Area Under The ROC Curve Research Articles

B-136 From Result to Response: The Development of Laboratory-Based Scoring Models to Predict COVID-19 Patient Outcomes

Abstract Background Patient characteristics relating to increased risk for COVID-19 severity are well-documented, including older age and pre-existing health concerns; however, methods to accurately predict patients’ acute reaction to SARS-CoV-2 infections remain lacking. Identification of specific laboratory tests informative of patient outcomes could effectively screen incoming patients and better inform physicians regarding optimal treatment plans. The aim of this study was to examine associations between frequently ordered laboratory markers and COVID-19 patient mortality to develop an objective laboratory-based scoring system that estimates patients’ risk of mortality. Methods Study participants included inpatients from 3 hospitals recruited into the GENCOV project based in Ontario, Canada. Participants (n=325) were at least 18 years of age, provided consent, and were hospitalized within 1 month of having a PCR confirmed COVID-19 infection between January 2020 and February 2022. Extensive clinical data including patient demographics, laboratory results, and treatment outcomes were extracted from patient’s electronic medical records (EMR). Results for 32 biochemical and hematological tests including complete blood cell counts, coagulation (activated partial thromboplastin time, D-dimer, fibrinogen, prothrombin time), general chemistry (albumin, blood gases, creatine kinase, electrolytes, glucose, triglycerides), inflammatory (lactate dehydrogenase, ferritin, C-reactive protein), liver (alanine aminotransferase, aspartate aminotransferase, total bilirubin), renal (creatinine, urea) and cardiac (troponin, NT-proBNP, BNP) markers were collected from each chart. Univariable logistic regression with nested likelihood ratio tests were used to determine significant associations between laboratory results and patient outcomes. Significant markers were incorporated into multivariable models and variable risk values were assigned. Total calculated risk scores for each participant were compared using univariable and multivariable risk values. Validation of each scoring method was performed on a 20% subset of inpatients. Receiver operating characteristic (ROC) curves evaluated model performance. Results Six laboratory markers were associated with COVID-19 patient mortality when controlling for age and sex. Univariable regression showed that elevated creatinine, elevated lactate, elevated white blood cell, low base excess, low bicarbonate, or low pH results upon admission were significantly associated with increased odds of mortality, in comparison to test results within the reference range for these same markers. Multivariable regression revealed fewer markers (only low bicarbonate and low base excess) showing significant associations with COVID-19 mortality. Area under the ROC curves (AUC) determined that risk scores derived from univariable values performed similarly to multivariable values in validation (0.800 vs 0.802) and development cohorts (0.821 vs 0.829). Total risk scores calculated from the univariable vs multivariable models suggest higher sensitivity (90% vs 85%) than specificity (58% vs 67%) in the validation cohort, whereas the development cohort had higher sensitivity (88%) than specificity (66%) with univariable scores, and higher specificity (81%) than sensitivity (74%) with multivariable scores. Conclusions Laboratory results associated with COVID-19 mortality following both univariable and multivariable analyses suggest hospitalized patients infected with SARS-CoV-2 may present with acidosis. Total risk scores derived from univariable and multivariable values performed similarly in predicting mortality; however, differences in marker significance between models indicate discrepancies with risk interpretation. Further comparison of the created risk score assessments with equivalent models in other populations is warranted.

A-021 Evaluation of a Novel N Terminal Pro B Type Natriuretic Peptide Assay as an aid in the Diagnosis of Acute Heart Failure in the Emergency Department

