Moderate Discriminative Ability Research Articles

Abstract 17712: The Discrimination Ability of the CHA2DS2-VASc Score for Predicting Ischemic Stroke in Patients With and Without Atrial Fibrillation: An Updated Meta-Analysis

BACKGROUND: The CHA2DS2-VASc score guides anticoagulation therapy for stroke prevention in non-valvular atrial fibrillation (NVAF) patients. A previous meta-analysis showed a moderate discrimination ability of CHA2DS2-VASc score to detect stroke in patients with & without NVAF. In the presence of newly published data over the past three years, there is a need to recalibrate the existing statistical evidence. METHODS: A comprehensive literature search was conducted across PubMed, Embase, CENTRAL (the Cochrane Library), & Google Scholar from inception till May 2023. All studies evaluating the discrimination ability of the CHA2DS2-VASc score, as measured by the C statistic, for short (<2 years), intermediate (2-5 years), & long-term (>5 years) follow-ups in patients with & without NVAF were included. The generic inverse variance weighted random effects method was used to pool C-statistic & corresponding standard errors (SEs) from each study. RESULTS: A total of 28 studies with a population of 639450 patients were included. Seventeen studies (n=500555) reported the discrimination ability of CHA2DS2-VASc score for NVAF patients, & 11 studies (n=138895) for patients without NVAF. The summary C statistic for short-term follow up was 0.66 (0.62-0.70) for NVAF subgroup. For mid-term follow-up, the pooled C statistic was 0.65 (0.63-0.67), while being 0.63 (0.60-0.66) & 0.68 (0.66-0.70) for the subgroup with NVAF & without NVAF respectively. The CHA2DS2-VASC score showed moderate discrimination [0.66 (0.62-0.70)] at long-term follow up (NVAF [0.58 (0.55-0.61)]; without NVAF [0.67 (0.64-0.70]). CONCLUSION: CHA2DS2-VASc score has moderate & similar discrimination ability for detecting stroke in both NVAF & non-NVAF patients.

Ideal serum non-ceruloplasmin bound copper prediction for long-term treated patients with Wilson disease: a nomogram model.

This study aimed to explore the factors associated with the optimal serum non-ceruloplasmin bound copper (NCBC) level and develop a flexible predictive model to guide lifelong therapy in Wilson disease (WD) and delay disease progression. We retrospectively collected clinical data from 144 patients hospitalized in the Encephalopathy Center of the first affiliated hospital of Anhui University of Chinese Medicine between May 2012 and April 2023. Independent variables were selected using variate COX and LASSO regressions, followed by multivariate COX regression analysis. A predictive nomogram was constructed and validated using the concordance index (C-index), calibration curves, and clinical decision curve analysis, of which nomogram pictures were utilized for model visualization. A total of 61 (42.36%) patients were included, with an average treatment duration of 55.0 (range, 28.0, 97.0) months. Multivariate regression analysis identified several independent risk factors for serum NCBC level, including age of diagnosis, clinical classification, laminin liver stiffness measurement, and copper to zinc ratio in 24-h urinary excretion. The C-index indicated moderate discriminative ability (48 months: 0.829, 60 months: 0.811, and 72 months: 0.819). The calibration curves showed good consistency and calibration; clinical decision curve analysis demonstrated clinically beneficial threshold probabilities at different time intervals. The predictive nomogram model can predict serum NCBC level; consequently, we recommend its use in clinical practice to delay disease progression and improve the clinical prognosis of WD.

Cross-Cultural Translation and Adaptation of the Consumer Ear Disease Risk Assessment (CEDRA) Questionnaire in Danish.

This study aimed to cross-culturally translate and adapt the Consumer Ear Disease Risk Assessment (CEDRA) questionnaire into Danish for remote ear, nose, and, throat assessments in adult, first-time hearing aid users when used in conjunction with audiometric measures and visual images of the tympanic membrane. Employing field-specific guidelines, the tool underwent a rigorous translation process. This was succeeded by field testing via cognitive debriefing with 30 intendent respondents and a pilot test involving 600 adult, potential first-time hearing aid users from 2020-2022. Test-retest reliability analysis in 113 respondents revealed high consistency and reproducibility, with most items showing Spearman's correlation coefficients of 0.82 or higher and a Pearson's correlation of 0.92 for the total score. The tool demonstrated moderate discriminative ability in identifying individuals at high and low risk of complicated hearing loss and targeted ear diseases, supported by an area under the curve of 0.82 on the receiver operating characteristics curve. Our findings suggest that the Danish-translated version of CEDRA is a reliable and effective screening instrument when used with audiometry and tympanometry, warranting further validation in a larger population.

