Risk factors of preoperative deep vein thrombosis in patients with non-traumatic osteonecrosis of the femoral head

PURPOSE

This study aims to identify independent risk factors for preoperative lower extremity deep venous thrombosis (DVT) in patients with non-traumatic osteonecrosis of the femoral head (NONFH), and to develop a prediction nomogram.

METHODS

Retrospective analysis of prospectively collected data on patients presenting with non-traumatic osteonecrosis of the femoral head between October 2014 and April 2019 was conducted. Duplex ultrasonography (DUS) was routinely used to screen for preoperative DVT of bilateral lower extremities. Data on demographics, chronic comorbidities, preoperative characteristics, and laboratory biomarkers were collected. Univariate analyses and multivariate logistic regression analyses were used to identify the independent risk factors associated with DVT which were combined and transformed into a nomogram model.

RESULT

Among 2824 eligible patients included, 35 (1.24%) had preoperative DVT, including 15 cases of proximal thrombosis, and 20 cases of distal thrombosis. Six independent risk factors were identified to be associated with DVT, including Sodium ≤ 137 mmol/L (OR = 2.116, 95% confidence interval [CI]: 1.036-4.322; P = 0.040), AGE ≥ 49 years (OR = 7.598, 95%CI: 1.763-32.735; P = 0.008), D-Dimer > 0.18 mg/L (OR = 2.351, 95%CI: 1.070-5.163; P = 0.033), AT III ≤ 91.5% (OR = 2.796, 95%CI: 1.387-5.634; P = 0.006), PLT ≥ 220.4*10⁹ /L (OR = 7.408, 95%CI: 3.434-15.981; P = 0.001) and ALB < 39 g/L (OR = 3.607, 95%CI: 1.084-12.696; P = 0.042). For the nomogram model, AUC was 0.845 (95%CI: 0.785-0.906), and C-index was 0.847 with the corrected value of 0.829 after 1000 bootstrapping validations. Moreover, the calibration curve and DCA exhibited the tool's good prediction consistency and clinical practicability.

CONCLUSION

These epidemiologic data and the nomogram may be conducive to the individualized assessment, risk stratification, and development of targeted prevention programs for preoperative DVT in patients with NONFH.

Low-density lipoprotein cholesterol-to-lymphocyte count ratio (LLR) is a promising novel predictor of postoperative new-onset deep vein thrombosis following open wedge high tibial osteotomy: a propensity score-matched analysis

BACKGROUND

The association of low-density lipoprotein cholesterol (LDL-C) and lymphocyte counts with the development of deep vein thrombosis (DVT) has been demonstrated in many fields but remains lacking in open wedge high tibial osteotomy (OWHTO). This study aimed to assess the predictive value of LDL-C to lymphocyte count ratio (LLR) in screening for postoperative new-onset DVT.

METHODS

Clinical data were retrospectively collected from patients who underwent OWHTO between June 2018 and May 2023. The limited restricted cubic spline (RCS) was conducted to evaluate the nonlinear relationship between LLR and the risk of postoperative new-onset DVT. The receiver operating characteristic (ROC) curves were plotted and the predictive value of biomarkers was assessed. After adjusting for intergroup confounders by propensity score matching, the univariate logistic regression was applied to assess the association between LLR and DVT.

RESULTS

1293 eligible patients were included. RCS analysis showed a linear positive correlation between LLR and the risk of DVT (P for overall = 0.008). We identified LLR had an area under the curve of 0.607, accuracy of 74.3%, sensitivity of 38.5%, and specificity of 80.7%, and LLR > 1.75 was independently associated with a 1.45-fold risk of DVT (95% CI: 1.01-2.08, P = 0.045). Furthermore, significant heterogeneities were observed in the subgroups of age, BMI, diabetes mellitus, hypertension, Kellgren-Lawrence grade, the American Society of Anesthesiologists (ASA) score, and intraoperative osteotomy correction size.

CONCLUSION

LLR is a valuable biomarker for predicting postoperative new-onset DVT in patients with OWHTO, and routine screening is expected to yield positive benefits.

