in six Alliance for Clinical Trials in Oncology trials were included (enrollment years 1997-2010). AS and PS were constructed using pathological tumor size, nodal status, estrogen receptor (ER), progesterone receptor (PR), HER2 status, and grade. Unadjusted Cox proportional hazard models were estimated to predict overall survival within 5 years, with AS and PS as predictor variables. The relative predictive power of staging models was assessed by comparing Harrell concordance indices (C-indices). Kaplan-Meier-based mortality estimates were compared by stage. Overall, 6924 women were included (median age 53years); 45.2% were diagnosed with ER+/PR+/HER2- tumors, 26.2% with HER2+ tumors, and 17.1% with ER-/PR-/HER2- tumors. Median follow-up time was 5 years (interquartile range 2.95-5.00). PS significantly improved predictive performance (C-index 0.721) for overall survival compared with AS (0.700) (p=0.020). Kaplan-Meier hazard estimates suggested PS did not distinguish mortality risk between patients with IIB and IIIA or IB and IIA disease. PS has significantly improved predictive performance for OS compared with AS. As systemic therapies evolve, it will be important to re-evaluate the prognostic staging system, particularly for patients with intermediate-stage cancers. gov Identifier: NCT02171078.

Performance of anthropometry-based and bio-electrical impedance-based muscle-mass indicators in the Global Leadership Initiative on Malnutrition criteria for predicting prognosis in patients with cancer

BackgroundReduced muscle mass is a criterion for diagnosing malnutrition using the Global Leadership Initiative on Malnutrition (GLIM) criteria; however, the choice of muscle-mass indicators within the GLIM criteria remains contentious. This study aimed to establish muscle-measurement-based GLIM criteria using data from bio-electrical impedance analysis (BIA) and anthropometric evaluations and evaluate their ability to predict overall survival (OS), short-term outcomes, and healthcare burden in patients with cancer. MethodsThis was a multicenter, prospective study that commenced in 2013 and enrolled participants from various clinical centers across China. We constructed GLIM criteria based on various muscle measurements, including fat-free mass index (FFMI), skeletal muscle index (SMI), calf circumference (CC), midarm circumference (MAC), midarm muscle circumference (MAMC), and midarm muscle area (MAMA). Survival was estimated using the Kaplan-Meier method and survival curves were compared using the log-rank test. Cox proportional hazards regression was used to assess the independent association between the GLIM criteria and OS. The discriminatory performance of different muscle-measurement-based GLIM criteria for mortality was evaluated using Harrell's concordance index (C-index). Logistic regression was used to evaluate the association of the GLIM criteria with short-term outcomes and healthcare burden. ResultsA total of 4,769 patients were included in the analysis, of whom 1,659 (34.8%) died during the study period. The Kaplan-Meier curves demonstrated that all muscle-measurement-based GLIM criteria significantly predicted survival in patients with cancer (all p<0.001). The survival rate of malnourished patients was approximately 10% lower than that of non-malnourished patients. Cox proportional hazards regression showed that all the muscle-measurement-based GLIM could independently predict the OS of patients (all p<0.001). The prognostic discrimination was as follows: MAMC (Chi-square: 79.61) > MAMA (Chi-square: 79.10) > MAC (Chi-square: 64.09) > FFMI (Chi-square: 62.33) > CC (Chi-square: 58.62) > ASMI (Chi-square: 57.29). In comparison to the FFMI-based GLIM criteria, the ASMI-based criteria (-0.002, 95% CI: -0.006-0.002, p=0.334) and CC-based criteria (-0.003, 95% CI: -0.007-0.002, p=0.227) did not exhibit a significant advantage. However, the MAC-based criteria (0.001, 95% CI: -0.003-0.004, p=0.776), MAMA-based criteria (0.004, 95% CI: 0.000-0.007, p=0.035), and MAMC-based criteria (0.005, 95% CI: 0.000-0.007, p=0.030) outperformed the FFMI-based GLIM criteria. Logistic regression showed that muscle measurement-based GLIM criteria predicted short-term outcomes and length of hospital stay in patients with cancer. ConclusionAll muscle measurement-based GLIM criteria can effectively predict OS, short-term outcomes, and healthcare burden in patients with cancer. Anthropometric measurement-based GLIM criteria have potential for clinical application as an alternative to BIA-based measurement.