Frailty in the prediction of delirium in the intensive care unit: A secondary analysis of the Deli study.

Delirium is an acute disorder of attention and cognition with an incidence of up to 70% in the adult intensive care setting. Due to the association with significantly increased morbidity and mortality, it is important to identify who is at the greatest risk of an acute episode of delirium while being cared for in the intensive care. The objective of this study was to determine the ability of the cumulative deficit frailty index and clinical frailty scale to predict an acute episode of delirium among adults admitted to the intensive care. This study is a secondary analysis of the Deli intervention study, a hybrid stepped-wedge cluster randomized controlled trial to assess the effectiveness of a nurse-led intervention to reduce the incidence and duration of delirium among adults admitted to the four adult intensive care units in the south-west of Sydney, Australia. Important predictors of delirium were identified using a bootstrap approach and the absolute risks, based on the cumulative deficit frailty index and the clinical frailty scale are presented. During the 10-mth data collection period (May 2019 and February 2020) 2566 patients were included in the study. Both the cumulative deficit frailty index and the clinical frailty scale on admission, plus age, sex, and APACHE III (AP III) score were able to discriminate between patients who did and did not experience an acute episode of delirium while in the intensive care, with AUC of 0.701 and 0.703 (moderate discriminatory ability), respectively. The addition of a frailty index to a prediction model based on age, sex, and APACHE III score, resulted in net reclassified of risk. Nomograms to individualize the absolute risk of delirium using these predictors are also presented. We have been able to show that both the cumulative deficits frailty index and clinical frailty scale predict an acute episode of delirium among adults admitted to intensive care.

Salivary metabolomics for oral leukoplakia with and without dysplasia

PurposeOral leukoplakia (OL) is a common potentially malignant oral disorder. Therefore, there is a need for simple screening methods for OL before its transformation into oral cancer. Furthermore, because invasive open biopsy is the sole method to determine if an OL lesion is dysplastic, there is also a clinical need for non-invasive methods to differentiate dysplastic OL from non-dysplastic OL. This study aimed to identify salivary metabolites that can help differentiate patients with OL from healthy controls (HC) and also dysplastic OL from non-dysplastic OL. Material & methodsWhole unstimulated saliva samples were collected from patients with OL (n = 30) and HCs (n = 29). The OL group included nine patients with dysplastic OL and 20 with non-dysplastic OL. Hydrophilic metabolites in the saliva samples were comprehensively analyzed through capillary electrophoresis mass spectrometry. To evaluate the discrimination ability of a combination of multiple markers, a multiple logistic regression (MLR) model was developed to differentiate patients with OL from HCs and dysplastic OL from non-dysplastic OL. ResultsTwenty-eight metabolites were evidently different between patients with OL and HCs. Finally, three metabolites (guanine, carnitine, and N-acetylputrescine) were selected to develop the MLR model, which resulted in a high area under curve (AUC) of the receiver operating characteristic (ROC) to differentiate patients with OL from HCs (AUC = 0.946, p < 0.001, 95% confidential interval [CI] = 0.889– 1.000). Similarly, two metabolites were evidently different between patients with dysplastic and non-dysplastic OL. Finally, only one metabolite (7-methylguanine) was selected in the MLR model, which revealed a moderate discrimination ability for dysplastic and non-dysplastic OL (AUC = 0761, p = 0.027, 95% CI = 0.551–0.972). ConclusionOur candidate salivary metabolites showed potential not only to discriminate OL from HC, but also to discriminate dysplastic OL from non-dysplastic OL.

External validity of the PRECISE-DAPT score in patients undergoing PCI: a systematic review and meta-analysis.

To summarise the totality of evidence validating the Predicting Bleeding Complications in Patients Undergoing Stent Implantation and Subsequent Dual Antiplatelet Therapy (PRECISE-DAPT) score, ascertaining its aggregate discrimination and validation power in multiple population subsets. We searched electronic databases from 2017 (PRECISE-DAPT proposal) up to March 2023 for studies that reported the occurrence of out-of-hospital bleedings according to the PRECISE-DAPT score in patients receiving DAPT following percutaneous coronary intervention (PCI). Pooled odds ratio (OR) with 95% confidence interval (CI) were used as summary statistics and were calculated using a random-effects model. Primary and secondary endpoints were the occurrence of any and major bleeding, respectively. A total of twenty-one studies and 67283 patients were included; 24.7% of patients (N=16603) were at high-bleeding risk (PRECISE-DAPT≥25) and when compared to those at low-bleeding risk, they experienced a significantly higher rate of any out-of-hospital bleeding (OR: 2.71; 95% CI: 2.24-3.29; p-value<0.001) and major bleedings (OR: 3.51; 95% CI: 2.71-4.55; p-value<0.001). Pooling data on c-stat whenever available, the PRECISE-DAPT score showed a moderate discriminative power in predicting major bleeding events at 1-year (pooled c-stat: 0.71; 95% CI: 0.64-0.77). This systematic review and meta-analysis confirms the external validity of the PRECISE-DAPT score in predicting out-of-hospital bleeding outcomes in patients on DAPT following PCI. The moderate discriminative ability highlights the need for future improved risk prediction tools in the field.