Combining multiple biomarkers linearly to minimize the Euclidean distance of the closest point on the receiver operating characteristic surface to the perfection corner in trichotomous settings

The performance of individual biomarkers in discriminating between two groups, typically the healthy and the diseased, may be limited. Thus, there is interest in developing statistical methodologies for biomarker combinations with the aim of improving upon the individual discriminatory performance. There is extensive literature referring to biomarker combinations under the two-class setting. However, the corresponding literature under a three-class setting is limited. In our study, we provide parametric and nonparametric methods that allow investigators to optimally combine biomarkers that seek to discriminate between three classes by minimizing the Euclidean distance from the receiver operating characteristic surface to the perfection corner. Using this Euclidean distance as the objective function allows for estimation of the optimal combination coefficients along with the optimal cutoff values for the combined score. An advantage of the proposed methods is that they can accommodate biomarker data from all three groups simultaneously, as opposed to a pairwise analysis such as the one implied by the three-class Youden index. We illustrate that the derived true classification rates exhibit narrower confidence intervals than those derived from the Youden-based approach under a parametric, flexible parametric, and nonparametric kernel-based framework. We evaluate our approaches through extensive simulations and apply them to real data sets that refer to liver cancer patients.

Cutoff estimation and construction of their confidence intervals for continuous biomarkers under ternary umbrella and tree stochastic ordering settings

Tuberculosis (TB) studies often involve four different states under consideration, namely, "healthy," "latent infection," "pulmonary active disease," and "extra-pulmonary active disease." While highly accurate clinical diagnosis tests do exist, they are expensive and generally not accessible in regions where they are most needed; thus, there is an interest in assessing the accuracy of new and easily obtainable biomarkers. For some such biomarkers, the typical stochastic ordering assumption might not be justified for all disease classes under study, and usual ROC methodologies that involve ROC surfaces and hypersurfaces are inadequate. Different types of orderings may be appropriate depending on the setting, and these may involve a number of ambiguously ordered groups that stochastically exhibit larger (or lower) marker scores than the remaining groups. Recently, there has been scientific interest on ROC methods that can accommodate these so-called "tree" or "umbrella" orderings. However, there is limited work discussing the estimation of cutoffs in such settings. In this article, we discuss the estimation and inference around optimized cutoffs when accounting for such configurations. We explore different cutoff alternatives and provide parametric, flexible parametric, and non-parametric kernel-based approaches for estimation and inference. We evaluate our approaches using simulations and illustrate them through a real data set that involves TB patients.

Development and validation of a nomogram for predicting the risk of immediate postoperative deep vein thrombosis after open wedge high tibial osteotomy

PURPOSE

This study aimed to identify independent risk factors for immediate postoperative deep vein thrombosis (DVT) in patients with open wedge high tibial osteotomy (OWHTO) and to develop and validate a predictive nomogram.

METHODS

Patients who underwent OWHTO for knee osteoarthritis (KOA) from June 2017 to December 2021 were retrospectively analyzed. Baseline data and laboratory test results were collected, and the occurrence of DVT in the immediate postoperative period was regarded as the study outcome event. Multivariable logistic regression identified independent risk factors associated with a higher incidence of immediate postoperative DVT. The predictive nomogram was constructed based on the analysis results. The stability of the model was further assessed in this study using patients from January to September 2022 as an external validation set.

RESULTS

741 patients were enrolled in the study, of which 547 were used in the training cohort and the other 194 for the validation cohort. Multivariate analysis revealed a higher Kellgren-Lawrence (K-L) grade (III vs. I-II OR 3.09, 95% CI 0.93-10.23. IV vs. I-II OR 5.23, 95% CI 1.27-21.48.), platelet to hemoglobin ratio (PHR) > 2.25 (OR 6.10, 95% CI 2.43-15.33), Low levels of albumin (ALB) (OR 0.79, 95% CI 0.70-0.90), LDL-C > 3.40 (OR 3.06, 95% CI 1.22-7.65), D-dimer > 1.26 (OR 2.83, 95% CI 1.16-6.87) and BMI ≥ 28 (OR 2.57, 95% CI 1.02-6.50) were the independent risk factors of immediate postoperative DVT. The concordance index (C-index) and Brier score of the nomogram were 0.832 and 0.036 in the training set, and the corrected values after internal validation were 0.795 and 0.038, respectively. The receiver-operating characteristic (ROC) curve, the calibration curve, the Hosmer-Lemeshow test, and the decision curve analysis (DCA) performed well in both the training and validation cohorts.

