Summary ROC curve based on a weighted Youden index for selecting an optimal cutpoint in meta-analysis of diagnostic accuracy

Assessment of fluid responsiveness using pulse pressure variation, stroke volume variation, plethysmographic variability index, central venous pressure, and inferior vena cava variation in patients undergoing mechanical ventilation: a systematic review and meta-analysis

IMPORTANCE

Maneuvers assessing fluid responsiveness before an intravascular volume expansion may limit useless fluid administration, which in turn may improve outcomes.

OBJECTIVE

To describe maneuvers for assessing fluid responsiveness in mechanically ventilated patients.

REGISTRATION

The protocol was registered at PROSPERO: CRD42019146781.

INFORMATION SOURCES AND SEARCH

PubMed, EMBASE, CINAHL, SCOPUS, and Web of Science were search from inception to 08/08/2023.

STUDY SELECTION AND DATA COLLECTION

Prospective and intervention studies were selected.

STATISTICAL ANALYSIS

Data for each maneuver were reported individually and data from the five most employed maneuvers were aggregated. A traditional and a Bayesian meta-analysis approach were performed.

RESULTS

A total of 69 studies, encompassing 3185 fluid challenges and 2711 patients were analyzed. The prevalence of fluid responsiveness was 49.9%. Pulse pressure variation (PPV) was studied in 40 studies, mean threshold with 95% confidence intervals (95% CI) = 11.5 (10.5-12.4)%, and area under the receiver operating characteristics curve (AUC) with 95% CI was 0.87 (0.84-0.90). Stroke volume variation (SVV) was studied in 24 studies, mean threshold with 95% CI = 12.1 (10.9-13.3)%, and AUC with 95% CI was 0.87 (0.84-0.91). The plethysmographic variability index (PVI) was studied in 17 studies, mean threshold = 13.8 (12.3-15.3)%, and AUC was 0.88 (0.82-0.94). Central venous pressure (CVP) was studied in 12 studies, mean threshold with 95% CI = 9.0 (7.7-10.1) mmHg, and AUC with 95% CI was 0.77 (0.69-0.87). Inferior vena cava variation (∆IVC) was studied in 8 studies, mean threshold = 15.4 (13.3-17.6)%, and AUC with 95% CI was 0.83 (0.78-0.89).

CONCLUSIONS

Fluid responsiveness can be reliably assessed in adult patients under mechanical ventilation. Among the five maneuvers compared in predicting fluid responsiveness, PPV, SVV, and PVI were superior to CVP and ∆IVC. However, there is no data supporting any of the above mentioned as being the best maneuver. Additionally, other well-established tests, such as the passive leg raising test, end-expiratory occlusion test, and tidal volume challenge, are also reliable.

Cytoplasmic Lipid Droplets Predict Worse Prognosis in Diffuse Large B-Cell Lymphoma: Next-Generation Sequencing Deciphering Lipogenic Genes

Burkitt lymphoma is characterized by high cell turnover and numerous cytoplasmic vacuoles that are demonstrated to be lipid droplets (LDs) decorated by adipophilin. By contrast, cytoplasmic vacuoles are variably observed in diffuse large B-cell lymphoma (DLBCL) and less well characterized. In this study, we first validated in DLBCL that cytoplasmic vacuoles are indeed LDs by Oil-red-O stain, Bodipy fluorescent stain, and electron microscopy. Second, in a cohort of DLBCL patients (n=52) we showed that LDs in effusional lymphoma cells were associated with a poorer prognosis (P=0.029, log-rank test) and higher International Prognostic Index (IPI) score (94% vs. 66%, P=0.026) than those without. Moreover, using adipophilin as a surrogate marker for LDs, we found in another cohort of biopsy specimen (n=85) that expression of adipophilin by lymphoma cells predicted a poorer prognosis (P=0.007, log-rank test) and higher IPI score (63% vs. 30%, P=0.005). In addition, whole exome sequencing of effusional DLBCL cells showed LD-positive DLBCL shared genetic features with the MCD (MYD88 and CD79B mutations) subtype and highlighted OSBPL10 and CUBN as the most frequently mutated genes involved in lipogenesis. Whole transcriptome analysis by comparing effusional DLBCL cells with versus without LDs showed upregulation of EHHADH, SLC1A1, CD96, INPP4B, and RNF183 relevant for lymphoma lipogenesis and upregulation of epithelial-mesenchymal transition and KRAS signaling pathways. Higher expression of EHHADH and CD96 were validated in LD-positive clinical samples and LD-rich cell lines than LD-poor cells along with the known lipogenic gene, FASN. Our findings highlight the roles of LDs and adipophilin expression in DLBCL, suggest that these markers may predict prognosis and show that lipogenic genes may be potential therapeutic targets.

Establishment of the Patient Acceptable Symptom State (PASS) for the Achilles Tendon Total Rupture Score in a Swedish Population

Background

As the use of patient-reported outcome measures (PROMs) is increasing in orthopaedic research, there is also a growing need for a standardized interpretation of these scores, such as the Patient Acceptable Symptom State (PASS), defined as the value beyond which patients consider themselves well. The Achilles tendon Total Rupture Score (ATRS) is the only PROM specific for Achilles tendon ruptures.

Purpose

To establish the PASS for the ATRS in a Swedish population.

Study Design

Cross-sectional study; Level of evidence, 3.

Methods

Patients treated for an acute Achilles tendon rupture at a single institution in Sweden (injured between July 1, 2018, and December 31, 2020) were asked to participate in this study. The patients completed a questionnaire consisting of the ATRS and an anchor question: "How satisfied are you with the result of your treatment?" Receiver operating characteristic curve analysis was performed to calculate the PASS threshold for a positive response to the anchor question.

Results

Of 516 eligible patients, 316 (61%) were included. The time from injury to completion of the questionnaire ranged from 12 to 27 months. The PASS threshold for the ATRS was found to be 75. The median ATRS of all patients was 80; 66% of patients reached an ATRS ≥75. Overall, 79% of patients were satisfied with the results of their treatment.

Conclusion

The estimated PASS for the ATRS was 75 in the general Swedish population at 12 to 27 months after an acute Achilles tendon rupture.

An evaluation of treatment response and remission definitions in adult obsessive-compulsive disorder: A systematic review and individual-patient data meta-analysis

INTRODUCTION

Expert consensus operationalized treatment response and remission in obsessive-compulsive disorder (OCD) as a Yale-Brown Obsessive-Compulsive Scale (Y-BOCS) reduction ≥35% and score ≤12 with ≤2 on Clinical Global Impressions Improvement (CGI-I) and Severity (CGI-S) scales, respectively. However, there has been scant empirical evidence supporting these definitions.

METHODS

We conducted a systematic review and an individual participant data meta-analysis of randomized-controlled trials (RCTs) in adults with OCD to determine optimal Y-BOCS thresholds for response and remission. We estimated pooled sensitivity/specificity for each percent reduction threshold (response) or posttreatment score (remission) to determine response and remission defined by a CGI-I and CGI-S ≤ 2, respectively.

RESULTS

Individual participant data from 25 of 94 eligible RCTs (1235 participants) were included. The optimal threshold for response was ≥30% Y-BOCS reduction and for remission was ≤15 posttreatment Y-BOCS. However, differences in sensitivity and specificity between the optimal and nearby thresholds for response and remission were small with some uncertainty demonstrated by the confidence ellipses.

CONCLUSION

While the empirically derived Y-BOCS thresholds in our meta-analysis differ from expert consensus, given the predominance of data from more recent trials of OCD, which involved more refractory participants and novel treatment modalities as opposed to first-line therapies, we recommend the continued use of the consensus definitions.

Characterizing the Increase in Artificial Intelligence Content Detection in Oncology Scientific Abstracts From 2021 to 2023

PURPOSE

Artificial intelligence (AI) models can generate scientific abstracts that are difficult to distinguish from the work of human authors. The use of AI in scientific writing and performance of AI detection tools are poorly characterized.