Triglyceride-glucose index is associated with the occurrence and prognosis of cardiac arrest: a multicenter retrospective observational study

BackgroundTriglyceride-glucose (TyG) index is an efficient indicator of insulin resistance and is proven to be a valuable marker in several cardiovascular diseases. However, the relationship between TyG index and cardiac arrest (CA) remains unclear. The present study aimed to investigate the association of the TyG index with the occurrence and clinical outcomes of CA.MethodsIn this retrospective, multicenter, observational study, critically ill patients, including patients post-CA, were identified from the eICU Collaborative Research Database and evaluated. The TyG index for each patient was calculated using values of triglycerides and glucose recorded within 24 h of intensive care unit (ICU) admission. In-hospital mortality and ICU mortality were the primary clinical outcomes. Logistic regression, restricted cubic spline (RCS), and correlation analyses were performed to explore the relationship between the TyG index and clinical outcomes. Propensity score matching (PSM), overlap weighting (OW), and inverse probability of treatment weighting (IPTW) were adopted to balance the baseline characteristics of patients and minimize selection bias to confirm the robustness of the results. Subgroup analysis based on different modifiers was also performed.ResultsOverall, 24,689 critically ill patients, including 1021 patients post-CA, were enrolled. The TyG index was significantly higher in patients post-CA than in those without CA (9.20 (8.72–9.69) vs. 8.89 (8.45–9.41)), and the TyG index had a moderate discrimination ability to identify patients with CA from the overall population (area under the curve = 0.625). Multivariate logistic regression indicated that the TyG index was an independent risk factor for in-hospital mortality (OR = 1.28, 95% CI: 1.03–1.58) and ICU mortality (OR = 1.27, 95% CI: 1.02–1.58) in patients post-CA. RCS curves revealed that an increased TyG index was linearly related to higher risks of in-hospital and ICU mortality (P for nonlinear: 0.225 and 0.271, respectively). Even after adjusting by PSM, IPTW, and OW, the TyG index remained a risk factor for in-hospital mortality and ICU mortality in patients experiencing CA, which was independent of age, BMI, sex, etc. Correlation analyses revealed that TyG index was negatively correlated with the neurological status of patients post-CA.ConclusionElevated TyG index is significantly associated with the occurrence of CA and higher mortality risk in patients post-CA. Our findings extend the landscape of TyG index in cardiovascular diseases, which requires further prospective cohort study.

MicroRNA biomarkers for diagnosis of mild traumatic brain injury and prediction of persistent symptoms: A prospective cohort study

The diagnosis of mild traumatic brain injury (mTBI) and early identification of patients who have persistent symptoms remains challenging. Symptoms are variably reported, and tests for cognitive impairment require specific expertise. The aim of this study was to assess the ability of plasma micro-ribonucleic acid (miRNA) biomarkers to distinguish between patients with mTBI and healthy controls. A secondary aim was to assess whether miRNA biomarker levels on the day of injury could predict persistent symptoms on day 7. Injured patients presented to an adult, tertiary referral hospital emergency department and were diagnosed with isolated mTBI (n = 75). Venous blood samples were collected within 6 h of injury. Symptom severity was assessed using the Rivermead Post-Concussion Symptom Questionnaire (RPQ) on the day of injury and at 7 days post-injury. The comparator group (n = 44) were healthy controls without any injury, who had bloods sampled and symptom severity assessed at the same time-point. Patients after mTBI reported higher symptom severity and had worse cognitive performance than the control group. Plasma miR423-3p levels were significantly higher among mTBI patients acutely post-injury compared to healthy controls and provided moderate discriminative ability (AUROC 0.67; 95 %CI: 0.57–0.77). None of the assessed miRNA biomarkers predicted persistent symptoms at 7 days. Plasma miR423-3p levels measured within 6 h of injury can discriminate for mTBI compared to healthy controls, with potential utility for screening after head injury or as an adjunct to the diagnosis of mTBI. Acute plasma miRNA levels did not predict patients who reported persistent symptoms at 7 days.

The Italian Family Satisfaction in the Intensive Care Unit Questionnaire: A Psychometric Evaluation Using the Rasch Model.