Abstract Background The signs and symptoms of acute heart failure (AHF) may overlap with other conditions and lead to diagnostic difficulty in the undifferentiated patient with dyspnea. The particularly high negative predictive value (NPV) of natriuretic peptides (NPs) in excluding AHF have rendered these assays especially useful in evaluating dyspnea in acute care setting. This study was conducted to assess the diagnostic performance of a novel N-terminal pro-B type natriuretic peptide (NT-proBNP) assay in the intended use patient population in the emergency department (ED). Methods A multicenter study was conducted in the U.S. on individuals presenting to the ED with dyspnea and a suspicion of AHF. Blood specimens were collected and tested using the Beckman Coulter Access NT-proBNP assay on the DxI 9000 Immunoassay Analyzer. Test results were compared to an adjudicated diagnosis clinical endpoint performed by an independent Clinical Events Committee blinded to the Access NT-proBNP result. Performance included NPV and sensitivity of an age-independent cut point of <300 ng/L to exclude AHF, and the positive predictive value (PPV) of the age-dependent cut points of >450, >900, and >1800 ng/L for ages <50 years, 50-75 years, and >75 years, respectively, for the diagnosis of AHF. The sample size required to achieve a 95% two tailed confidence interval (CI) was 150 subjects per age group to provide clinical performance at 90% (±10%) sensitivity. Sample size was based upon CLSI EP24-A2, Assessment of the Diagnostic Accuracy of Laboratory Tests Using Receiver Operating Characteristic Curves. Sensitivity and specificity were calculated at a series of cutoff points and plotted in separate receiver operation curves (ROC) for an overall age-independent rule-out analysis and for individual age-dependent groups separately. Area under the receiver operator characteristic curve (AUROC) was reported with 95% 2-sided CI. NPV and PPV with respective 95% 2-sided score CI was computed for the overall age-independent cut point and for each age group cohort. Results Of 490 enrolled patients, 41% were adjudicated with AHF with a median (Q1, Q3) age of 60 (47-76) years, and distributions of 46.1% female, 60.8% White and 38.2% Black. The assay had an AUROC for AHF of 0.88 (P<0.001), comparable to other commercially available NT-proBNP assays. The rule-out cut point of <300 ng/L had 96% sensitivity and NPV of 95%. Age-dependent cut points had sensitivity of 84%, 90%, and 87%, specificity of 81%, 70% and 61%, and PPV of 72%, 62%, and 74%, respectively. Conclusions The Access NT-proBNP assay demonstrated high clinical performance in the diagnosis and exclusion of acute HF in the undifferentiated dyspneic patient population and performed similarly well to validated assays used in current clinical practice. Given the continued increase in importance and reach of natriuretic peptide testing, these results are useful to extend availability of this assay to a larger audience.

Technological Advancements in Menstrual Health: The Role of Generative Pre-Trained Transformer and Bees Algorithm

This paper introduces the concept of a smart menstrual cup, incorporating advanced technology to address challenges associated with traditional menstrual health management methods. The proposed method, Hybrid Bidirectional Encoder Generative Transformer based Bees Search (Hybrid BEGT-BS), aims to enhance the longevity and reliability of smart menstrual cups. The Generative Pre-trained Transformer (GPT) is employed to extract key information and determine emotional tones related to menstruation, while the Bidirectional Encoder Representations from Transformer (BERT) classifies menstrual-related data such as cycle tracking, hygiene, and symptoms. Further different validation metrics such as F1-score, specificity, Area Under the Curve-Receiver Operating Characteristic (AUC-ROC), Mean Squared Error (MSE), recall, energy consumption, accuracy, and precision are employed to assess the method's effectiveness. From the comparative results it demonstrates the superior performance of the Hybrid BEGT-BS method in enhancing the performance of smart menstrual cups.

Evaluating Respiratory Muscle Strength in Sarcopenia Screening among Older Men in South Korea: A Retrospective Analysis.