METHODS

We extracted text from published scientific abstracts from the ASCO 2021-2023 Annual Meetings. Likelihood of AI content was evaluated by three detectors: GPTZero, Originality.ai, and Sapling. Optimal thresholds for AI content detection were selected using 100 abstracts from before 2020 as negative controls, and 100 produced by OpenAI's GPT-3 and GPT-4 models as positive controls. Logistic regression was used to evaluate the association of predicted AI content with submission year and abstract characteristics, and adjusted odds ratios (aORs) were computed.

RESULTS

Fifteen thousand five hundred and fifty-three abstracts met inclusion criteria. Across detectors, abstracts submitted in 2023 were significantly more likely to contain AI content than those in 2021 (aOR range from 1.79 with Originality to 2.37 with Sapling). Online-only publication and lack of clinical trial number were consistently associated with AI content. With optimal thresholds, 99.5%, 96%, and 97% of GPT-3/4-generated abstracts were identified by GPTZero, Originality, and Sapling respectively, and no sampled abstracts from before 2020 were classified as AI generated by the GPTZero and Originality detectors. Correlation between detectors was low to moderate, with Spearman correlation coefficient ranging from 0.14 for Originality and Sapling to 0.47 for Sapling and GPTZero.

CONCLUSION

There is an increasing signal of AI content in ASCO abstracts, coinciding with the growing popularity of generative AI models.

Korean listeners' identification and discrimination of lengthened /s/ as prolongations

This study aimed to examine Korean listeners' judgement of sound duration as prolongation and the mode of perceiving prolongation, whether discrete or continuous. A total of 75 Korean undergraduate students listened to the Korean segment /s/, each of which was lengthened by 0-380 ms (ranging from the original 205 to 585 ms) in 20-ms increments. Then, the participants were asked to complete two different primary tasks: determine whether the sound was normal (0) or abnormal (1) and rate each version of the sound based on a rating of 1 to 100 (the closer to 100, the less fluent). The minimum duration for the Korean sound to be perceived as abnormally prolonged was calculated by analysing the receiver operating characteristic (ROC) curves using Youden's index. To examine whether listeners perceived durational variations for the fricative segment discretely or continuously, a curve was estimated using the best fitting regression model for the observed data with the highest adjusted R-squared value. The minimum duration identified as abnormal prolongation for the Korean lenis fricative /s/ was 375 ms, corresponding to 182.9% of the original, unaltered sound's length. The mode of perceiving durational variations for the segment was continuous (or gradient) rather than discrete. No gender difference was found in the minimum durational threshold and the mode of perceiving prolongation. The findings of this study were further discussed in relation to the existing body of research, and some clinical implications for the assessment of stuttering were also presented.

Transfer learning enhanced graph neural network for aldehyde oxidase metabolism prediction and its experimental application

Aldehyde oxidase (AOX) is a molybdoenzyme that is primarily expressed in the liver and is involved in the metabolism of drugs and other xenobiotics. AOX-mediated metabolism can result in unexpected outcomes, such as the production of toxic metabolites and high metabolic clearance, which can lead to the clinical failure of novel therapeutic agents. Computational models can assist medicinal chemists in rapidly evaluating the AOX metabolic risk of compounds during the early phases of drug discovery and provide valuable clues for manipulating AOX-mediated metabolism liability. In this study, we developed a novel graph neural network called AOMP for predicting AOX-mediated metabolism. AOMP integrated the tasks of metabolic substrate/non-substrate classification and metabolic site prediction, while utilizing transfer learning from 13C nuclear magnetic resonance data to enhance its performance on both tasks. AOMP significantly outperformed the benchmark methods in both cross-validation and external testing. Using AOMP, we systematically assessed the AOX-mediated metabolism of common fragments in kinase inhibitors and successfully identified four new scaffolds with AOX metabolism liability, which were validated through in vitro experiments. Furthermore, for the convenience of the community, we established the first online service for AOX metabolism prediction based on AOMP, which is freely available at https://aomp.alphama.com.cn.

Diagnostic cut-off values for the urinary corticoid:creatinine ratio for the diagnosis of canine Cushing's syndrome using an automated chemiluminescent assay

BACKGROUND

Cushing's syndrome is one of the most common endocrinopathies in dogs. The preferred screening test for spontaneous Cushing's syndrome is the low-dose dexamethasone suppression test (LDDST). The diagnostic value of urinary cortisol:creatinine ratios (UCCR) is questionable.

OBJECTIVES

The aim of this study was to determine diagnostic cut-off values for UCCR testing in comparison with LDDST as a clinical reference standard and to calculate the sensitivity and specificity.

METHODS

Data from 2018 to 2020 were obtained retrospectively from a commercial laboratory. Both LDDST and UCCR were measured by automated chemiluminescent immunoassay (CLIA). The maximum interval between both tests was 14 days. The optimal cut-off value for UCCR testing was calculated by the Youden index. The sensitivity and specificity of these cut-off values for the UCCR test and LDDST were assessed by Bayesian latent class models (BLCMs).

RESULTS

This study included 324 dogs with both UCCR test and LDDST results. The optimal UCCR cut-off value, calculated by the Youden index, was 47.4 × 10-6 . Any UCCR <40 × 10-6 was interpreted as a negative result, 40-60 × 10-6 as values in a gray zone, and >60 × 10-6 as positive. Using the cut-off of 60 × 10-6 , BLCM showed 91% (LDDST) and 86% (UCCR test) sensitivity and a specificity of 54% (LDDST) and 63% (UCCR test).

CONCLUSIONS

Considering an 86% sensitivity and a 63% specificity, UCCR testing may be considered a first-line investigation to rule out Cushing's syndrome using CLIA analysis. Urine samples can be collected noninvasively at home by the owner, reducing the potential impact of stress.

Accuracy of pulse CO-oximetry to evaluate blood carboxyhemoglobin level: a systematic review and meta-analysis of diagnostic test accuracy studies

Carbon monoxide (CO) poisoning is one of the most common causes of poisoning death and its diagnosis requires an elevated carboxyhemoglobin (COHb) level. Noninvasive CO saturation by pulse oximetry (SpCO) has been available since 2005 and has the advantage of being portable and easy to use, but its accuracy in determining blood COHb level is controversial. To evaluate the accuracy of SpCO (index test) to estimate COHb (reference test). Systematic review and meta-analysis of diagnostic test accuracy (DTA) studies. Four electronic databases were searched (Medline, Embase, Cochrane Central Register of Controlled Trials, and OpenGrey) on 2 August 2022. All studies of all designs published since the 2000s evaluating the accuracy and reliability of SpCO measurement compared to blood COHb levels in human volunteers or ill patients, including children, were included. The primary outcome was to assess the diagnostic accuracy of SpCO for estimating COHb by blood sampling by modeling receiver operating characteristic (ROC) curves and calculating sensitivity and specificity (primary measures). The secondary measures were to calculate the limits of agreement (LOA) and the mean bias. This systematic review was conducted according to the Preferred Reporting Items for a Systematic Review and Meta-analysis-DTA 2018 guidelines and has been registered on International Prospective Register of Systematic Reviews (PROSPERO, CRD42020177940). The risk of bias was evaluated using the Quality Assessment of Diagnostic Accuracy Studies-2 tool. Twenty-one studies were eligible for the systematic review; 11 could be included for the quantitative analysis of the primary measures and 18 for the secondary measures. No publication bias was found. The area under the summary ROC curve was equal to 86%. The mean sensitivity and specificity were 0.77, 95% confidence interval (CI, 0.66-0.85) and 0.83, 95% CI (0.74-0.89), respectively (2089 subjects and 3381 observations). The mean bias was 0.75% and the LOA was -7.08% to 8.57%, 95% CI (-8.89 to 10.38) (2794 subjects and 4646 observations). Noninvasive measurement of COHb (SpCO) using current pulse CO oximeters do not seem to be highly accurate to estimate blood COHb (moderate sensitivity and specificity, large LOA). They should probably not be used to confirm (rule-in) or exclude (rule-out) CO poisoning with certainty.