Quality measurement of the intensive care unit (ICU) should include families' perspectives, their satisfaction with the care process and outcomes, and the evaluation of actions to improve their psychological health and wellbeing. The current study was designed to validate the Italian version of the Family Satisfaction in the Intensive Care Unit (FS-ICU) using the Rasch model. Results included reliability and separation for items and persons, item fit statistics, unidimensionality, and item characteristic curve. The study was conducted between August 2022 and February 2023. A total of 108 family members (mean age 54.9 years) completed the FS-ICU questionnaire. The instrument had a moderate discrimination ability and only five items (#21, #23, #10, #22, and #24) exhibited a misfit. The Rasch dimension explained 52.1% of the variance in the data, while the unexplained variance in the first contrast is 7.2%, which indicates a possible second dimension. FS-ICU was shown to be beneficial as an assessment instrument for family member satisfaction in the ICU, despite some flaws that need to further be addressed to improve the scale.

Charlson comorbidity index to predict 28-day mortality in critically ill COVID-19 patients

BACKGROUND Severe COVID-19 patients may become critically ill and require treatment in the intensive care unit (ICU). As intensive care resources are limited, mortality predictors should be used to guide resource allocation. This study aimed to validate the Charlson comorbidity index (CCI) as the mortality predictor of critical COVID-19 patients in the ICU.&#x0D; METHODS A retrospective cohort study was done in adult patients admitted to the ICU with severe COVID-19 at Cipto Mangunkusumo Hospital and Universitas Indonesia Hospital from March to August 2020. We extracted the subject’s CCI score from the medical records and the 28-day mortality after ICU admission. The CCI score was validated by the Hosmer–Lemeshow calibration test, determination of area under the curve (AUC), and optimal cut-off point for the critical patients in the ICU. We used the chi-square test to examine the association of comorbidities with mortality.&#x0D; RESULTS Mortality was higher in CCI scores &gt;4 (odds ratio [OR]: 8.83; 95% confidence interval [CI] = 1.81–43.01). The CCI score had moderate discrimination ability (AUC 76.1%; 95% CI = 0.661–0.881). Chronic kidney disease (CKD) (OR: 18.00, 95% CI = 2.19–147.51), congestive heart failure (CHF) (OR: 4.25, 95% CI = 1.23–14.75), and uncontrolled diabetes mellitus (DM) (OR: 18.429, 95% CI = 2.19–155.21) increased the risk of 28-day mortality.&#x0D; CONCLUSIONS The CCI score could predict the 28-day mortality of critical COVID-19 patients. The coexistence of CKD, CHF, DM, peripheral vascular disease, and peptic ulcer in COVID-19 patients should be considered for patient management.

Validation of Segmental Post-PCI Physiological Gradients With IVUS-Detected Focal Lesions and Stent Underexpansion

BackgroundSegmental postpercutaneous coronary intervention (PCI) pressure gradients may detect residual disease and potential targets for optimization. However, universal definitions of relevant segmental gradients are lacking. ObjectivesThis study sought to evaluate the diagnostic performance of post-PCI fractional flow reserve (FFR), distal coronary pressure-to-aortic pressure ratio (Pd/Pa), and diastolic pressure ratio (dPR) gradients to detect residual focal lesions and stent underexpansion as observed by intravascular ultrasound (IVUS). MethodsPatients from the IVUS-guided optimization arm of the FFR REACT (FFR-guided PCI Optimization Directed by High-Definition IVUS Versus Standard of Care) trial with complete IVUS and FFR pullback data were included. Patients with angiographically successful PCI and post-PCI FFR <0.90 underwent FFR, Pd/Pa, and IVUS pullbacks. dPR was calculated offline using dedicated software. Segmental pressure gradients (distal, in stent, and proximal) in segments ≥5 mm were evaluated against IVUS-detected residual disease (distal or proximal focal lesions and stent underexpansion). ResultsA total of 139 vessels were included (mean post-PCI FFR: 0.83 ± 0.05, range 0.56-0.89). Focal distal and proximal lesions were detected by IVUS in 23 (17.4%) of 132 and 14 (12.6%) of 111 vessels, respectively, whereas stent underexpansion was present in 86 (61.9%) vessels. Diagnostic ability of segmental FFR gradients to predict IVUS-detected distal and proximal lesions was moderate-to-good (area under the curve [AUC]: 0.69 and 0.84, respectively) and poor to moderate for segmental Pd/Pa and dPR gradients (AUC ranging from 0.58 to 0.69). In-stent gradients had no discriminative ability to detect stent underexpansion (FFR AUC: 0.52; Pd/Pa AUC: 0.54; dPR AUC: 0.55). ConclusionsIn patients with post-PCI FFR <0.90, segmental post-PCI pressure gradients have moderate discriminative ability to identify IVUS-detected focal lesions but no discriminative ability to identify IVUS-detected stent underexpansion.