#426 “Predict-PD”: a machine learning tool to predict patient survival in peritoneal dialysis patients

Abstract Background and Aims With the rising prevalence of chronic kidney disease (CKD), particularly among the elderly, the need for effective renal replacement therapy has become increasingly crucial. Peritoneal dialysis (PD), recognized as a gentler alternative to haemodialysis, offers advantages in cognitive preservation and overall quality of life. Patient survival on PD, a tangible outcome measure, gains particular importance, especially in an aging demographic confronting end-stage kidney disease. Predicting survival thus becomes vital in facilitating individualized treatment planning and has implications on resource allocation and patient counseling. Method Our study analysed data from the UK Renal Registry and included 22 711 incident dialysis patients choosing PD between 2007 and 2022. Our objective was to employ Artificial Intelligence (AI) algorithms in developing an advanced risk stratification indicator with high predictive capabilities, specifically tailored for use in the United Kingdom's transplant selection procedure. We conducted experiments using three distinct machine learning models and evaluated their performance in terms of calibration and discrimination, utilizing metrics such as the integrated Brier score (IBS) and Harrell's concordance index. We assessed the potential clinical utility using decision curve analysis. Results The XGBoost model demonstrated excellent predictive efficacy for patient survival, with a concordance of 0.78. Further scrutiny revealed AUC values at 1 year (0.81), 3 years (0.78), and 5 years (0.77). The integrated Brier score, a comprehensive measure of predictive accuracy, was 0.09. There was no statistical difference between the actual and predicted survival probabilities (P = 0.72). Decision curve analysis accentuated the model's clinical applicability. Conclusion This study illustrates the potential of machine learning in prognosticating patient survival in PD, offering a pathway towards individualized patient-centric management in nephrology. The findings advocate for the integration of data-driven methodologies in clinical decision-making, paving the way for enhanced personalized care in CKD management and can serve as a guide for future research.

#1427 Comparison of METS-IR and HOMA-IR for predicting new-onset CKD in middle-aged and older adults

Abstract Background and Aims Chronic kidney disease (CKD) has emerged as a mounting public health issue worldwide; therefore, prompt identification and prevention are imperative in mitigating CKD-associated complications and mortality rate. We aimed to compare the predictive powers of the homeostatic model assessment for insulin resistance (HOMA-IR) and the metabolic score for insulin resistance (METS-IR) for CKD incidence in middle-aged and older adults. Method This study used longitudinal prospective cohort data from the Korean Genome and Epidemiology Study. A total of 10,030 participants, aged 40–69 years, residing in the Ansung or Ansan regions of the Republic of Korea, were recruited between 2001 and 2002 through a two-stage cluster sampling method. We compared the predictive powers of METS-IR and HOMA-IR for CKD prevalence and incidence, respectively. CKD prevalence was measured by the area under the receiver operating characteristic (ROC) curve (AUC), and the indices’ predictive performance for CKD incidence were assessed using Harrell's concordance index and time-dependent ROC curve analysis. Results A total of 9261 adults aged 40–69 years at baseline and 8243 adults without CKD were included in this study. The AUCs and 95% confidence intervals (CIs) of HOMA-IR and METS-IR for CKD prevalence at baseline were 0.577 (0.537–0.618) and 0.599 (0.560–0.637), respectively, with no significant difference (p = 0.337). The Heagerty's integrated AUC for METS-IR in predicting CKD incidence was 0.772 (95% CI 0.750–0.799), which was significantly higher than that of HOMA-IR (0.767 [95% CI 0.742–0.791], p = 0.015). Conclusion METS-IR surpassed HOMA-IR in predicting CKD incidence and was as effective as HOMA-IR in predicting CKD prevalence. This implies that METS-IR could be a valuable indicator for early detection and prevention of CKD among Korean adults.

Assessing blood sugar measures for predicting new-onset diabetes and cardiovascular disease in community-dwelling adults.