To evaluate the potential of incorporating respiratory muscle strength, specifically maximal inspiratory pressure (MIP), and maximal expiratory pressure (MEP), along with traditional sarcopenia screening measures such as hand grip strength (HGS) and skeletal muscle mass index (SMI), to identify sarcopenia in older men. A retrospective analysis was conducted involving male patients aged 65 years and older who underwent measurements of respiratory muscle strength, HGS, and muscle mass at a general hospital in Korea from July 2016 to May 2022. Statistical analysis utilized independent t-tests and receiver operating characteristic (ROC) curves to assess the sensitivity and specificity of MIP, MEP, HGS, and SMI in sarcopenia screening. The cut-off values for sarcopenia screening were determined based on the area under the ROC curve (AUC). The analysis of 282 study participants revealed the following cut-off values for sarcopenia based on the AUC: for MIP, the cut-off value was 65.50 cmH2O (AUC=0.70, sensitivity: 0.63, specificity: 0.61), while for MEP, it was 84.50 cmH2O (AUC=0.74, sensitivity: 0.66, specificity: 0.68). This study showed the utility of respiratory muscle strength in screening for sarcopenia among older men. We suggest the screening cut-off values as 65.50 cmH2O for MIP and 84.50 cmH2O for MEP. Even when HGS and SMI measurements are not feasible, sarcopenia can be reasonably predicted based on respiratory muscle strength.

Predicting Mortality in Trauma Research: Evaluating the Performance of Trauma Scoring Tools in a South African Population.

Background Trauma is a leading cause of death and disability in low-resource settings, yet trauma severity scores are seldom validated in these contexts. There is a pressing need to better characterize and compare trauma scoring tools, especially within research frameworks. This study aimed to evaluate the performance of various trauma scoring tools in predicting in-hospital mortality among trauma patients in South Africa. Methods This study conducted a secondary analysis of existing data from the multicenter Epidemiology and Outcomes of Prolonged Trauma Care (EpiC) study, which included 13,548 adult trauma patients aged 18 years and older, collected between August 2021 and March 2024. The predictive ability of the scoring tools was assessed by calculating the area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (AUPRC). Results The mortality rate was 2.5% (n = 298). The Kampala Trauma Score (KTS) demonstrated the highest predictive ability for seven-day in-hospital mortality, with an AUROC of 0.95 and an AUPRC of 0.53. Similarly, the Trauma and Injury Severity Score (TRISS) and the New Injury Severity Score (NISS) also exhibited strong predictive capabilities, with AUROC values of 0.96 and AUPRC values of 0.62 for TRISS and an AUROC of 0.96 and AUPRC of 0.53 for NISS. In contrast, the Revised Trauma Score and Mechanism, Glasgow Coma Scale, Age, and Arterial Pressure (MGAP) showed lower predictive performance, with AUROC values of 0.87 (AUPRC = 0.51) and 0.86 (AUPRC = 0.47), respectively. Conclusions The KTS exhibited optimal performance characteristics for retrospectively predicting mortality in our cohort, outperforming other scoring tools. Notably, it is also the simplest scoring tool, featuring the fewest variables compared to other trauma severity assessments. These findings highlight the necessity for external validation of trauma scoring tools in resource-limited populations to ensure their applicability and effectiveness in trauma research across diverse healthcare settings.

Correlation between Bone Marrow Microvascular Density, Angiogenesis Factors and Bortezomib Resistance in Multiple Myeloma

To investigate the correlation between bone marrow microvascular density, angiogenesis factors and bortezomib resistance in multiple myeloma (MM). The data of 200 patients with MM treated in our hospital from January 2020 to August 2023 were retrospectively analyzed, and the patients with MM were divided into drug-resistant group(n=68) and non-drug-resistant group(n=132) according to their drug resistance during bortezomib treatment. The univariate and multivariate logistic analysis were used to screen the independent influencing factors of bortezomib resistance in MM patients during treatment. The receiver operating characteristic (ROC) curve and clinical decision curve (DCA) were used to evaluate the predictive performance and clinical application value of the risk prediction model, the consistency between the actual incidence rate and the predicted incidence rate was judged by validating the calibration chart, and the goodness-of-fit of the model was judged by H-L test. 68 of the 200 MM patients developed resistance and poor clinical efficacy during bortezomib treatment, and the clinical resistance rate of bortezomib was 34.0%. The results of multivariate analysis showed that high bone marrow microvessel density (MVD) and high bone marrow supernatant VEGF, HGF, and bFGF expression levels were independent risk factors for bortezomib resistance in MM patients (P < 0.05). The area under the ROC curve (AUC) of the model jointly constructed by bone marrow MVD, serum VEGF, HGF, bFGF and TNF-α levels was 0.924, and its sensitivity and specificity were 92.6% and 78.8%, which were higher than those of the bone marrow MVD model (AUC=0.743) and the vasogenesis factor model (AUC=0.878). The calibration curve of the joint prediction model was close to the standard curve, indicating that the model is more consistent. The results of H-L goodness -of - fit test showed χ2=14.748， P =0.164, the joint prediction model had a good fit. The DCA curve showed that the clinical net benefit of intervention in the range of 0.0~1.0 was greater than that of full intervention and no intervention. The prediction model based on bone marrow MVD and vasogenesis factors (VEGF, HGF, bFGF) in MM patients has higher clinical evaluation performance and predictive value.