Development and internal validation of a non-invasive clinical tool to predict sufficient omega-3 levels in early pregnancy

BackgroundComplications from preterm birth (PTB) are the leading cause of death and disability in those under five years. Whilst the role of omega-3 (n-3) supplementation in reducing PTB is well-established, growing evidence suggests supplementation use in those replete may increase the risk of early PTB. Aim: To develop a non-invasive tool to identify individuals with total n-3 serum levels above 4.3% of total fatty acids in early pregnancy.MethodsWe conducted a prospective observational study recruiting 331 participants from three clinical sites in Newcastle, Australia. Eligible participants (n = 307) had a singleton pregnancy between 8 and 20 weeks’ gestation at recruitment. Data on factors associated with n-3 serum levels were collected using an electronic questionnaire; these included estimated intake of n-3 (including food type, portion size, frequency of consumption), n-3 supplementation, and sociodemographic factors. The optimal cut-point of estimated n-3 intake that predicted mothers with total serum n-3 levels likely above 4.3% was developed using multivariate logistic regression, adjusting for maternal age, body mass index, socioeconomic status, and n-3 supplementation use. Total serum n-3 levels above 4.3% was selected as previous research has demonstrated that mothers with these levels are at increased risk of early PTB if they take additional n-3 supplementation during pregnancy. Models were evaluated using various performance metrics including sensitivity, specificity, area under receiver operator characteristic (AUROC) curve, true positive rate (TPR) at 10% false positive rate (FPR), Youden Index, Closest to (0,1) Criteria, Concordance Probability, and Index of Union. Internal validation was performed using 1000-bootstraps to generate 95% confidence intervals for performance metrics generated.ResultsOf 307 eligible participants included for analysis, 58.6% had total n-3 serum levels above 4.3%. The optimal model had a moderate discriminative ability (AUROC 0.744, 95% CI 0.742–0.746) with 84.7% sensitivity, 54.7% specificity and 37.6% TPR at 10% FPR.ConclusionsOur non-invasive tool was a moderate predictor of pregnant women with total serum n-3 levels above 4.3%; however, its performance is not yet adequate for clinical use.Trial registrationThis trial was approved by the Hunter New England Human Research Ethics Committee of the Hunter New England Local Health District (Reference 2020/ETH00498 on 07/05/2020 and 2020/ETH02881 on 08/12/2020).

#5569 ASSOCIATION OF URINARY CCL14 WITH PLASMA PROTEIN BIOMARKERS IN CRITICALLY ILL SEPSIS PATIENTS WITH PERSISTENT ACUTE KIDNEY INJURY

Abstract Background and Aims The urinary concentration of CC chemokine ligand 14 (CCL14) – a chemoattractant for T-cells and monocytes – has recently been identified as a strong predictor of persistence of severe acute kidney injury (AKI). The mechanisms by which CCL14 mediates persistent severe AKI are not yet completely understood. With sepsis being the most frequent cause of AKI, we wanted to externally validate whether urinary CCL14 concentration can differentiate sepsis patients with and without persistence of AKI, and its association with plasma protein biomarkers reflective of three pathophysiological pathways. Method Patients admitted with sepsis to the intensive care unit (ICU) were categorized according to the evolution of AKI (no-AKI, transient-AKI, and persistent-AKI, defined as lasting &amp;gt; 48h), and matched to a 1:1:1 ratio. AKI was assessed prospectively and daily, using the risk, injury, failure, loss, and end-stage kidney disease (RIFLE) classification. Transient- and persistent-AKI patients had at least injury or failure (severe) RIFLE stages. Urine samples collected at ICU admission were analyzed for CCL14 level; in the same patient cohort we measured 9 plasma biomarkers reflective of systemic inflammation and cytokine responses, endothelial cell activation, and loss of endothelial barrier function. We applied a recently proposed cutoff of 1.3 ng/mL for high CCL14 for the identification of patients at high risk for persistent severe AKI. Results of 211 sepsis patients enrolled, 72 patients did not have AKI, 71 had transient AKI and 68 had persistent AKI. Majority of patients with high CCL14 had a persistent AKI (23 out of 31; 74.2%). CCL14 showed good discrimination between persistent-AKI and no-AKI (Area Under the ROC Curve [AUC], 0.82, 95%CI 0.75-0.89), and moderate discrimination between persistent-AKI and transient-AKI (AUC 0.71, 95%CI 0.55-0.87). High CCL14 was associated with higher disease severity as measured by APACHE-IV score, and source of infection being the urinary tract, and skin. The plasma protein analysis revealed that high urinary CCL14 was associated with more prominent systemic (anti-)inflammatory and cytokine responses, and signs of disrupted endothelial barrier function (shown by elevated interleukin [IL]-8 and IL-10, and decreased angiopoietin-1 [P = .005; P = .004; P = .009, respectively]). Within the persistent-AKI group, baseline characteristics (age, chronic comorbidities, and disease severity) did not differ between CCL14 groups, and high CCL14 was only associated with decreased angiopoietin-1 (P = .003). Conclusion External validation of urinary CCL14 showed good discrimination in critically ill sepsis patients between persistent AKI and no-AKI, and moderate discriminative ability to separate transient-AKI. High urinary CCL14 identifies persistent-AKI sepsis patients with more profound dysregulation of inflammatory pathways coupled with loss of endothelial barrier function. This could guide future therapeutic strategies targeting subgroups within the AKI population.