Derivation and validation of a risk prediction score for nonsteroidal anti-inflammatory drug-related serious gastrointestinal complications in the elderly

AIMS

Few studies have quantified the impact of risk factors on GI complications in elderly nonsteroidal anti-inflammatory drug (NSAID) users. This study aimed to develop and validate a risk prediction score for severe GI complications to identify high-risk elderly patients using NSAID.

METHODS

We used the following two Korean claims datasets: customized data with an enrolment period 2016-2017 for model development, and the sample data in 2019 for external validation. We conducted a nested case-control study for model development and validation. NSAID users were identified as the elderly (≥65 years) who received NSAIDs for more than 30 days. Serious GI complications were defined as hospitalizations or emergency department visits, with a main diagnosis of GI bleeding or perforation. We applied the logistic least absolute shrinkage and selection operator (LASSO) regression model for variable selection and model fitting.

RESULTS

We identified 8176 cases and 81 760 controls with a 1:10 matched follow-up period in the derivation cohort. In the external validation cohort, we identified 372 cases from 254 551 patients. The risk predictors were high-dose NSAIDs, nonselective NSAID, complicated GI ulcer history, male sex, concomitant gastroprotective agents, relevant co-medications, severe renal disease and cirrhosis. Area under the receiver operating characteristic curve was 0.79 (95% confidence interval, 0.77-0.81) in the external validation dataset.

CONCLUSIONS

The prediction model may be a useful tool for reducing the risk of serious GI complications by identifying high-risk elderly patients.

Diagnostic Accuracy of Fecal Calprotectin for Predicting Relapse in Inflammatory Bowel Disease: A Meta-Analysis

Fecal calprotectin (FC) levels correlate with the disease activity of inflammatory bowel diseases (IBD); however, the utility of FC in predicting IBD relapse remains to be determined. We aim to evaluate the efficacy of fecal calprotectin in predicting the relapse of inflammatory bowel disease. We searched Pubmed (MEDLINE), Embase, Web of Science, and the Cochrane library databases up to 7 July 2021. Our study estimated the pooled sensitivity and specificity, summary receiver operating characteristic (SROC) curve, and the optimal cut-off value for predicting IBD relapse using a multiple threshold model. A total of 24 prospective studies were included in the meta-analysis. The optimal FC cut-off value was 152 μg/g. The pooled sensitivity and specificity of FC was 0.720 (0.528 to 0.856) and 0.740 (0.618 to 0.834), respectively. FC is a useful, non-invasive, and inexpensive biomarker for the early prediction of IBD relapse. An FC value of 152 μg/g is an ideal threshold to identify patients with a high relapse probability.

Prediction of malignant lymph nodes in NSCLC by machine-learning classifiers using EBUS-TBNA and PET/CT

Accurate determination of lymph-node (LN) metastases is a prerequisite for high precision radiotherapy. The primary aim is to characterise the performance of PET/CT-based machine-learning classifiers to predict LN-involvement by endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) in stage-III NSCLC. Prediction models for LN-positivity based on [¹⁸F]FDG-PET/CT features were built using logistic regression and machine-learning models random forest (RF) and multilayer perceptron neural network (MLP) for stage-III NSCLC before radiochemotherapy. A total of 675 LN-stations were sampled in 180 patients. The logistic and RF models identified SUV_max, the short-axis LN-diameter and the echelon of the considered LN among the most important parameters for EBUS-positivity. Adjusting the sensitivity of machine-learning classifiers to that of the expert-rater of 94.5%, MLP (P = 0.0061) and RF models (P = 0.038) showed lower misclassification rates (MCR) than the standard-report, weighting false positives and false negatives equally. Increasing the sensitivity of classifiers from 94.5 to 99.3% resulted in increase of MCR from 13.3/14.5 to 29.8/34.2% for MLP/RF, respectively. PET/CT-based machine-learning classifiers can achieve a high sensitivity (94.5%) to detect EBUS-positive LNs at a low misclassification rate. As the specificity decreases rapidly above that level, a combined test of a PET/CT-based MLP/RF classifier and EBUS-TBNA is recommended for radiation target volume definition.

Development and Validation of a Nomogram to Predict Hemiplegic Shoulder Pain in Patients With Stroke: A Retrospective Cohort Study

•

We developed a nomogram for the individualized prediction of hemiplegic shoulder pain in patients with stroke during inpatient rehabilitation.

•

The nomogram demonstrated good performance in terms of discrimination, calibration, and clinical utility.

•

The variables incorporated in the nomogram were readily available from clinical characteristics, making the nomogram a simple and practical tool for clinicians.

•

A web application of the nomogram was developed to facilitate its use for clinicians.

Objective

The development and validation of a nomogram for the individualized prediction of hemiplegic shoulder pain (HSP) during the inpatient rehabilitation of patients with stroke.

Design

Retrospective cohort study.

Setting

The rehabilitation department at a tertiary hospital.

Participants

A total of 376 patients (N=376) with stroke admitted to inpatient rehabilitation from January 2018 to April 2021 were included in this study.

Interventions

Not applicable.

Main Outcome Measures

The outcome measure was shoulder pain on the patients’ hemiplegic side occurring at rest or with movement during hospitalization.

Results

Among the 376 patients with stroke, 113 (30.05%) developed HSP. Five independent predictors were included in the nomogram: subluxation, Brunnstrom stage, hand edema, spasticity, and sensory disturbance. The nomogram was a good predictor, with a C-index of 0.85 (95% confidence interval, 0.81-0.89) and corrected C-index of 0.84. The Homer-Lemeshow test (χ²=13.854, P=.086) and calibration plot suggested good calibration ability of the nomogram. The optimal cutoff value for the predicted probability of HSP was 0.30 (sensitivity, 0.73; specificity, 0.83). Moreover, the decision curve analysis revealed that the nomogram would add net clinical benefits if the threshold possibility of HSP risk was from 5%-88%.

Conclusions

Our nomogram could accurately predict HSP, which may help clinicians accurately quantify the HSP risk in individuals and implement early interventions.

Circulating Tumor DNA in Adults With Glioma: A Systematic Review and Meta-Analysis of Biomarker Performance

BACKGROUND

Circulating tumor DNA (ctDNA) has emerged as a promising noninvasive biomarker to capture tumor genetics in patients with brain tumors. Research into its clinical utility, however, has not been standardized because the sensitivity and specificity of ctDNA remain undefined.

OBJECTIVE

To (1) review the primary literature about ctDNA in adults with glioma to compare the sensitivity and specificity of ctDNA in the cerebrospinal fluid vs the plasma and (2) to evaluate the effect of tumor grade on detection of ctDNA.

METHODS

PRISMA-guided systematic review and meta-analysis was performed using published studies that assessed ctDNA in either plasma or cerebrospinal fluid among adult patients with confirmed glioma. Summary receiver operating characteristic curves were generated using the Rücker-Schumacher method, and area under the curve (AUC) was calculated.

RESULTS

Meta-analysis revealed improved biomarker performance for CSF (AUC = 0.947) vs plasma (AUC = 0.741) ctDNA, although this did not reach statistical significance (P = .141). Qualitative analysis revealed greater sensitivities among single-allele PCR and small, targeted next-generation sequencing panels compared with broader panels. It additionally demonstrated higher sensitivity of ctDNA detection in high-grade vs low-grade gliomas, although these analyses were limited by a lack of specificity reporting in many studies.

CONCLUSION

ctDNA seems to be a highly sensitive and specific noninvasive biomarker among adults with gliomas. To maximize its performance, CSF should be studied with targeted genetic analysis platforms, particularly in high-grade gliomas. Further studies on ctDNA are needed to define its clinical utility in diagnosis, prognostication, glioblastoma pseudoprogression, and other scenarios wherein neoadjuvant therapies may be considered.

Balanced Hybrid Nutrient Density Score Compared to Nutri-Score and Health Star Rating Using Receiver Operating Characteristic Curve Analyses

Background

Nutrient profiling (NP) models that are used to assess the nutrient density of foods can be based on a combination of key nutrients and desirable food groups.