Utility of multi-biomarker panel on discriminating disease activity in patients with psoriatic arthritis

ObjectiveTo investigate the correlation of serum protein biomarkers and disease activity in patients with PsA. Methods176 patients fulfilled the CASPAR (ClASsification criteria for Psoriatic ARthritis) were recruited in this cross-sectional study. The level of 48 protein biomarkers, cartilage and bone turn-over markers were assessed. The patients were randomly divided into a derivation-cohort and a validation-cohort at a ratio of 7:3. Patients were further categorized based on their disease activity states using cDAPSA (remission/low disease activity and moderate/high disease activity). Least absolute shrinkage and selection operator (LASSO) was used to select biomarkers which were associated with moderate/high disease activity in the derivation cohort. Receiver operating characteristic (ROC) curve, GiViTI calibration belt were used to assess the performance of the model in both cohorts. ResultsThe cohort [age: 55.5 (44.0–62.75) years, male: 80 (45.5 %)] had moderate disease activity [DAPSA: 15.9 (8.3–26.9); PASI: 3.2 (0.5–6.8)]. 101 PsA patients (57.4 %) had clinical DAPSA moderate/high disease activity. Biomarker levels associated with moderate/high disease activity included SAA (Serum amyloid A), IL-8 (Interleukin 8), IP10 (Interferon gamma-induced protein 10)/CXCL10, M-CSF (Macrophage colony-stimulating factor), SCGF-β (Stem cell growth factor), SDF-1α (Stromal cell-derived factor 1α)/CXCL12. The model’s equation including the 6 biomarker levels was applied to the validation-cohort. The area under the ROC curve (AUC) for discriminating moderate/high disease activity was 0.802 and 0.835 for the derivation-and-validation-cohorts, respectively. The multi-biomarkers panel model had higher-AUC when compared with that of C-reactive protein (CRP) (AUC = 0.727, p = 0.022). The P-values of calibration charts in the two sets were 0.902 and 0.123. ConclusionsThe multi-biomarkers panel demonstrated the ability to discriminate patients with moderate/high disease activity from those with low disease activity/remission.

Difference in hematocrit and plasma albumin levels as an early biomarker of severity and prognosis in patients with severe fever and thrombocytopenia syndrome.