Ultrasound-based radiomics analysis in the assessment of renal fibrosis in patients with chronic kidney disease.

Assessment of renal fibrosis non-invasively in chronic kidney disease (CKD) patients is still a clinical challenge. In this study, we aimed to establish a radiomics model integrating radiomics features derived from ultrasound (US) images with clinical characteristics for the assessment of renal fibrosis severity in CKD patients. A total of 160 patients with CKD who underwent kidney biopsy and renal US examination were prospectively enrolled. Patients were classified into the mild or moderate-severe fibrosis group based on pathology results. Radiomics features were extracted from the US images, and a radiomics signature was constructed using the maximum relevance minimum redundancy (mRMR) and least absolute shrinkage and selection operator (LASSO) regression algorithms. Multivariable logistic regression was employed to construct the radiomics model, which incorporated the radiomics signature and the selected clinical variables. The established model was evaluated for discrimination, calibration, and clinical utility in the derivation cohort and internal cross-validation (CV) analysis, respectively. The radiomics signature, consisting of nine identified fibrosis-related features, achieved moderate discriminatory ability with an area under the receiver operating characteristic curve (AUC) of 0.72 (95% confidence interval (CI) 0.64-0.79). By combining the radiomics signature with significant clinical risk factors, the radiomics model showed satisfactory discrimination performance, yielding an AUC of 0.85 (95% CI 0.79-0.91) in the derivation cohort and a mean AUC of 0.84 (95% CI 0.77-0.92) in the internal CV analysis. It also demonstrated fine accuracy via the calibration curve. Furthermore, the decision curve analysis indicated that the model was clinically useful. The proposed radiomics model showed favorable performance in determining the individualized risk of moderate-severe renal fibrosis in patients with CKD, which may facilitate more effective clinical decision-making.

Advancing heart failure research using machine learning

Advancing heart failure research using machine learning

Validity of Acute Physiology and Chronic Health Evaluation (APACHE) IV for the Prediction of Prolonged Intensive Care Unit (ICU) Length of Stay in Dr. Sardjito General Hospital in the COVID Era

Introduction: APACHE IV was a good predictor of ICU length of stay in the USA and some countries outside the USA but poor in others. It is important to develop a scoring system for the Indonesian population, especially in this scope, Dr. Sardjito General Hospital. To develop such a scoring system, it is reasonable to study the validity of APACHE IV in ICU Dr. Sardjito General Hospital for predicting prolonged length of stay.&#x0D; Methods: A retrospective cohort observational study using data from January 1st, 2020, to December 31st, 2020, taken from the ICU of Dr. Sardjito General Hospital. The data are the patient's observed ICU LOS and data required in calculating APACHE IV score and ICU LOS prediction. Discrimination is calculated using the area under (AUC) the receiver operating characteristic curve (ROC) and calibration by the Hosmer-Lemeshow test.&#x0D; Results: Samples were 329 patients. APACHE IV ICU length of stay prediction showed moderate discriminatory ability (AUC-ROC: 0.74) and poor calibration (p &lt;0.001) to predict prolonged ICU stay. The APACHE IV score has a strong discriminatory ability (AUC-ROC: 0.83). Using the DeLong method, the AUC from ROC APACHE IV score was greater than the AUC from ROC predicted length of stay in APACHE IV ICU (p &lt;0.001). APACHE IV predicted ICU length of stay overestimated observed ICU length of stay.&#x0D; Conclusion: APACHE IV ICU length of stay prediction has moderate discrimination and poor calibration to predict prolonged ICU stay. The APACHE IV score has better discrimination than the APACHE IV ICU length of stay prediction in predicting prolonged ICU stay.