CONCLUSION

This study developed a personalized predictive nomogram with six predictors, which allows surgeons to stratify risk and recommended immediate ultrasound scans for patients with any of these factors.

LEVEL OF EVIDENCE

III.

Salvage radiotherapy strategy and its prognostic significance for patients with locoregional recurrent cervical cancer after radical hysterectomy: a multicenter retrospective 10-year analysis

OBJECTIVE

We aimed to evaluate the clinical efficacy and prognostic significance of intensity-modulated radiotherapy (IMRT)-based salvage concurrent chemoradiotherapy (CCRT) for patients with locoregional recurrence cervical cancer after radical hysterectomy and evaluated two salvage radiotherapy modes-regional RT (involved-field RT combined with regional lymph nodes) and local RT (involved-field RT).

METHODS

Patients were enrolled retrospectively from January 2011 to January 2022 in three medical centers. Clinical outcomes were analyzed using the Kaplan-Meier method and a Cox proportional hazards model. Propensity score (PS) matching analysis was used to compare the two RT groups.

RESULTS

There were 72 patients underwent IMRT-based salvage CCRT. The 5-year overall survival and progression-free survival rates were 65.9% and 57.6%, respectively. Univariate analysis showed that patients with stump recurrence, a lower systemic inflammation response index (SIRI), only one metastatic lesion, and received regional RT had better prognosis than their counterparts. In multivariate analysis, recurrence site was the independent prognostic factor of OS, and SIRI was that of PFS. After PS matching, there were 15 patients each in the regional RT group and local RT group. The 5-year OS rate of regional RT group was better than that of local RT group (90.9 vs. 42.4, p = 0.021). However, there was no significant difference between them in terms of PFS rate (47.1 vs. 38.1, p = 0.195).

CONCLUSION

Locoregional recurrent cervical cancer treated with IMRT-based salvage therapy has a good prognosis. Recurrence site and SIRI were independent prognostic factors. Regional RT may be a better option for patients with locoregional recurrent.

Statistical inference for the two-sample problem under likelihood ratio ordering, with application to the ROC curve estimation

The receiver operating characteristic (ROC) curve is a powerful statistical tool and has been widely applied in medical research. In the ROC curve estimation, a commonly used assumption is that larger the biomarker value, greater severity the disease. In this article, we mathematically interpret "greater severity of the disease" as "larger probability of being diseased." This in turn is equivalent to assume the likelihood ratio ordering of the biomarker between the diseased and healthy individuals. With this assumption, we first propose a Bernstein polynomial method to model the distributions of both samples; we then estimate the distributions by the maximum empirical likelihood principle. The ROC curve estimate and the associated summary statistics are obtained subsequently. Theoretically, we establish the asymptotic consistency of our estimators. Via extensive numerical studies, we compare the performance of our method with competitive methods. The application of our method is illustrated by a real-data example.

Risk factors for deep vein thrombosis in patients with pelvic or lower-extremity fractures in the emergency intensive care unit

Introduction

This study aimed to investigate the incidence of deep vein thrombosis (DVT) in patients with pelvic or lower-extremity fractures in the emergency intensive care unit (EICU), explore the independent risk factors for DVT, and investigate the predictive value of the Autar scale for DVT in these patients.

Methods

The clinical data of patients with single fractures of the pelvis, femur, or tibia in the EICU from August 2016 to August 2019 were retrospectively examined. The incidence of DVT was statistically analyzed. Logistic regression was used to analyze the independent risk factors for DVT in these patients. The receiver-operating characteristic (ROC) curve was used to evaluate the predictive value of the Autar scale for the risk of DVT.