Severe fever with thrombocytopenia syndrome (SFTS) is a widespread infectious disease with high mortality. Hence, identifying valuable biomarkers for detecting the early changes in SFTS is crucial. In this study, we investigated the relationship between the difference in hematocrit (HCT) and serum albumin (ALB) levels (HCT-ALB) and the prognosis of patients with SFTS virus infection. After excluding the patients who did not meet the SFTS diagnostic criteria, those with SFTS from the First Affiliated Hospital of Wannan Medical College were divided into a fatal and Nonfatal group based on their disease prognosis. A dynamic analysis of the daily laboratory data was conducted for 14 days following SFTS onset. A receiver operating characteristic (ROC) curve was used to evaluate the predictive value of HCT-ALB. Another sample of patients with SFTS admitted to the First Affiliated Hospital of Nanjing Medical University was utilized to verify the study conclusions. A total of 158 patients with SFTS were included. Among them, 126 patients were categorized in the Nonfatal group and 32 in the fatal group, leading to a mortality rate of 20.25% (32/158). Univariate analysis of the laboratory test findings and ROC curve analysis showed that alanine aminotransferase (ALT), aspartate aminotransferase (AST), HCT-ALB, and lactate dehydrogenase (LDH) had a relatively better ability to discriminate the disease condition of the patients with SFTS. Moreover, HCT-ALB served as a predictor of SFTS prognosis. Additionally, an area under the ROC curve (AUC) of 0.777 and a critical HCT-ALB value of 4.75 on day 7 were associated with a sensitivity of 83.3% and a specificity of 73.9%. On day 8 (AUC = 0.882), the critical value of HCT-ALB was 9.25, while the sensitivity was 100% and specificity was 76.5%. Further verification based on the data of 91 patients with SFTS admitted to the First Affiliated Hospital of Nanjing Medical University demonstrated a mortality rate of 51% (24/47) among those with HCT-ALB values >4.75 on day 7 of the disease course, highlighting the potential of the HCT-ALB value of >4.75 for predicting SFTS prognosis. High HCT-ALB values are closely related to the mortality of patients with SFTS. HCT-ALB is a sensitive and independent predictor of early disease in patients with SFTS.

Enhancing readmission prediction models by integrating insights from home healthcare notes: Retrospective cohort study

BackgroundHospital readmission is an important indicator of inpatient care quality and a significant driver of increasing medical costs. Therefore, it is important to explore the effects of postdischarge information, particularly from home healthcare notes, on enhancing readmission prediction models. Despite the use of natural language processing (NLP) and machine learning in prediction model development, current studies often overlook insights from home healthcare notes. ObjectiveThis study aimed to develop prediction models for 30-day readmissions using home healthcare notes and structured data. In addition, it explored the development of 14- and 180-day prediction models using variables in the 30-day model. DesignA retrospective observational cohort study. Setting(s)This study was conducted at Ajou University School of Medicine in South Korea. ParticipantsData from electronic health records, encompassing demographic characteristics of 1819 participants, along with information on conditions, drug, and home healthcare, were utilized. MethodsTwo distinct models were developed for each prediction window (30-, 14-, 180-day): the traditional model, which utilized structured variables alone, and the common data model (CDM)-NLP model, which incorporated structured and topic variables extracted from home healthcare notes. BERTopic facilitated topic generation and risk probability, representing the likelihood of documents being assigned to specific topics. Feature selection involved experimenting with various algorithms. The best-performing algorithm, determined using the area under the receiver operating characteristic curve (AUROC), was used for model development. Model performance was assessed using various learning metrics including AUROC. ResultsAmong 1819 patients, 251 (13.80 %) experienced 30-day readmission. The least absolute shrinkage and selection operator was used for feature extraction and model development. The 15 structured features were used in the traditional model. Moreover, five additional topic variables from the home healthcare notes were applied in the CDM-NLP model. The AUROC of the traditional model was 0.739 (95 % CI: 0.672–0.807). The AUROC of the CDM-NLP model was high at 0.824 (95 % CI: 0.768–0.880), which indicated an outstanding performance. The topics in the CDM-NLP model included emotional distress, daily living functions, nutrition, postoperative status, and cardiorespiratory issues. In extended prediction model development for 14- and 180-day readmissions, the CDM-NLP consistently outperformed the traditional model. ConclusionsThis study developed effective prediction models using both structured and unstructured data, thereby emphasizing the significance of postdischarge information from home healthcare notes in readmission prediction.