Clinicopathological value of hematopoietic cell kinase overexpression in laryngeal squamous cell carcinoma tissues

Laryngeal squamous cell carcinoma (LSCC) is the most lethal cancer in head and neck tumors. Although hematopoietic cell kinase (HCK) has been proven to be an oncogene in several solid tumors, its roles in LSCC remain obscure. This is the first study to evaluate the clinical value of HCK in LSCC, with the aim of exploring its expression status and potential molecular mechanisms underlying LSCC. LSCC tissue-derived gene chips and RNA-seq data were collected for a quantitive integration of HCK mRNA expression level. To confirm the protein expression level of HCK, a total of 82 LSCC tissue specimens and 56 non-tumor laryngeal epithelial controls were collected for in-house tissue microarrays and immunohistochemical staining. Kaplan-Meier curves were generated to determine the ability of HCK in predicting overall survival, progress-free survival, and disease-free survival of LSCC patients. LSCC overexpressed genes and HCK co-expressed genes were intersected to preliminarily explore the enriched signaling pathways of HCK. It was noticed that HCK mRNA was markedly overexpressed in 323 LSCC tissues compared with 196 non-LSCC controls (standardized mean difference = 0.81, p < 0.0001). Upregulated HCK mRNA displayed a moderate discriminatory ability between LSCC tissues and non-tumor laryngeal epithelial controls (area under the curve = 0.78, sensitivity = 0.76, specificity = 0.68). The higher expression level of HCK mRNA could predict worse overall survival and disease-free survival for LSCC patients (p = 0.041 and p = 0.013). Lastly, upregulated co-expression genes of HCK were significantly enriched in leukocyte cell-cell adhesion, secretory granule membrane, and extracellular matrix structural constituent. Immune-related pathways were the predominantly activated signals, such as cytokine-cytokine receptor interaction, Th17 cell differentiation, and Toll-like receptor signaling pathway. In conclusion, HCK was upregulated in LSCC tissues and could be utilized as a risk predictor. HCK may promote the development of LSCC by disturbing immune signaling pathways.

External validation of 18 F-FDG PET-based radiomic models on identification of residual oesophageal cancer after neoadjuvant chemoradiotherapy.

Detection of residual oesophageal cancer after neoadjuvant chemoradiotherapy (nCRT) is important to guide treatment decisions regarding standard oesophagectomy or active surveillance. The aim was to validate previously developed 18 F-FDG PET-based radiomic models to detect residual local tumour and to repeat model development (i.e. 'model extension') in case of poor generalisability. This was a retrospective cohort study in patients collected from a prospective multicentre study in four Dutch institutes. Patients underwent nCRT followed by oesophagectomy between 2013 and 2019. Outcome was tumour regression grade (TRG) 1 (0% tumour) versus TRG 2-3-4 (≥1% tumour). Scans were acquired according to standardised protocols. Discrimination and calibration were assessed for the published models with optimism-corrected AUCs >0.77. For model extension, the development and external validation cohorts were combined. Baseline characteristics of the 189 patients included [median age 66 years (interquartile range 60-71), 158/189 male (84%), 40/189 TRG 1 (21%) and 149/189 (79%) TRG 2-3-4] were comparable to the development cohort. The model including cT stage plus the feature 'sum entropy' had best discriminative performance in external validation (AUC 0.64, 95% confidence interval 0.55-0.73), with a calibration slope and intercept of 0.16 and 0.48 respectively. An extended bootstrapped LASSO model yielded an AUC of 0.65 for TRG 2-3-4 detection. The high predictive performance of the published radiomic models could not be replicated. The extended model had moderate discriminative ability. The investigated radiomic models appeared inaccurate to detect local residual oesophageal tumour and cannot be used as an adjunct tool for clinical decision-making in patients.

Automated Breast Ultrasound (ABUS)-based radiomics nomogram: an individualized tool for predicting axillary lymph node tumor burden in patients with early breast cancer