Diabetes mellitus (DM) is a global health concern linked to various complications, including cardiovascular disease (CVD). However, long-term follow-up studies on the risk of DM and CVD using different blood glucose assessment methods in the general Korean population are lacking. This study aimed to assess the predictive abilities of fasting plasma glucose (FPG), 2-h oral glucose tolerance test (OGTT), and glycosylated hemoglobin (HbA1c) for new-onset DM and high CVD risk in a middle-aged and older Korean population. This study used data from the Korean Genome and Epidemiology Study, a population-based prospective cohort. Blood sugar measures (FPG, OGTT, and HbA1c) were examined. The primary endpoint was the development of new-onset DM, and CVD risk was evaluated using the Framingham risk score. The predictive abilities for new-onset DM based on glycemic values were evaluated using Harrell's Concordance index and 95% confidence intervals. Among the 10,030 participants, data of 6813 participants without DM at baseline were analyzed. The study revealed that OGTT outperformed FPG and HbA1c in predicting new-onset DM. The combination of FPG and HbA1c did not significantly enhance predictions for DM compared with OGTT alone. OGTT also outperformed FPG and HbA1c in predicting high CVD risk, and this difference remained significant even after adjusting for additional confounders. OGTT has superior predictive capabilities in identifying new-onset DM and high CVD risk in the Korean population. This suggests that relying solely on individual blood sugar measures may be insufficient for assessing DM and CVD risks.

Development and validation of a nomogram integrating marital status for 5-year overall survival of chondrosarcoma: a population-based study.

The objective of this study was to evaluate the influence of marital status on overall survival (OS) and develop a nomogram for predicting 5-year OS in chondrosarcoma (CHS) patients. We utilized the Surveillance, Epidemiology, and End Results (SEER) database to identify CHS patients diagnosed between 2010 and 2018. Survival rates were calculated using Kaplan-Meier analysis. Prognostic factors were identified through univariate and multivariate analyses. An independent cohort was used for external validation of the nomogram. Performance evaluation of the nomogram was conducted using Harrell's concordance index (C-index), calibration plot, and decision curve analysis (DCA). In the SEER cohort, Kaplan-Meier analysis showed significant differences in OS among CHS patients with different marital statuses (P < 0.001), with widowed patients having the lowest OS. In terms of gender, there were significant survival differences based on marital status in females (P < 0.001), but not in males (P = 0.067). The OS of married and single females is significantly higher than that of married (P < 0.001) and single male (P = 0.006), respectively. Kaplan-Meier curves showed no significant difference in OS between groups stratified by either gender or marital status in the external cohort. Univariate and multivariate analyses confirmed that age at diagnosis, gender, marital status, tumor size, histological type, tumor grade, SEER stage, and surgery were independent prognostic factors for OS. The nomogram demonstrated high internal and external validation C-indexes of 0.818 and 0.88, respectively. Calibration plots, DCA curve, and Kaplan-Meier curve (P < 0.001) confirmed the excellent performance and clinical utility of the nomogram. Marital status was an independent factor influencing OS in CHS patients, with widowed patients having the worst prognosis. The OS of both married and single females is significantly higher than that of their male counterparts. However, these findings require further validation in a large independent cohort. While the contribution of marital status on predicting OS appears modest, our nomogram accurately predicted 5-year OS and identified high-risk groups, providing a valuable tool for clinical decision-making.

Nomogram for predicting post-therapy recurrence in BCLC A/B hepatocellular carcinoma with Child-Pugh B cirrhosis.

This study conducts a retrospective analysis on patients with BCLC stage A/B hepatocellular carcinoma (HCC) accompanied by Child-Pugh B cirrhosis, who underwent transarterial chemoembolization (TACE) in combination with local ablation therapy. Our goal was to uncover risk factors contributing to post-treatment recurrence and to develop and validate an innovative 1-, 3-, and 5-year recurrence free survival (RFS) nomogram. Data from 255 BCLC A/B HCC patients with Child-Pugh B cirrhosis treated at Beijing You'an Hospital (January 2014 - January 2020) were analyzed using random survival forest (RSF), LASSO regression, and multivariate Cox regression to identify independent risk factors for RFS. The prognostic nomogram was then constructed and validated, categorizing patients into low, intermediate, and high-risk groups, with RFS assessed using Kaplan-Meier curves. The nomogram, integrating the albumin/globulin ratio, gender, tumor number, and size, showcased robust predictive performance. Harrell's concordance index (C-index) values for the training and validation cohorts were 0.744 (95% CI: 0.703-0.785) and 0.724 (95% CI: 0.644-0.804), respectively. The area under the curve (AUC) values for 1-, 3-, and 5-year RFS in the two cohorts were also promising. Calibration curves highlighted the nomogram's reliability and decision curve analysis (DCA) confirmed its practical clinical benefits. Through meticulous patient stratification, we also revealed the nomogram's efficacy in distinguishing varying recurrence risks. This study advances recurrence prediction in BCLC A/B HCC patients with Child-Pugh B cirrhosis following TACE combined with ablation. The established nomogram accurately predicts 1-, 3-, and 5-year RFS, facilitating timely identification of high-risk populations.