Objective

To compare the diagnostic accuracy of a new balanced hybrid nutrient density score (bHNDS) to Nutri-Score and Health Star Rating (HSR) front-of-pack systems using receiver operating characteristic (ROC) curve analyses. The diet-level bHNDS was first validated against Healthy Eating Index (HEI-2015) using data from the 2017-18 National Health and Nutrition Examination Survey (2017-18 NHANES). Food-level bHNDS values were then compared to both the Nutri-Score and HSR using ROC curve analyses.

Results

The bHNDS was based on 6 nutrients to encourage (protein, fiber, calcium, iron, potassium, and vitamin D); 5 food groups to encourage (whole grains, nuts and seeds, dairy, vegetables, and fruit), and 3 nutrients (saturated fat, added sugar, and sodium) to limit. The algorithm balanced components to encourage against those to limit. Diet-level bHNDS values correlated well with HEI-2015 (r = 0.67; p < 0.001). Food-level correlations with both Nutri-Score (r = 0.60) and with HSR (r = 0.58) were significant (both p < 0.001). ROC estimates of the Area Under the Curve (AUC) showed high agreement between bHNDS values and optimal Nutri-Score and HSR ratings (>0.90 in most cases). ROC analysis identified those bHNDS cut-off points that were predictive of A-grade Nutri-Score or 5-star HSR. Those cut-off points were highly category-specific.

Conclusion

The new bHNDS model showed high agreement with two front-of-pack labeling systems. Cross-model comparisons based on ROC curve analyses are the first step toward harmonization of proliferating NP methods that aim to "diagnose" high nutrient-density foods.

Deep learning radiomics of ultrasonography can predict response to neoadjuvant chemotherapy in breast cancer at an early stage of treatment: a prospective study

OBJECTIVES

Breast cancer (BC) is the most common cancer in women worldwide, and neoadjuvant chemotherapy (NAC) is considered the standard of treatment for most patients with BC. However, response rates to NAC vary among patients, which leads to delays in appropriate treatment and affects the prognosis for patients who ineffectively respond to NAC. This study aimed to investigate the feasibility of deep learning radiomics (DLR) in the prediction of NAC response at an early stage.

METHODS

In total, 168 patients with clinicopathologically confirmed BC were enrolled in this prospective study, from March 2016 to December 2020. All patients completed NAC treatment and underwent ultrasonography (US) at three time points (before NAC, after the second course, and after the fourth course). We developed two DLR models, DLR-2 and DLR-4, for predicting responses after the second and fourth courses of NAC. Furthermore, a novel deep learning radiomics pipeline (DLRP) was proposed for stepwise prediction of response at different time points of NAC administration.

RESULTS

In the validation cohort, DLR-2 achieved an AUC of 0.812 (95% CI: 0.770-0.851) with an NPV of 83.3% (95% CI: 76.5-89.6). DLR-4 achieved an AUC of 0.937 (95% CI: 0.913-0.955) with a specificity of 90.5% (95% CI: 86.3-94.2). Moreover, 19 of 21 non-response patients were successfully identified by DLRP, suggesting that they could benefit from treatment strategy adjustment at an early stage of NAC.

CONCLUSIONS

The proposed DLRP strategy holds promise for effectively predicting NAC response at its early stage for BC patients.

KEY POINTS

• We proposed two novel deep learning radiomics (DLR) models to predict response to neoadjuvant chemotherapy (NAC) in breast cancer (BC) patients based on US images at different NAC time points. • Combining two DLR models, a deep learning radiomics pipeline (DLRP) was proposed for stepwise prediction of response to NAC. • The DLRP may provide BC patients and physicians with an effective and feasible tool to predict response to NAC at an early stage and to determine further personalized treatment options.

Development of Machine Learning Models for Prediction of Osteoporosis from Clinical Health Examination Data

Osteoporosis is treatable but often overlooked in clinical practice. We aimed to construct prediction models with machine learning algorithms to serve as screening tools for osteoporosis in adults over fifty years old. Additionally, we also compared the performance of newly developed models with traditional prediction models. Data were acquired from community-dwelling participants enrolled in health checkup programs at a medical center in Taiwan. A total of 3053 men and 2929 women were included. Models were constructed for men and women separately with artificial neural network (ANN), support vector machine (SVM), random forest (RF), k-nearest neighbor (KNN), and logistic regression (LoR) to predict the presence of osteoporosis. Area under receiver operating characteristic curve (AUROC) was used to compare the performance of the models. We achieved AUROC of 0.837, 0.840, 0.843, 0.821, 0.827 in men, and 0.781, 0.807, 0.811, 0.767, 0.772 in women, for ANN, SVM, RF, KNN, and LoR models, respectively. The ANN, SVM, RF, and LoR models in men, and the ANN, SVM, and RF models in women performed significantly better than the traditional Osteoporosis Self-Assessment Tool for Asians (OSTA) model. We have demonstrated that machine learning algorithms improve the performance of screening for osteoporosis. By incorporating the models in clinical practice, patients could potentially benefit from earlier diagnosis and treatment of osteoporosis.

Joint confidence region estimation on predictive values

For evaluating diagnostic accuracy of inherently continuous diagnostic tests/biomarkers, sensitivity and specificity are well-known measures both of which depend on a diagnostic cut-off, which is usually estimated. Sensitivity (specificity) is the conditional probability of testing positive (negative) given the true disease status. However, a more relevant question is "what is the probability of having (not having) a disease if a test is positive (negative)?". Such post-test probabilities are denoted as positive predictive value (PPV) and negative predictive value (NPV). The PPV and NPV at the same estimated cut-off are correlated, hence it is desirable to make the joint inference on PPV and NPV to account for such correlation. Existing inference methods for PPV and NPV focus on the individual confidence intervals and they were developed under binomial distribution assuming binary instead of continuous test results. Several approaches are proposed to estimate the joint confidence region as well as the individual confidence intervals of PPV and NPV. Simulation results indicate the proposed approaches perform well with satisfactory coverage probabilities for normal and non-normal data and, additionally, outperform existing methods with improved coverage as well as narrower confidence intervals for PPV and NPV. The Alzheimer's Disease Neuroimaging Initiative (ADNI) data set is used to illustrate the proposed approaches and compare them with the existing methods.

Deep learning model for the prediction of microsatellite instability in colorectal cancer: a diagnostic study

BACKGROUND

Detecting microsatellite instability (MSI) in colorectal cancer is crucial for clinical decision making, as it identifies patients with differential treatment response and prognosis. Universal MSI testing is recommended, but many patients remain untested. A critical need exists for broadly accessible, cost-efficient tools to aid patient selection for testing. Here, we investigate the potential of a deep learning-based system for automated MSI prediction directly from haematoxylin and eosin (H&E)-stained whole-slide images (WSIs).

METHODS

Our deep learning model (MSINet) was developed using 100 H&E-stained WSIs (50 with microsatellite stability [MSS] and 50 with MSI) scanned at 40× magnification, each from a patient randomly selected in a class-balanced manner from the pool of 343 patients who underwent primary colorectal cancer resection at Stanford University Medical Center (Stanford, CA, USA; internal dataset) between Jan 1, 2015, and Dec 31, 2017. We internally validated the model on a holdout test set (15 H&E-stained WSIs from 15 patients; seven cases with MSS and eight with MSI) and externally validated the model on 484 H&E-stained WSIs (402 cases with MSS and 77 with MSI; 479 patients) from The Cancer Genome Atlas, containing WSIs scanned at 40× and 20× magnification. Performance was primarily evaluated using the sensitivity, specificity, negative predictive value (NPV), and area under the receiver operating characteristic curve (AUROC). We compared the model's performance with that of five gastrointestinal pathologists on a class-balanced, randomly selected subset of 40× magnification WSIs from the external dataset (20 with MSS and 20 with MSI).