Comparison of pulse pressure and stroke volume variations measured by three monitors in high-risk surgical patients

IntroductionDynamic indices of fluid responsiveness (FR) such as pulse pressure variation (PPV) and stroke volume variation (SVV) differ among hemodynamic monitors, which use proprietary algorithms, and vary even over a short period of time. We aimed to compare the baseline values, fluctuation and predictive value for FR of PPV and SVV measured by three minimally invasive monitors. Patients and methodsTwenty patients undergoing high-risk abdominal surgery were included and 45 fluid challenges were analysed. The patients were simultaneously monitored using Carescape B650, LiDCO Rapid and FloTrac/Vigileo system. Cardiac output (CO), PPV and SVV were recorded before and after the fluid challenge of 500ml of balanced crystalloid solution. An increase in CO ≥ 15 % defined fluid responders. Concurrently recorded arterial waveform was used for offline calculation of PPV. ResultsMean baseline values of the indices ranged between 8.6% and 13.4%. LiDCO showed higher fluctuation of indices compared to the other monitors. Area under the receiver operating characteristic curve (AUROC) ranged from 0.71 to 0.76 with optimal cut-off value between 7.5% and 13.9%. AUROC increased for all indices when FR was defined as an increase in stroke volume. Furthermore, a decrease in PPV or SVV after fluid challenge (ΔPPV, ΔSVV) proved a better marker of FR (AUROC 0.82 – 0.93) than baseline values with a uniform threshold of approximately -3%. ConclusionsAlthough a significant range of baselines variations and optimal cut-off values was observed, the predictive value of PPV and SVV from all the monitors was only moderate and comparable. Nevertheless, ΔPPV and ΔSVV appear to be a reliable and device-independent markers of FR.

Coagulation profiles and percentiles in neonates with hypoxic-ischemic encephalopathy undergoing therapeutic hypothermia: A step toward more accurate transfusion thresholds.

In the literature, there are no studies about the transfusion threshold for neonates with hypoxic-ischemic encephalopathy (HIE) undergoing therapeutic hypothermia (TH). In order to facilitate accurate interpretation of coagulation results in these neonates, we aimed to generate specific reference intervals in this specific population. This retrospective study included all HIE neonates admitted from 2014 to 2022 to undergo TH. All infants during TH underwent blood exams, including the coagulation profile. Our primary outcome was to assess the estimates of the 3rd, 10th, 25th, 50th, 75th, 90th, and 97th percentiles for each parameter on admission (before transfusion). By the receiver operating characteristic (ROC) analysis, the area under the ROC curve (AUC) and the best cut-off point were used to evaluate the ability of the prothrombin time expressed as international normalized ratio (PT-INR) to predict the risk of any bleeding. A total of 143 infants were included in this study. On admission, the median fibrinogen value was 205mg/dL, prothrombin time 18.6seconds, PT-INR 1.50, activated partial thromboplastin time 38.3seconds, thrombin time 18.6seconds, antithrombin 57.0%. The optimal cut-off of PT-INR in predicting the risk of any bleeding was greater than 1.84 (AUC.623, p=.024). For the first time, we proposed the percentiles of coagulation parameters in our cohort of neonates with HIE. Furthermore, we found that a PT-INR greater than 1.84 can significantly predict the risk of any bleeding. Further studies are needed to determine if a restrictive versus a liberal transfusion approach can be equally safer for these high-risk infants.

Contrast-Enhanced Ultrasound-Based Radiomics for the Prediction of Axillary Lymph Nodes Status in Breast Cancer.

Breast cancer is the leading cause of cancer-related deaths in the female population. Axillary lymph nodes (ALN) are a group of the most common metastatic sites of breast cancer. Timely assessment of ALN status is of paramount clinical importance for medical decision making. To utilize contrast-enhanced ultrasound (CEUS)-based radiomics models for noninvasive pretreatment prediction of ALN status. Clinical data and pretreatment CEUS images of primary breast tumors were retrospectively studied to build radiomics signatures for pretreatment prediction of nodal status between May 2015 and July 2021. The cases were divided into the training cohorts and test cohorts in a 9:1 ratio. The mRMR approach and stepwise forward logistic regression technique were used for feature selection, followed by the multivariate logistic regression technique for building radiomics signatures in the training cohort. The confusion matrix and receiver operating characteristic (ROC) analysis were used for accessing the prediction efficacy of the radiomics models. The radiomics models, which consist of six features, achieved predictive accuracy with the area under the ROC curve (AUC) of 0.713 in the test set for predicting lymph node metastasis. The CEUS-based radiomics is promising to be developed as a reliable noninvasive tool for predicting ALN status.