Performance of the EORTC and CUETO Models to Predict Recurrence and Progression in High-risk Non-muscle-invasive Bladder Cancer Patients.

To evaluate the performance of the European Organization for Research and Treatment of Cancer (EORTC) and the Spanish Urological Club for Oncological Treatment (CUETO) risk scoring models in non-muscle-invasive bladder cancer (NMIBC) patients defined as high risk according to European Association of Urology guidelines and managed based on current recommendations. Data from 187 high-risk NMIBC patients treated at a tertiary center between July 2010 and November 2021 were analyzed retrospectively. One- and five-year recurrence- and progression-free survival were assessed for each patient using the EORTC and CUETO risk scores. The patients were divided into four risk groups according to their risk scores as low, medium-low, medium-high and high risk, as indicated in the models. Discriminative ability was evaluated with the Harrell's concordance index (c-index). Both risk scoring models overestimated the risk of recurrence and progression at one and five years. Only the prediction of recurrence at five years in the high risk group according to the CUETO model was compatible with our cohort. CUETO (c-indices for recurrence and progression were 0.802 and 0.834, respectively) exhibited better discrimination than EORTC (0.722 for recurrence and 0.752 for progression) in the prediction of disease recurrence and progression. The CUETO model was superior to the EORTC model in predicting recurrence and progression and stratifying patients with different prognoses in our high-risk NMIBC patient population treated according to current guideline recommendations. However, both models overestimated the probability of disease recurrence and progression. Only the probability of recurrence at five years in the high-risk group of the CUETO model was compatible with our cohort.

Development and Validation of a Multimodality Model Based on Whole-Slide Imaging and Biparametric MRI for Predicting Postoperative Biochemical Recurrence in Prostate Cancer.

Purpose To develop and validate a machine learning multimodality model based on preoperative MRI, surgical whole-slide imaging (WSI), and clinical variables for predicting prostate cancer (PCa) biochemical recurrence (BCR) following radical prostatectomy (RP). Materials and Methods In this retrospective study (September 2015 to April 2021), 363 male patients with PCa who underwent RP were divided into training (n = 254; median age, 69 years [IQR, 64-74 years]) and testing (n = 109; median age, 70 years [IQR, 65-75 years]) sets at a ratio of 7:3. The primary end point was biochemical recurrence-free survival. The least absolute shrinkage and selection operator Cox algorithm was applied to select independent clinical variables and construct the clinical signature. The radiomics signature and pathomics signature were constructed using preoperative MRI and surgical WSI data, respectively. A multimodality model was constructed by combining the radiomics signature, pathomics signature, and clinical signature. Using Harrell concordance index (C index), the predictive performance of the multimodality model for BCR was assessed and compared with all single-modality models, including the radiomics signature, pathomics signature, and clinical signature. Results Both radiomics and pathomics signatures achieved good performance for BCR prediction (C index: 0.742 and 0.730, respectively) on the testing cohort. The multimodality model exhibited the best predictive performance, with a C index of 0.860 on the testing set, which was significantly higher than all single-modality models (all P ≤ .01). Conclusion The multimodality model effectively predicted BCR following RP in patients with PCa and may therefore provide an emerging and accurate tool to assist postoperative individualized treatment. Keywords: MR Imaging, Urinary, Pelvis, Comparative Studies Supplemental material is available for this article. © RSNA, 2024.

Deep learning model integrating radiologic and clinical data to predict mortality after ischemic stroke

Most prognostic indexes for ischemic stroke mortality lack radiologic information. We aimed to create and validate a deep learning-based mortality prediction model using brain diffusion weighted imaging (DWI), apparent diffusion coefficient (ADC), and clinical factors. Data from patients with ischemic stroke who admitted to tertiary hospital during acute periods from 2013 to 2019 were collected and split into training (n=1109), validation (n=437), and internal test (n=654). Data from patients from secondary cardiovascular center was used for external test set (n=507). The algorithm for predicting mortality, based on DWI and ADC (DLP_DWI), was initially trained. Subsequently, important clinical factors were integrated into this model to create the integrated model (DLP_INTG). The performance of DLP_DWI and DLP_INTG was evaluated by using time-dependent area under the receiver operating characteristic curves (TD AUCs) and Harrell concordance index (C-index) at one-year mortality. The TD AUC of DLP_DWI was 0.643 in internal test set, and 0.785 in the external dataset. DLP_INTG had a higher performance at predicting one-year mortality than premise score in internal dataset (TD- AUC: 0.859 vs. 0.746; p=0.046), and in external dataset (TD- AUC: 0.876 vs. 0.808; p=0.007). DLP_DWI and DLP_INTG exhibited strong discrimination for the high-risk group for one-year mortality. A deep learning model using brain DWI, ADC and the clinical factors was capable of predicting mortality in patients with ischemic stroke.