Results

A total of 817 patients were enrolled in this study; of these, 142 (17.38%) had DVT. Significant differences were found in the incidence of DVT among the pelvic fractures, femoral fractures, and tibial fractures (P < 0.001). The multivariate logistic regression analysis showed multiple injuries (OR = 2.210, 95% CI: 1.166-4.187, P = 0.015), fracture site (compared with tibia fracture group, femur fracture group OR = 4.839, 95% CI: 2.688-8.711, P < 0.001; pelvic fracture group OR = 2.210, 95% CI: 1.225-3.988, P = 0.008), and Autar score (OR = 1.198, 95% CI: 1.016-1.353, P = 0.004) were independent risk factors for DVT in patients with pelvic or lower-extremity fractures in the EICU. The area under the ROC curve (AUROC) of the Autar score for predicting DVT was 0.606. When the Autar score was set as the cutoff value of 15.5, the sensitivity and specificity for predicting DVT in patients with pelvic or lower-extremity fractures were 45.1% and 70.7%, respectively.

Discussion

Fracture is a high-risk factor for DVT. Patients with a femoral fracture or multiple injuries have a higher risk of DVT. In the case of no contraindications, DVT prevention measures should be taken for patients with pelvic or lower-extremity fractures. Autar scale has a certain predictive value for the occurrence of DVT in patients with pelvic or lower-extremity fractures, but it is not ideal.

Confidence intervals and sample size planning for optimal cutpoints

Various methods are available to determine optimal cutpoints for diagnostic measures. Unfortunately, many authors fail to report the precision at which these optimal cutpoints are being estimated and use sample sizes that are not suitable to achieve an adequate precision. The aim of the present study is to evaluate methods to estimate the variance of cutpoint estimations based on published descriptive statistics ('post-hoc') and to discuss sample size planning for estimating cutpoints. We performed a simulation study using widely-used methods to optimize the Youden index (empirical, normal, and transformed normal method) and three methods to determine confidence intervals (the delta method, the parametric bootstrap, and the nonparametric bootstrap). We found that both the delta method and the parametric bootstrap are suitable for post-hoc calculation of confidence intervals, depending on the sample size, the distribution of marker values, and the correctness of model assumptions. On average, the parametric bootstrap in combination with normal-theory-based cutpoint estimation has the best coverage. The delta method performs very well for normally distributed data, except in small samples, and is computationally more efficient. Obviously, not every combination of distributions, cutpoint optimization methods, and optimized metrics can be simulated and a lot of the literature is concerned specifically with cutpoints and confidence intervals for the Youden index. This complicates sample size planning for studies that estimate optimal cutpoints. As a practical tool, we introduce a web-application that allows for running simulations of width and coverage of confidence intervals using the percentile bootstrap with various distributions and cutpoint optimization methods.

Statistical assessment of the prognostic and the predictive value of biomarkers-A biomarker assessment framework with applications to traumatic brain injury biomarker studies

Studies that investigate the performance of prognostic and predictive biomarkers are commonplace in medicine. Evaluating the performance of biomarkers is challenging in traumatic brain injury (TBI) and other conditions when both the time factor (i.e. time from injury to biomarker measurement) and different levels or doses of treatments are in play. Such factors need to be accounted for when assessing the biomarker’s performance in relation to a clinical outcome. The Hyperbaric Oxygen in Brain Injury Treatment (HOBIT) trial, a phase II randomized control clinical trial seeks to determine the dose of hyperbaric oxygen therapy (HBOT) for treating severe TBI that has the highest likelihood of demonstrating efficacy in a phase III trial. Hyperbaric Oxygen in Brain Injury Treatment will study up to 200 participants with severe TBI. This paper discusses the statistical approaches to assess the prognostic and predictive performance of the biomarkers studied in this trial, where prognosis refers to the association between a biomarker and the clinical outcome while the predictiveness refers to the ability of the biomarker to identify patient subgroups that benefit from therapy. Analyses based on initial biomarker levels accounting for different levels of HBOT and other baseline clinical characteristics, and analyses of longitudinal changes in biomarker levels are discussed from a statistical point of view. Methods for combining biomarkers that are of complementary nature are also considered and the relevant algorithms are illustrated in detail along with an extensive simulation study that assesses the performance of the statistical methods. Even though the discussed approaches are motivated by the HOBIT trial, their applications are broader. They can be applied in studies assessing the predictiveness and prognostic ability of biomarkers in relation to a well-defined therapeutic intervention and clinical outcome.