Clinical Diagnostic Significance of Combined Measurement of Lipoprotein(a) and Neck Circumference in Patients with Coronary Heart Disease.

The study aimed to explore the clinical diagnostic significance of lipoprotein(a) [Lp(a)] and neck circumference (NC) in patients with coronary heart disease (CHD). This cross-sectional study was conducted at Chengde Central Hospital from September 2021 to June 2023, enrolling 791 patients with suspected CHD who underwent selective coronary angiography (CAG). Patients were categorized into CHD and non-CHD groups based on the severity of arterial narrowing. Subsequently, the diagnostic value of Lp(a) combined with NC in patients with CHD was assessed using receiver operating characteristic (ROC) curves. Based on the results of multivariate logistic regression, a nomogram was constructed, and its clinical applicability was validated using decision curve analysis (DCA) and clinical impact curve (CIC). Multivariate logistic regression proved that high Lp(a) and high NC are risk factors for CHD, with OR of 1.836 (95% CI: 1.282-2.630) and 1.383 (1.0.978-1.955), respectively. Patients in the high NC or Lp(a) group exhibited a higher prevalence of multi-vessel disease. The area under the ROC curve (AUC) of the predictive model combining high Lp(a) and high NC was 0.710 (95% CI: 0.670-0.751) and also demonstrated good calibration (Hosmer-Lemeshow goodness-of-fit test P value=0.494). The DCA and CIC confirmed the clinical utility of the nomogram developed to predict CHD based on the combination of high Lp(a) and high NC. The levels of Lp(a) and NC exhibit a significant correlation with the presence of CHD, and their combined assessment holds specific clinical value in the diagnosis of CHD.

A Comprehensive Analysis Exploring the Vital Role of the Systemic Immune-Inflammatory Index Upon Admission in Severe Hemorrhagic Fever with Renal Syndrome.

To explore the value of the systemic immune-inflammatory index (SII) and the systemic inflammatory response index (SIRI) in patients with severe hemorrhagic fever with renal syndrome (HFRS) upon admission. This study included a total of 165 patients with HFRS, who were divided into mild and severe groups based on the severity of the disease. By reviewing medical records, we collected the white blood cell (WBC), SII, and SIRI values of patients upon admission. Univariate and multivariate logistics regression analyses were performed to identify risk factors for severe HFRS. The receiver operating characteristic (ROC) curve was applied to calculate the area under the ROC curve (AUC) to analyze the predictive value of SII and SIRI for severe HFRS, and the results were compared with WBC and SIRI. Compared with the mild HFRS group, patients in the severe HFRS group had a longer duration of illness (P < 0.05), higher levels of WBC, neutrophil (NEUT), lymphocyte (LYMP), monocyte (MONO), procalcitonin (PCT), SIRI, alanine transaminase (ALT), and creatinine (Scr) (P < 0.05), while lower levels of ALB, platelet (PLT), platelet-to-lymphocyte rate (PLR), and SII, with statistically significant differences (P < 0.05). Binary logistics regression analysis indicated that WBC (OR: 1.190, 95% CI: 1.032-1.371), SII (OR: 0.967, 95% CI: 0.951-0.984), and SIRI (OR: 4.743, 95% CI: 2.077-10.830) were risk factors for severe HFRS. The AUCs of WBC, SII, and SIRI for predicting severe HFRS were 0.765, 0.803, and 0.785, respectively. Low levels of SII and high levels of WBC and SIRI upon admission are risk factors for severe HFRS and have certain value in predicting the progression of HFRS to severe cases, among which SII exhibits the best predictive value.