FINDINGS

The MSINet model achieved an AUROC of 0·931 (95% CI 0·771-1·000) on the holdout test set from the internal dataset and 0·779 (0·720-0·838) on the external dataset. On the external dataset, using a sensitivity-weighted operating point, the model achieved an NPV of 93·7% (95% CI 90·3-96·2), sensitivity of 76·0% (64·8-85·1), and specificity of 66·6% (61·8-71·2). On the reader experiment (40 cases), the model achieved an AUROC of 0·865 (95% CI 0·735-0·995). The mean AUROC performance of the five pathologists was 0·605 (95% CI 0·453-0·757).

INTERPRETATION

Our deep learning model exceeded the performance of experienced gastrointestinal pathologists at predicting MSI on H&E-stained WSIs. Within the current universal MSI testing paradigm, such a model might contribute value as an automated screening tool to triage patients for confirmatory testing, potentially reducing the number of tested patients, thereby resulting in substantial test-related labour and cost savings.

FUNDING

Stanford Cancer Institute and Stanford Departments of Pathology and Biomedical Data Science.

A panel of two miRNAs correlated to systolic blood pressure is a good diagnostic indicator for stroke

BACKGROUND

We aimed to develop a diagnostic indicator of stroke based on serum miRNAs correlated to systolic blood pressure.

METHODS

Using miRNA expression profiles in GSE117604 from the Gene Expression Omnibus (GEO), we utilized the WGCNA to identify hub miRNAs correlated to systolic blood pressure (SBP). Differential analysis was applied to highlight hub differentially expressed miRNAs (DE-miRNAs), whereby we built a miRNA-based diagnostic indicator for stroke using bootstrap ranking Least Absolute Shrinkage and Selection Operator (LASSO) regression with 10-fold cross-validation. The classification value of the indicator was validated with receiver operating characteristic (ROC) analysis in both the training set and test set, as well as quantitative real-time PCR (qRT-PCR) for the feature miRNAs. Further, target genes of hub miRNAs and hub DE-miRNAs were retrieved for functional enrichment.

RESULTS

A total of 447 hub miRNAs in the blue modules were significantly correlated with systolic blood pressure (r = 0.32, false discovery rate = 10-6). Target genes predicted with the hub miRNAs were mostly implicated in the Kyoto Encyclopedia of Genes and Genomes (KEGG) terms including mitogen-activated protein kinase (MAPK) pathway, senescence, and TGF-β signaling pathway. The diagnostic indicator with miR-4420 and miR-6793-5p showed remarkable performance in the training set (area under curve [AUC]= 0.953), as well as in the test set (AUC = 0.894). Results of qRT-PCR validated the diagnostic value of the two miRNAs embedded in the proposed indicator.

CONCLUSIONS

We developed a panel of two miRNAs, which is a good diagnostic indicator for stroke. These results require further investigation.

Breath biopsy of breast cancer using sensor array signals and machine learning analysis

Breast cancer causes metabolic alteration, and volatile metabolites in the breath of patients may be used to diagnose breast cancer. The objective of this study was to develop a new breath test for breast cancer by analyzing volatile metabolites in the exhaled breath. We collected alveolar air from breast cancer patients and non-cancer controls and analyzed the volatile metabolites with an electronic nose composed of 32 carbon nanotubes sensors. We used machine learning techniques to build prediction models for breast cancer and its molecular phenotyping. Between July 2016 and June 2018, we enrolled a total of 899 subjects. Using the random forest model, the prediction accuracy of breast cancer in the test set was 91% (95% CI: 0.85–0.95), sensitivity was 86%, specificity was 97%, positive predictive value was 97%, negative predictive value was 97%, the area under the receiver operating curve was 0.99 (95% CI: 0.99–1.00), and the kappa value was 0.83. The leave-one-out cross-validated discrimination accuracy and reliability of molecular phenotyping of breast cancer were 88.5 ± 12.1% and 0.77 ± 0.23, respectively. Breath tests with electronic noses can be applied intraoperatively to discriminate breast cancer and molecular subtype and support the medical staff to choose the best therapeutic decision.

A Deep Learning Prognosis Model Help Alert for COVID-19 Patients at High-Risk of Death: A Multi-Center Study

Since its outbreak in December 2019, the persistent coronavirus disease (COVID-19) became a global health emergency. It is imperative to develop a prognostic tool to identify high-risk patients and assist in the formulation of treatment plans. We retrospectively collected 366 severe or critical COVID-19 patients from four centers, including 70 patients who died within 14 days (labeled as high-risk patients) since their initial CT scan and 296 who survived more than 14 days or were cured (labeled as low-risk patients). We developed a 3D densely connected convolutional neural network (termed De-COVID19-Net) to predict the probability of COVID-19 patients belonging to the high-risk or low-risk group, combining CT and clinical information. The area under the curve (AUC) and other evaluation techniques were used to assess our model. The De-COVID19-Net yielded an AUC of 0.952 (95% confidence interval, 0.928-0.977) on the training set and 0.943 (0.904-0.981) on the test set. The stratified analyses indicated that our model's performance is independent of age, sex, and with/without chronic diseases. The Kaplan-Meier analysis revealed that our model could significantly categorize patients into high-risk and low-risk groups (p < 0.001). In conclusion, De-COVID19-Net can non-invasively predict whether a patient will die shortly based on the patient's initial CT scan with an impressive performance, which indicated that it could be used as a potential prognosis tool to alert high-risk patients and intervene in advance.

Identification of a novel anti‑heat shock cognate 71 kDa protein antibody in patients with Kawasaki disease

Kawasaki disease (KD) is an idiopathic form of acute systemic vasculitis, which clinically mimics febrile diseases. Although it has been hypothesized that immune system malfunction is associated with KD, its etiology remains unclear. The aim of the present study was to identify a KD‑associated antibody. Immunoproteomic methods were used to identify KD‑associated antigens that could be recognized in the sera of patients with KD. HeLa cells were used as an antigen source and KD sera were used as probe antibodies to determine the binding of the antibodies using an indirect immunofluorescence assay. Western blotting was performed to identify KD‑associated antigens in HeLa whole cell lysates. Eight out of 12 serum samples obtained from patients with KD demonstrated immunoreactive bands at ~70 kDa, which was later determined to be heat shock cognate 71 kDa protein (HSP7C) by mass spectrometry. The diagnostic value of serum anti‑HSP7C antibodies for KD was assessed using ELISA. Using a cut‑off value of 0.267, anti‑HSP7C antibodies were observed to be present in the sera of 60.00% (30/50) of patients with KD, in 21.05% (8/38) of non‑KD febrile controls, and in 5.26% (2/38) of healthy controls. High serum levels of anti‑HSP7C antibodies were detected in the peripheral circulation of patients with KD. To the best of our knowledge, the present study is the first to observe the high expression levels of anti‑HSP7C antibodies in patients with KD. Therefore, anti‑HSP7C antibodies may be used as a diagnostic marker to detect KD.

Prediction of Prolonged Length of Stay for Stroke Patients on Admission for Inpatient Rehabilitation Based on the International Classification of Functioning, Disability, and Health (ICF) Generic Set: A Study from 50 Centers in China

Background

This study aimed to develop a risk prediction model for prolonged length of stay (LOS) in stroke patients in 50 inpatient rehabilitation centers in 20 provinces across mainland China based on the International Classification of Functioning, Disability, and Health (ICF) Generic Set case mix on admission.

Material/Methods

In this cohort study, 383 stroke patients were included from inpatient rehabilitation settings of 50 hospitals across mainland China. Independent predictors of prolonged LOS were identified using multivariate logistic regression analysis. A prediction model was established and then evaluated by receiver operating characteristic (ROC) curve analysis and the Hosmer-Lemeshow test.

Results

Multivariate logistic regression analysis showed that the type of medical insurance and the performance of daily activities (ICF, d230) were associated with prolonged LOS (P<0.05). Age and mobility level measured by the ICF Generic Set demonstrated no significant predictive value. The prediction model showed acceptable discrimination shown by an area under the curve (AUC) of 0.699 (95% CI, 0.646–0.752) and calibration (χ²=11.66; P=0.308).