Comparison of diagnostic validity of two autism rating scales for suspected autism in a large Chinese sample

BACKGROUND

Autism is the most common clinical developmental disorder in children. The childhood autism rating scale (CARS) and autistic autism behavior checklist (ABC) are the most commonly used assessment scales for diagnosing autism. However, the diagnostic validations and the corresponding cutoffs for CARS and ABC in individuals with suspected autism spectrum disorder (ASD) remain unclear. Furthermore, for suspected ASD in China, it remains unclear whether CARS is a better diagnostic tool than ABC. Also unclear is whether the current cutoff points for ABC and CARS are suitable for the accurate diagnosis of ASD.

AIM

To investigate the diagnostic validity of CARS and ABC based on a large Chinese sample.

METHODS

A total of 591 outpatient children from the ASD Unit at Beijing Children’s Hospital between June and November 2019 were identified. First, the Clancy autism behavior scale (CABS) was used to screen out suspected autism from these children. Then, each suspected ASD was evaluated by CARS and ABC. Receiver operating characteristic (ROC) curve analysis was used to compare diagnostic validations. We also calculated the area under the curve (AUC) for both CARS and ABC.

RESULTS

We found that the Cronbach alpha coefficients of CARS and ABC were 0.772 and 0.426, respectively. Therefore, the reliability of the CARS was higher than that of the ABC. In addition, we found that the correlation between CARS and CABS was 0.732. Next, we performed ROC curve analysis for CARS and ABC, which yielded AUC values of 0.846 and 0.768, respectively. The cutoff value, which is associated with the maximum Youden index, is usually applied as a decision threshold. We found that the cutoff values of CARS and ABC were 34 and 67, respectively.

CONCLUSION

This result indicated that CARS is superior to ABC in the Chinese population with suspected ASD.

Joint inference about the AUC and Youden index for paired biomarkers

It is common to compare biomarkers' diagnostic or prognostic performance using some summary ROC measures such as the area under the ROC curve (AUC) or the Youden index. We propose to compare two paired biomarkers using both the AUC and the Youden index since the two indices describe different aspects of the ROC curve. This comparison can be made by estimating the joint confidence region (an elliptical area) of the differences of the paired AUCs and the Youden indices. Furthermore, for deciding if one marker is better than the other in terms of both the A U C and the Youden index (J), we can test H 0 : A U C a ≤ A U C b or J a ≤ J b against H a : A U C a > A U C b and J a > J b using the paired differences. The construction of such a joint hypothesis is an example of the multivariate order-restricted hypotheses. For such a hypothesis, we propose and compare three testing procedures: (1) the intersection-union test ( I U T ); (2) the conditional test; and (3) the joint test. The performance of the proposed inference methods was evaluated and compared through simulations. The simulation results demonstrate that the proposed joint confidence region maintains the desired confidence level, and all three tests maintain the type I error under the null. Furthermore, among the three proposed testing methods, the conditional test is the preferred approach with markedly larger power consistently than the other two competing methods.

Inference on the overlap coefficient: The binormal approach and alternatives

The overlap coefficient (OVL) measures the similarity between two distributions through the overlapping area of their distribution functions. Given its intuitive description and ease of visual representation by the straightforward depiction of the amount of overlap between the two corresponding histograms based on samples of measurements from each one of the two distributions, the development of accurate methods for confidence interval construction can be useful for applied researchers. The overlap coefficient has received scant attention in the literature since it lacks readily available software for its implementation, while inferential procedures that can cover the whole range of distributional scenarios for the two underlying distributions are missing. Such methods, both parametric and non-parametric are developed in this article, while R-code is provided for their implementation. Parametric approaches based on the binormal model show better performance and are appropriate for use in a wide range of distributional scenarios. Methods are assessed through a large simulation study and are illustrated using a dataset from a study on human immunodeficiency virus-related cognitive function assessment.