ObjectivesPreoperative evaluation of axillary lymph node (ALN) status is an essential part of deciding the appropriate treatment. According to ACOSOG Z0011 trials, the new goal of the ALN status evaluation is tumor burden (low burden, < 3 positive ALNs; high burden, ≥ 3 positive ALNs), instead of metastasis or non-metastasis.We aimed to develop a radiomics nomogram integrating clinicopathologic features, ABUS imaging features and radiomics features from ABUS for predicting ALN tumor burden in early breast cancer.MethodsA total of 310 patients with breast cancer were enrolled. Radiomics score was generated from the ABUS images. Multivariate logistic regression analysis was used to develop the predicting model, we incorporated the radiomics score, ABUS imaging features and clinicopathologic features, and this was presented with a radiomics nomogram. Besides, we separately constructed an ABUS model to analyze the performance of ABUS imaging features in predicting ALN tumor burden. The performance of the models was assessed through discrimination, calibration curve, and decision curve.ResultsThe radiomics score, which consisted of 13 selected features, showed moderate discriminative ability (AUC 0.794 and 0.789 in the training and test sets). The ABUS model, comprising diameter, hyperechoic halo, and retraction phenomenon, showed moderate predictive ability (AUC 0.772 and 0.736 in the training and test sets). The ABUS radiomics nomogram, integrating radiomics score with retraction phenomenon and US-reported ALN status, showed an accurate agreement between ALN tumor burden and pathological verification (AUC 0.876 and 0.851 in the training and test sets). The decision curves showed that ABUS radiomics nomogram was clinically useful and more excellent than US-reported ALN status by experienced radiologists.ConclusionsThe ABUS radiomics nomogram, with non-invasive, individualized and precise assessment, may assist clinicians to determine the optimal treatment strategy and avoid overtreatment.

PD34-05 A NOVEL MACHINE-LEARNING ALGORITHM TO PREDICT STONE RECURRENCE WITH 24-HOUR URINE DATA

You have accessJournal of UrologyCME1 Apr 2023PD34-05 A NOVEL MACHINE-LEARNING ALGORITHM TO PREDICT STONE RECURRENCE WITH 24-HOUR URINE DATA Kevin Shee, Carter Chan, Andrew Liu, Heiko Yang, Wilson Sui, Sunita Ho, Thomas Chi, and Marshall Stoller Kevin SheeKevin Shee More articles by this author , Carter ChanCarter Chan More articles by this author , Andrew LiuAndrew Liu More articles by this author , Heiko YangHeiko Yang More articles by this author , Wilson SuiWilson Sui More articles by this author , Sunita HoSunita Ho More articles by this author , Thomas ChiThomas Chi More articles by this author , and Marshall StollerMarshall Stoller More articles by this author View All Author Informationhttps://doi.org/10.1097/JU.0000000000003327.05AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookLinked InTwitterEmail Abstract INTRODUCTION AND OBJECTIVE: There are limited predictive markers of stone recurrence, which poses a challenge for the clinical management of stone disease. Furthermore, inability to predict stone events is a significant limitation for clinical trials, where many patients must be enrolled to obtain sufficient stone events for analysis. In this study, we sought to use machine learning methods to identify a novel algorithm to predict stone recurrence. METHODS: Patients enrolled in the Registry for Stones of the Kidney and Ureter (ReSKU), a multi-institutional registry of nephrolithiasis patients collected between 2015-2020, with at least one prospectively collected 24-hour urine test (Litholink) were included in the training set. A test set of stone patients not enrolled in ReSKU with 24-hour urine test results were collected. Stone events were defined as either an office visit where a patient reported symptomatic passage of stone(s) or surgery for stone removal. Seven prediction classification methods were evaluated: Decision Tree, Logistic regression with ElasticNet, Extra Trees, kNeighbors, LightGBM, Logistic Regression, and Random Forest. Probability of Recurrent Nephrolithiasis (PRN) score ranging from 0-1 was generated from the models. Predictive analyses and ROC curve generation was performed in R. RESULTS: A training set of 423 kidney stone patients with stone event data and 24-hour urine samples were trained using the prediction classification methods. The highest performing prediction model was a Logistic Regression with ElasticNet machine learning model (AUC=0.65; Figure 1A). Selecting for high confidence predictions (PRN 0-0.15, 0.8-1.0) improved model accuracy (AUC=0.82; Figure 1B). The model was validated on a test set of 172 stone patients. Patients predicted to have high likelihood of stone recurrence had an increase in the average number of stone events compared to those with low likelihood (3.6 vs 1.7 events; p=2e-7). Prediction accuracy in the test set demonstrated moderate discriminative ability (AUC=0.64; Figure 1C). CONCLUSIONS: Using machine-learning trained on 24-hour urine and stone outcome data, a novel algorithm was developed to predict stone recurrence. Initial validation studies suggest ability of algorithm to identify patients with higher likelihood of stone events and warrants further study. Source of Funding: n/a © 2023 by American Urological Association Education and Research, Inc.FiguresReferencesRelatedDetails Volume 209Issue Supplement 4April 2023Page: e921 Advertisement Copyright & Permissions© 2023 by American Urological Association Education and Research, Inc.MetricsAuthor Information Kevin Shee More articles by this author Carter Chan More articles by this author Andrew Liu More articles by this author Heiko Yang More articles by this author Wilson Sui More articles by this author Sunita Ho More articles by this author Thomas Chi More articles by this author Marshall Stoller More articles by this author Expand All Advertisement PDF downloadLoading ...