Conclusions

The risk prediction model for prolonged LOS in stroke patients in 50 rehabilitation centers in China, based on the ICF Generic Set, showed that the scores for the type of medical insurance and the performance of daily activities (ICF, d230) on admission were independent predictors of prolonged LOS. This prediction model may allow stakeholders to estimate the risk of prolonged LOS on admission quantitatively, facilitate the financial planning, treatment regimens during hospitalization, referral after discharge, and reimbursement.

Diagnostic accuracy of midkine on hepatocellular carcinoma: A meta-analysis

OBJECTIVE

To evaluate the dependability and accuracy of midkine (MK) in the diagnosis of hepatocellular carcinoma (HCC).

METHODS

PubMed, EMBASE, Web of Science, China Biology Medicine disc and grey literature sources were searched from the date of database inception to January 2019. Two authors (B-H.Z. and B.L.) independently extracted the data and evaluated the study quality using the Quality Assessment of Diagnostic Accuracy Studies-2 tool. The sensitivity, specificity, positive likelihood ratio (LR+) and negative likelihood ratio (LR-) were estimated using a bivariate model. Moreover, hierarchical summary receiver operating characteristic curves were generated. The diagnostic odds ratio (DOR) and area under the curve (AUC) were pooled using a univariate model.

RESULTS

Nine articles (11 studies) were included (1941 participants). The bivariate analysis revealed that the sensitivity and specificity of MK for HCC diagnosis were 0.85 (95% CI 0.78-0.91) and 0.83 (95% CI 0.76-0.88), respectively. We also found a LR+ of 5.05 (95% CI 3.33-7.40), a LR- of 0.18 (95% CI 0.11-0.28), a DOR of 31.74 (95% CI 13.98-72.09) and an AUC of 0.91 (95% CI 0.84-0.99). Subgroup analyses showed that MK provided the best efficiency for HCC diagnosis when the cutoff value was greater than 0.5 ng/mL.

CONCLUSIONS

MK has an excellent diagnostic value for hepatocellular carcinoma.

Prognostication via early computed tomography head in patients treated with targeted temperature management after cardiac arrest

BACKGROUND

We evaluated computed tomography head (CTH) imaging obtained prior to targeted temperature management (TTM) in patients after cardiac arrest, and its role in prognostication.

METHODS

In this retrospective cohort study in a tertiary-care hospital, 341 adults presenting with out-of-hospital cardiac arrest received a CTH prior to TTM. Associations between outcomes and neuroimaging variables were evaluated with Chi-square analysis for significant associations that yielded a composite neuroimaging score-Tennessee Early Neuroimaging Score (TENS). Univariable and multivariable logistic regression analysis including TENS as an independent variable and the four outcome dependent variables were analyzed.

RESULTS

Four of the neuroimaging variables-sulcal effacement, partial gray-white matter effacement, total gray-white matter effacement, deep nuclei effacement-had significant associations with each of the four outcome variables and yielded TENS. In multivariable logistic regression models adjusted for potential confounders, TENS was associated with poor discharge CPC (OR 2.15, 95%CI 1.16-3.98, p = .015), poor disposition (OR 2.62, 95%CI 1.37-5.02, p = .004), in-hospital mortality (OR 1.99, 95%CI 1.09-3.62, p = .024), and ICU mortality (OR 1.89, 95%CI 1.12-3.20, p = .018).

CONCLUSION

Imaging prior to TTM may help identify post-cardiac arrest patients with severe anoxic brain injury and poor outcomes.

Indicators of unresponsiveness after initial i.v. immunoglobulin treatment in acute Kawasaki disease

BACKGROUND

The aim of this study was to identify the indicators of unresponsiveness to initial i.v. immunoglobulin (IVIG) treatment for Kawasaki disease (KD).

METHODS

One hundred and forty-five patients with KD, who had received initial treatment consisting of a single IVIG dose (1 g/kg or 2 g/kg) and oral aspirin (30 mg/kg), were studied. Laboratory parameters, including C-reactive protein (CRP) and serum sodium (Na), were measured before and after IVIG treatment, and during the convalescent phase, and the laboratory data compared with regards to IVIG response. Multiple logistic regression models, which included laboratory data obtained immediately after the IVIG treatment, were constructed to determine the indicators of IVIG unresponsiveness immediately after the completion of the initial treatment.

RESULTS

On logistic regression analysis, serum Na after IVIG treatment was the only independent factor related to initial IVIG unresponsiveness (β = 0.53, P < 0.01; OR, 1.69; 95%CI: 1.15-2.49). On receiver operating characteristic curve analysis, the optimal serum Na cut-off immediately after IVIG treatment was 135.5 mEq/L with a sensitivity of 0.75, and a specificity of 0.79.

CONCLUSIONS

Prolonged hyponatremia after completion of the initial IVIG therapy was an indicator of the need for subsequent IVIG therapy. Treatment plans should be established for patients with acute KD that pay particular attention to prolonged hyponatremia.

The prognostic significance of the neutrophil-to-lymphocyte ratio and the platelet-to-lymphocyte ratio in giant cell tumor of the extremities

Background

In this study, the influence of the neutrophil-to-lymphocyte ratio (NLR) and the platelet-to-lymphocyte ratio (PLR) on the prognosis of giant cell tumor (GCT) of the extremities were investigated.

Methods

The clinical parameters of 163 patients who were diagnosed with GCT of the extremities between July 2008 and January 2018 were retrospectively analyzed. Optimal cutoff values of NLR and PLR were determined using receiver operating characteristic (ROC) analysis. According to optimal cutoff values, patients were divided into high NLR and low NLR groups or high PLR and low PLR groups. Kaplan-Meier and log-rank methods were used to compare the recurrence-free survival (RFS) between the high and low NLR groups, and between the high and low PLR groups. Univariate analysis was performed to determine the influence of age, gender, neutrophil count, lymphocyte count, platelet count, white blood cell count, tumor size, surgical approach and Campanacci stage on the prognosis of giant cell tumor of bone. The main predictors of RFS were determined by Cox multivariate regression analysis.

Results

The optimal cutoff value of NLR in giant cell tumor of the extremities was 2.32, which was used to classify patients into high and low NLR groups. The optimal cutoff value of PLR was 116.81, and was used to classify patients into high and low PLR groups. Campanacci stage, tumor maximum diameter, alkaline phosphatase, and C-reactive protein (CRP) were significantly associated with the high NLR and PLR. Cox multivariate regression analysis revealed that the Campanacci stage (HR = 3.28, 95% CI: 1.24~8.69) and NLR (HR = 4.18, 95% CI: 1.83~9.57) were independent prognostic factors for giant cell tumor of the extremities.

Conclusion

As a novel inflammatory index, NLR has some predictive power for the prognosis of patients with giant cell tumor of the extremities.

Differential leukocyte counts on admission predict outcomes in patients with acute ischaemic stroke treated with intravenous thrombolysis

BACKGROUND AND PURPOSE

To determine the association of differential leukocyte counts on admission with efficacy and safety outcomes in patients with acute ischaemic stroke (AIS) treated with intravenous thrombolysis (IVT).

METHODS

Consecutive patients with AIS receiving IVT were evaluated at two stroke centers. Differential leukocyte counts and neutrophil:lymphocyte ratio (NLR) were determined during the initial 12 h of admission. Efficacy outcomes were favorable functional outcome (FFO) (modified Rankin Scale scores of 0-1) and functional independence (FI) (modified Rankin Scale scores of 0-2) at 3 months, whereas safety outcomes were symptomatic intracranial hemorrhage and 3-month mortality.