Statistical inference for the difference between two maximized Youden indices obtained from correlated biomarkers

Currently, there is global interest in deriving new promising cancer biomarkers that could complement or substitute the conventional ones. Clinical decisions can often be based on the cutoff that corresponds to the maximized Youden index when maximum accuracy drives decisions. When more than one classification criteria are measured within the same individuals, correlated measurements arise. In this work, we propose hypothesis tests and confidence intervals for the comparison of two correlated receiver operating characteristic (ROC) curves in terms of their corresponding maximized Youden indices. We explore delta-based techniques under parametric assumptions, or power transformations. Nonparametric kernel-based methods are also examined. We evaluate our approaches through simulations and illustrate them using data from a metabolomic study referring to the detection of pancreatic cancer.

Estimation and construction of confidence intervals for biomarker cutoff-points under the shortest Euclidean distance from the ROC surface to the perfection corner

Pancreatic ductal adenocarcinoma (PDAC) is an aggressive type of cancer with a 5-year survival rate of less than 5%. As in many other diseases, its diagnosis might involve progressive stages. It is common that in biomarker studies referring to PDAC, recruitment involves three groups: healthy individuals, patients that suffer from chronic pancreatitis, and PDAC patients. Early detection and accurate classification of the state of the disease are crucial for patients' successful treatment. ROC analysis is the most popular way to evaluate the performance of a biomarker and the Youden index is commonly employed for cutoff derivation. The so-called generalized Youden index has a drawback in the three-class case of not accommodating the full data set when estimating the optimal cutoffs. In this article, we explore the use of the Euclidean distance of the ROC to the perfection corner for the derivation of cutoffs in trichotomous settings. We construct an inferential framework that involves both parametric and nonparametric techniques. Our methods can accommodate the full information of a given data set and thus provide more accurate estimates in terms of the decision-making cutoffs compared with a Youden-based strategy. We evaluate our approaches through extensive simulations and illustrate them on a PDAC biomarker study.

Confidence interval estimation for sensitivity and difference between two sensitivities at a given specificity under tree ordering

This article considers a setting in diagnostic studies (or biomarker study) which involves a healthy class and a diseased class and the latter consists of several subclasses. The problem of interest is to evaluate the accuracy of a biomarker (or a diagnostic test) measured on a continuous scale correctly identifying healthy subjects from diseased subjects without requiring specification of an ordering in terms of marker values for subclasses relative to each other within the diseased class. Such setting is quite common in practice and it falls in the framework of tree ordering or umbrella ordering. This article explores several parametric and nonparametric approaches for estimating confidence intervals of sensitivity of single biomarker and difference between sensitivities of two correlated biomarkers under tree ordering at a given specificity. The performances of all the methods are evaluated and compared by a comprehensive simulation study. A published microarray data set is analyzed using the proposed methods.

Systemic Inflammation Response Index Is a Predictor of Poor Survival in Locally Advanced Nasopharyngeal Carcinoma: A Propensity Score Matching Study

Introduction

Nasopharyngeal carcinoma (NPC) is a common malignancy in China and known prognostic factors are limited. In this study, neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), systemic immune inflammation index (SII), and systemic inflammation response index (SIRI) were evaluated as prognostic factors in locally advanced NPC patients.

Materials and Methods

NPC patients who received curative radiation or chemoradiation between January 2012 and December 2015 at the Second Xiangya Hospital were retrospectively reviewed, and a total of 516 patients were shortlisted. After propensity score matching (PSM), 417 patients were eventually enrolled. Laboratory and clinical data were collected from the patients' records. Receiver operating characteristic curve analysis was used to determine the optimal cut-off value. Survival curves were analyzed using the Kaplan-Meier method. The Cox proportional hazard model was used to identify prognostic variables.

Results

After PSM, all basic characteristics between patients in the high SIRI group and low SIRI group were balanced except for sex (p=0.001) and clinical stage (p=0.036). Univariate analysis showed that NLR (p=0.001), PLR (p=0.008), SII (p=0.001), and SIRI (p<0.001) were prognostic factors for progression-free survival (PFS) and overall survival (OS). However, further multivariate Cox regression analysis showed that only SIRI was an independent predictor of PFS and OS (hazard ratio (HR):2.83; 95% confidence interval (CI): 1.561-5.131; p=0.001, HR: 5.19; 95% CI: 2.588-10.406; p<0.001), respectively.

Conclusion

Our findings indicate that SIRI might be a promising predictive indicator of locally advanced NPC patients.