Development and validation of [18F]-PSMA-1007 PET-based radiomics model to predict biochemical recurrence-free survival following radical prostatectomy.

Biochemical recurrence (BCR) following radical prostatectomy (RP) is a significant concern for patients with prostate cancer. Reliable prediction models are needed to identify patients at risk for BCR and facilitate appropriate management. This study aimed to develop and validate a clinical-radiomics model based on preoperative [18F]PSMA-1007 PET for predicting BCR-free survival (BRFS) in patients who underwent RP for prostate cancer. A total of 236 patients with histologically confirmed prostate cancer who underwent RP were retrospectively analyzed. All patients had a preoperative [18F]PSMA-1007 PET/CT scan. Radiomics features were extracted from the primary tumor region on PET images. A radiomics signature was developed using the least absolute shrinkage and selection operator (LASSO) Cox regression model. The performance of the radiomics signature in predicting BRFS was assessed using Harrell's concordance index (C-index). The clinical-radiomics nomogram was constructed using the radiomics signature and clinical features. The model was externally validated in an independent cohort of 98 patients. The radiomics signature comprised three features and demonstrated a C-index of 0.76 (95% CI: 0.60-0.91) in the training cohort and 0.71 (95% CI: 0.63-0.79) in the validation cohort. The radiomics signature remained an independent predictor of BRFS in multivariable analysis (HR: 2.48, 95% CI: 1.47-4.17, p < 0.001). The clinical-radiomics nomogram significantly improved the prediction performance (C-index: 0.81, 95% CI: 0.66-0.95, p = 0.007) in the training cohort and (C-index: 0.78 95% CI: 0.63-0.89, p < 0.001) in the validation cohort. We developed and validated a novel [18F]PSMA-1007 PET-based clinical-radiomics model that can predict BRFS following RP in prostate cancer patients. This model may be useful in identifying patients with a higher risk of BCR, thus enabling personalized risk stratification and tailored management strategies.

A nomogram for predicting postoperative delirium in pediatric patients following cardiopulmonary bypass: A prospective observational study

ObjectivesTo create a nomogram for early delirium detection in pediatric patients following cardiopulmonary bypass. Research Methodology/DesignThis prospective, observational study was conducted in the Cardiac Intensive Care Unit at a Children's Hospital, enrolling 501 pediatric patients from February 2022 to January 2023. Perioperative data were systematically collected through the hospital information system. Postoperative delirium was assessed using the Cornell Assessment of Pediatric Delirium (CAPD). For model development, Least Absolute Shrinkage and Selection Operator (LASSO) regression was employed to identify the most relevant predictors. These selected predictors were then incorporated into a multivariable logistic regression model to construct the predictive nomogram. The performance of the model was evaluated by Harrell's concordance index, receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis. External validity of the model was confirmed through the C-index and calibration plots. ResultsFive independent predictors were identified: age, SpO2 levels, lymphocyte count, diuretic use, and midazolam administration, integrated into a predictive nomogram. This nomogram demonstrated strong predictive capacity (AUC 0.816, concordance index 0.815) with good model fit (Hosmer-Lemeshow test p = 0.826) and high accuracy. Decision curve analysis showed a significant net benefit, and external validation confirmed the nomogram's reliability. ConclusionsThe study successfully developed a precise and effective nomogram for identifying pediatric patients at high risk of post-cardiopulmonary bypass delirium, incorporating age, SpO2 levels, lymphocyte counts, diuretic use, and midazolam medication. Implications for Clinical PracticeThis nomogram aids early delirium detection and prevention in critically ill children, improving clinical decisions and treatment optimization. It enables precise monitoring and tailored medication strategies, significantly contributes to reducing the incidence of delirium, thereby enhancing the overall quality of patient care.