RESULTS

Among 657 IVT-treated patients with AIS, the mean age was 64 ± 14 years, 50% were female and median National Institutes of Health Stroke Scale score was 7 points (interquartile range, 4-13). Lower neutrophil and leukocyte counts and NLR counts were observed in patients with 3-month FFO and FI, whereas higher counts were observed in patients who died at 3 months. The best discriminative factors for 3-month FFO and FI were NLR < 2.2 (sensitivity 51.4%, specificity 63.1%) and leukocyte count <8100/μL (sensitivity 57.5%, specificity 55.1%), respectively. After adjustment for potential confounders, NLR < 2.2 was associated with higher odds of FFO [odds ratio (OR), 1.56; 95% confidence interval (CI), 1.08-2.24; P = 0.018], whereas leukocyte count <8100/μL demonstrated higher odds of 3-month FI (OR, 1.69; 95% CI, 1.11-2.57; P = 0.014) and lower odds of 3-month mortality (OR, 0.31; 95% CI, 0.16-0.60; P = 0.001). Combined neutrophil (<6800/μL) and leukocyte (<8100/μL) counts demonstrated a strong interaction for 3-month FI (OR, 1.73; 95% CI, 1.13-2.67; P interaction = 0.012).

CONCLUSIONS

Differential leukocyte counts on admission were independently associated with clinical outcomes in patients with AIS treated with IVT. These inflammatory biomarkers are potential targets for adjunctive neuroprotection in this stroke subgroup.

Meta-analysis of full ROC curves using bivariate time-to-event models for interval-censored data

Systematic reviews and meta-analyses are the cornerstones of evidence-based medicine and inform treatment, diagnosis, or prevention of individual patients as well as policy decisions in health care. Statistical methods for the meta-analysis of intervention studies are well established today. Meta-analysis for diagnostic accuracy trials has also been a vivid research area in recent years, which is especially due to the increased complexity of their bivariate outcome of sensitivity and specificity. The situation is even more challenging when single studies report a full ROC curve with several pairs of sensitivity and specificity, each pair for a different threshold. Researchers frequently ignore this information and use only 1 pair of sensitivity and specificity from each study to arrive at meta-analytic estimates. Although methods to deal with the full information have been proposed, they have some disadvantages, eg, the numbers or values of thresholds have to be identical across studies, or the precise values of thresholds are ignored. We propose an approach for the meta-analysis of full ROC curves including the information from all thresholds by using bivariate time-to-event models for interval-censored data with random effects. This approach avoids the problems of previous methods and comes with the additional advantage that it allows for various distributions of the underlying continuous test values. The results from a small simulation study are given, which show that the approach works well in practice. Furthermore, we illustrate our new model using an example based on the population-based screening for type 2 diabetes mellitus.

Z-score of Mitral Annular Plane Systolic Excursion is a Useful Indicator of Evaluation of Left Ventricular Function in Patients with Acute-Phase Kawasaki Disease

We previously reported the clinical usefulness of the mitral annular plane systolic excursion (MAPSE) to evaluate the left ventricular (LV) function in patients with Kawasaki disease (KD) in the acute-phase. However, the feasibility of the MAPSE z-score has not been evaluated in patients with acute KD. We prospectively studied 60 KD patients without coronary aneurysms. The MAPSE z-scores were calculated using our standard MAPSE data. Brain natriuretic peptide (BNP) was measured as a parameter of LV function. In total, 281 healthy age- and body size-matched subjects were chosen as the control group. The MAPSE z-score decreased in the acute-phase (median value, -1.4) and increased in the convalescent phase (median value, 0.18; P < 0.0001). However, there was no significant difference in the MAPSE z-score between patients in the convalescent phase and the control patients (0.18 vs. 0.02, P = 0.199). Multivariate regression analysis revealed that BNP was an independent predictor of the MAPSE z-score (β = 0.40, P < 0.005). According to the receiver operating characteristic (ROC) analysis, the optimal cutoff value for the MAPSE z-score to judge LV dysfunction was -0.9. The MAPSE z-score is a useful index to evaluate LV function, and the cutoff value of -0.9 can be an indicator to judge LV dysfunction in the patients with acute-phase KD.

Association between Low-density lipoprotein cholesterol and occipital periventricular hyperintensities in a group of Chinese patients: an observational study

BACKGROUND

While occipital periventricular hyperintensities (OPVHs) are among the most common mild white matter hyperintensities, the clinical factors associated with OPVHs remain unclear. In this study, we investigated the role of clinical factors in development of pure OPVHs.

METHODS

This study included 97 patients with OPVHs and 73 healthy controls. Univariate analysis of clinical factors in OPVH patients and controls was followed by binomial logistic regression analysis to identify clinical factors significantly associated with OPVHs.

RESULT

Univariate analysis indicated that age, total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C) and apolipoprotein-B (Apo-B) levels differed significantly between the OPVH patients and controls (p < 0.05). Age and gender were correlated with OPVH scores (p < 0.05), while LDL-C, triglycerides, Apo-B and TC were anti-correlated with OPVHs scores (p < 0.05). Multivariate analysis indicated that LDL-C is negatively correlated with OPVHs (p < 0.05), and age is positively correlated with OPVHs (p < 0.001).

CONCLUSION

In summary, LDL-C was negatively and age was positively associated with OPVHs among Chinese patients in a hospital.

Modelling multiple thresholds in meta-analysis of diagnostic test accuracy studies

BACKGROUND

In meta-analyses of diagnostic test accuracy, routinely only one pair of sensitivity and specificity per study is used. However, for tests based on a biomarker or a questionnaire often more than one threshold and the corresponding values of true positives, true negatives, false positives and false negatives are known.

METHODS

We present a new meta-analysis approach using this additional information. It is based on the idea of estimating the distribution functions of the underlying biomarker or questionnaire within the non-diseased and diseased individuals. Assuming a normal or logistic distribution, we estimate the distribution parameters in both groups applying a linear mixed effects model to the transformed data. The model accounts for across-study heterogeneity and dependence of sensitivity and specificity. In addition, a simulation study is presented.

RESULTS

We obtain a summary receiver operating characteristic (SROC) curve as well as the pooled sensitivity and specificity at every specific threshold. Furthermore, the determination of an optimal threshold across studies is possible through maximization of the Youden index. We demonstrate our approach using two meta-analyses of B type natriuretic peptide in heart failure and procalcitonin as a marker for sepsis.

CONCLUSIONS

Our approach uses all the available information and results in an estimation not only of the performance of the biomarker but also of the threshold at which the optimal performance can be expected.

Accuracy of FENO for diagnosing asthma: a systematic review

BACKGROUND

Measurement of FE_NO might substitute bronchial provocation for diagnosing asthma. We aimed to investigate the diagnostic accuracy of FE_NO measurement compared with established reference standard.

METHODS

Systematic review and diagnostic meta-analysis. Data sources were Medline, Embase and Scopus up to 29 November 2015. Sensitivity and specificity were estimated using a bivariate model. Additionally, summary receiver-operating characteristic curves were estimated.

RESULTS

26 studies with 4518 participants (median 113) were included. Risk of bias was considered low for six of seven items in five studies and for five items in seven studies. The overall sensitivity in the meta-analysis was 0.65 (95% CI 0.58 to 0.72), the overall specificity 0.82 (0.76 to 0.86), the diagnostic OR 9.23 (6.55 to 13.01) and the area under the curve 0.80 (0.77 to 0.85). In meta-regression analyses, higher cut-off values were associated with increasing specificity (OR 1.46 per 10 ppb increase in cut-off) while there was no association with sensitivity. Sensitivities varied significantly within the different FE_NO devices, but not specificities. Neither prevalence, age, use of bronchoprovocation in >90% of participants or as exclusive reference standard test, nor risk of bias were significantly associated with diagnostic accuracy.

CONCLUSIONS

There appears to be a fair accuracy of FE_NO for making the diagnosis of asthma. The overall specificity was higher than sensitivity, which indicates a higher diagnostic potential for ruling in than for ruling out the diagnosis of asthma.