Impact of radiation source activity on short- and long-term outcomes of cervical carcinoma patients treated with high-dose-rate brachytherapy: A retrospective cohort study

OBJECTIVE

High-dose-rate (HDR) afterloading brachytherapy using Iridium-192 source involves large radiation activity varieties due to fast decay. It was unknown but clinically desirable to evaluate its impacts on patient outcomes to support more informed decisions.

METHODS

Data of 510 cervical carcinoma (CC) patients were retrospectively included. High-radioactive (HR) and low-radioactive (LR) groups were statistically defined per patient-specific average mean-dose-rate (MDR) of all fractions. The cutoffs were calculated using R-3.6.1 packages based on significance of correlation with binary outcome or survival time. Categorized 1-month and 3-month follow-up results were analyzed as short-term outcomes. Long-term outcomes were evaluated using local recurrence-free survival (LRFS) and metastatic recurrence-free survival (MRFS). Propensity-score-matched (PSM) pairs were generated to reduce bias.

RESULTS

The median follow-up time was 47.1 months (interquartile range: 33.9 months-66.4 months), involving MDR varieties of up to 9 folds ranging from 6059.99 cGy/h to 54013.66 cGy/h due to 17 source replacements at intervals ranging from 93 days-199 days. Both short-term (1-month: p = 0.22; 3-month: p = 0.79) and long-term (LRFS: p = 0.10; MRFS: p = 0.46) outcomes showed no significant difference between HR and LR. Subgroup analysis displayed significantly better results in LR for stage I-II (3-month, p = 0.02) and stage II (LRFS, p = 0.04) patients. Both LRFS and MRFS of LR were significantly non-inferior to HR (p ≤ 0.02).

CONCLUSIONS

LR is clinically non-inferior or partially superior to HR for CC treatment using HDR, which dispels concerns of potentially undermined patient outcomes when source replacement is delayed.

Color Doppler Ultrasound Improves Machine Learning Diagnosis of Breast Cancer

Color Doppler is used in the clinic for visually assessing the vascularity of breast masses on ultrasound, to aid in determining the likelihood of malignancy. In this study, quantitative color Doppler radiomics features were algorithmically extracted from breast sonograms for machine learning, producing a diagnostic model for breast cancer with higher performance than models based on grayscale and clinical category from the Breast Imaging Reporting and Data System for ultrasound (BI-RADS_US). Ultrasound images of 159 solid masses were analyzed. Algorithms extracted nine grayscale features and two color Doppler features. These features, along with patient age and BI-RADS_US category, were used to train an AdaBoost ensemble classifier. Though training on computer-extracted grayscale features and color Doppler features each significantly increased performance over that of models trained on clinical features, as measured by the area under the receiver operating characteristic (ROC) curve, training on both color Doppler and grayscale further increased the ROC area, from 0.925 ± 0.022 to 0.958 ± 0.013. Pruning low-confidence cases at 20% improved this to 0.986 ± 0.007 with 100% sensitivity, whereas 64% of the cases had to be pruned to reach this performance without color Doppler. Fewer borderline diagnoses and higher ROC performance were both achieved for diagnostic models of breast cancer on ultrasound by machine learning on color Doppler features.

Nonparametric confidence regions for the symmetry point-based optimal cutpoint and associated sensitivity of a continuous-scale diagnostic test

In medical research, diagnostic tests with continuous values are widely employed to attempt to distinguish between diseased and non-diseased subjects. The diagnostic accuracy of a test (or a biomarker) can be assessed by using the receiver operating characteristic (ROC) curve of the test. To summarize the ROC curve and primarily to determine an "optimal" threshold for test results to use in practice, several approaches may be considered, such as those based on the Youden index, on the so-called close-to-(0,1) point, on the concordance probability and on the symmetry point. In this paper, we focus on the symmetry point-based approach, that simultaneously controls the probabilities of the two types of correct classifications (healthy as healthy and diseased as diseased), and show how to get joint nonparametric confidence regions for the corresponding optimal cutpoint and the associated sensitivity (= specificity) value. Extensive simulation experiments are conducted to evaluate the finite sample performances of the proposed method. Real datasets are also used to illustrate its application.