Right to Left Ventricular Diameter Ratio ≥0.42 is the Warning Flag for Suspecting Atrial Septal Defect in Preschool Children: Age- and Body Surface Area-Related Reference Values Determined by M-Mode Echocardiography

It is not always easy to observe and screen atrial septal defects (ASD) using echocardiography. In addition, there are no established echocardiographic reference indices for screening patients with ASDs. We retrospectively reviewed our database and recruited 151 isolated ASD patients and 2769 healthy subjects. In total, 307 echocardiographic studies were performed for ASD patients. Surgical repairs were done in 75 of the ASD patients. The ratio of right to left ventricular end-diastolic dimensions (RVD/LVD), which was determined by M-mode echocardiography, was used as an index of RV dilatation. After obtaining age- and body surface area (BSA)-related RVD/LVD nomograms in healthy subjects, we calculated the z-scores of RVD/LVD for all subjects and obtained the optimal cut-off values to differentiate patients with ASD from healthy subjects. The optimal cut-off values were high in neonates and gradually decreased with an increase in the age and BSA, but were almost constant in children aged >4 years or whose BSA was >0.65 m(2). The cut-off values of RVD/LVD for suspected ASD were ≥0.42 in children aged >4 years or those whose BSA was >0.65 m(2). Those for an ASD operation were ≥0.46 in those whose BSA > 0.65 m(2). The RVD/LVD determined by M-mode echocardiography is a useful index to evaluate RV dilatation in patients with ASDs. The RVD/LVD ≥ 0.42 is the warning flag for suspecting ASD in preschool children and that ≥0.46 may be a clinical important sign to determine ASD operation.

Meta-analysis of Diagnostic Accuracy and ROC Curves with Covariate Adjusted Semiparametric Mixtures

Many screening tests dichotomize a measurement to classify subjects. Typically a cut-off value is chosen in a way that allows identification of an acceptable number of cases relative to a reference procedure, but does not produce too many false positives at the same time. Thus for the same sample many pairs of sensitivities and false positive rates result as the cut-off is varied. The curve of these points is called the receiver operating characteristic (ROC) curve. One goal of diagnostic meta-analysis is to integrate ROC curves and arrive at a summary ROC (SROC) curve. Holling, Böhning, and Böhning (Psychometrika 77:106-126, 2012a) demonstrated that finite semiparametric mixtures can describe the heterogeneity in a sample of Lehmann ROC curves well; this approach leads to clusters of SROC curves of a particular shape. We extend this work with the help of the [Formula: see text] transformation, a flexible family of transformations for proportions. A collection of SROC curves is constructed that approximately contains the Lehmann family but in addition allows the modeling of shapes beyond the Lehmann ROC curves. We introduce two rationales for determining the shape from the data. Using the fact that each curve corresponds to a natural univariate measure of diagnostic accuracy, we show how covariate adjusted mixtures lead to a meta-regression on SROC curves. Three worked examples illustrate the method.

Latent class bivariate model for the meta-analysis of diagnostic test accuracy studies

Background

Several types of statistical methods are currently available for the meta-analysis of studies on diagnostic test accuracy. One of these methods is the Bivariate Model which involves a simultaneous analysis of the sensitivity and specificity from a set of studies. In this paper, we review the characteristics of the Bivariate Model and demonstrate how it can be extended with a discrete latent variable. The resulting clustering of studies yields additional insight into the accuracy of the test of interest.

Methods

A Latent Class Bivariate Model is proposed. This model captures the between-study variability in sensitivity and specificity by assuming that studies belong to one of a small number of latent classes. This yields both an easier to interpret and a more precise description of the heterogeneity between studies. Latent classes may not only differ with respect to the average sensitivity and specificity, but also with respect to the correlation between sensitivity and specificity.

Results

The Latent Class Bivariate Model identifies clusters of studies with their own estimates of sensitivity and specificity. Our simulation study demonstrated excellent parameter recovery and good performance of the model selection statistics typically used in latent class analysis. Application in a real data example on coronary artery disease showed that the inclusion of latent classes yields interesting additional information.

Conclusions

Our proposed new meta-analysis method can lead to a better fit of the data set of interest, less biased estimates and more reliable confidence intervals for sensitivities and specificities. But even more important, it may serve as an exploratory tool for subsequent sub-group meta-analyses.

Estimating the optimal threshold for a diagnostic biomarker in case of complex biomarker distributions

BACKGROUND

Estimating the optimal threshold (and especially the confidence interval) of a quantitative biomarker to be used as a diagnostic test is essential for medical decision-making. This is often done with simple methods that are not always reliable. More advanced methods work well but only for biomarkers with very simple distributions. In fact, biomarker distributions are often complex because of a natural heterogeneity in marker expression and other heterogeneities due to various disease stages, laboratory equipments, etc. Methods are required to estimate a biomarker optimal threshold in case of heterogeneity and complex distributions.

METHODS

A previously described Bayesian method developed for normally distributed biomarkers is applied to two flexible distributions; namely, a Student-t and a mixture of Dirichlet processes. Here, numerical studies assess the adequacy of the previous method with both distributions. Two applications are presented: the diagnosis of treatment failure after prostate cancer treated by ultrasound and the early diagnosis of cancers of the upper aerodigestive tract.

RESULTS

Bayesian inference provided reliable credible intervals in terms of bias and coverage probability. The two distributions analysed gave meaningful clinical interpretations in both applications.

CONCLUSIONS

Reliable methods can be used to estimate a biomarker optimal threshold, even in case of complex distributions.

Meta-analysis and meta-modelling for diagnostic problems

BACKGROUND

A proportional hazards measure is suggested in the context of analyzing SROC curves that arise in the meta-analysis of diagnostic studies. The measure can be motivated as a special model: the Lehmann model for ROC curves. The Lehmann model involves study-specific sensitivities and specificities and a diagnostic accuracy parameter which connects the two.

METHODS

A study-specific model is estimated for each study, and the resulting study-specific estimate of diagnostic accuracy is taken as an outcome measure for a mixed model with a random study effect and other study-level covariates as fixed effects. The variance component model becomes estimable by deriving within-study variances, depending on the outcome measure of choice. In contrast to existing approaches - usually of bivariate nature for the outcome measures - the suggested approach is univariate and, hence, allows easily the application of conventional mixed modelling.

RESULTS

Some simple modifications in the SAS procedure proc mixed allow the fitting of mixed models for meta-analytic data from diagnostic studies. The methodology is illustrated with several meta-analytic diagnostic data sets, including a meta-analysis of the Mini-Mental State Examination as a diagnostic device for dementia and mild cognitive impairment.

CONCLUSIONS

The proposed methodology allows us to embed the meta-analysis of diagnostic studies into the well-developed area of mixed modelling. Different outcome measures, specifically from the perspective of whether a local or a global measure of diagnostic accuracy should be applied, are discussed as well. In particular, variation in cut-off value is discussed together with recommendations on choosing the best cut-off value. We also show how this problem can be addressed with the proposed methodology.

A HYBRID PSO-SA OPTIMIZING APPROACH FOR SVM MODELS IN CLASSIFICATION

Support vector machine (SVM) is a widely used tool in the field of image processing and pattern recognition. However, the parameters selection of SVMs is a dilemma in disease identification and clinical diagnosis. This paper proposed an improved parameter optimization method based on traditional particle swarm optimization (PSO) algorithm by changing the fitness function in the traditional evolution process of SVMs. Then, this PSO method was combined with simulated annealing global searching algorithm to avoid local convergence that traditional PSO algorithms usually run into. And this method has achieved better results which reflected in the receiver-operating characteristic curves in medical images classification and has gained considerable identification accuracy in clinical disease detection.

Quantitative metabolomic and lipidomic profiling reveals aberrant amino acid metabolism in type 2 diabetes

Type 2 diabetes (T2DM) is a multi-factorial disease with a complex pathogenic mechanism; however a complete understanding of precise biochemical alterations accompanying the onset and progression of T2DM is lacking. Using a combination of untargeted and targeted metabolomic profiling approach we were able to delineate significantly altered metabolites in the diabetic (T2DM) group. Our results indicate significant perturbations in amino acid metabolism, TCA cycle and glycerol-phospholipid metabolism possibly impacting the overall glucose homeostasis in T2DM. A systems approach offers promise towards identification of clinically relevant markers of T2DM and novel molecular targets to foster drug discovery for effective therapeutic development for diabetes.