Mortality rate analysis of patients on invasive mechanical ventilation in the intensive care unit on day 28.

Outcomes in patients receiving invasive mechanical ventilation (IMV) are currently unclear. The present study aimed to explore the prognostic factors of the mortality rate on day 28 in patients treated in the intensive care unit (ICU) and undergoing IMV. The IMV Mortality Prediction Score (IMPRES) of 129 patients in the ICU receiving IMV after emergency (or selective) endotracheal intubation from March 2018 to August 2020 was calculated. The patients were divided into survival (n=73) and death groups (n=56) on day 28. The predictive factors of independent and combined mortality rates were determined using a receiver operating characteristic (ROC) curve and the area under the ROC curve (AUC). The AUC of the IMPRES for predicting patient death on day 28 was 0.785 (95% confidence interval (CI): 0.704-0.864, P<0.01). When the IMPRES cut-off was 4.50, the Youden index was at its maximum (0.487) with a sensitivity of 85.7% and a specificity of 63.0%. The AUC of the ventilator use time (days) at 12.5 days cut-off was 0.653 (95% CI: 0.56-0.746, P<0.01), the Youden index was 0.235 with a sensitivity of 52.1% and a specificity of 71.4%. The AUC of the IMPRES combined with the duration of ventilator use was 0.856 (95% CI: 0.789-0.922, P<0.001), the Youden index was 0.635 with a sensitivity of 84.9% and a specificity of 78.6%. The IMPRES was observed to be the main factor influencing the mortality rate of patients receiving IMV at the ICU on day 28, and the IMPRES combined with the duration of ventilator use had a significant predictive value for the 28-day mortality rates of these patients.

Clinical Crrelation and Prognostic Analysis of ALBI Score in Secondary Hemophagocytic Syndrome in Children

To explore the clinical correlation and prognostic value of the Albumin-Bilirubin (ALBI) score in children with secondary hemophagocytic syndrome(sHLH). A retrospective analysis was conducted on the data of children's sHLH cases clearly diagnosed in the Affiliated Hospital of Zunyi Medical University from January 2012 to March 2023. Survival analysis was conducted according to the ALBI classification. Spearman correlation analysis was conducted between the ALBI score and clinical indicators. The Receiver Operating Characteristic(ROC) curve was used to evaluate the ALBI score, select the best cutoff value, and evaluate the accuracy of prognostic prediction value. Kaplan-Meier method was used to draw the survival curve. Log-rank method was used to compare the differences of survival curve between groups. Cox regression was used for prognostic analysis and restricted cubic spline curves used to calculate the relationship between ALBI scores and the risk of death in children with sHLH. A total of 128 children with sHLH were included in this study, with a median age of 38(13.25, 84) months. There were 70 males (54.69%) and 58 females (45.31%). The survival analysis results of ALBI grading showed that the survival rate of HLH patients with ALBI grade 3 was significantly lower than those with ALBI grades 1 and 2. Spearman correlation analysis results showed that ALBI score was positively correlated with splenomegaly, respiratory failure, disseminated intravascular coagulation(DIC), pulmonary hemorrhage, gastrointestinal hemorrhage, central nervous system involvement, ALT, AST, TG, LDH, PT, APTT, and SF (the correlation coefficients are: r =0.181, 0.362, 0.332, 0.221, 0.351, 0.347, 0.391, 0.563, 0.180, 0.448, 0.483, 0.37, 0.356), and was negatively correlated with HB, PLT, and FIB (the correlation coefficients are: r =-0.321, -0.316, -0.423), but was not significantly correlated with EBV infection, fungal infection, hepatomegaly, and ANC (P ＞0.05). Using the ROC curve, the cutoff value of ALBI was -1.76. Single factor Cox regression analysis results showed that HB< 90 g/L, ALT≥80 U/L, AST≥200 U/L, LDH≥1 000 U/L, PT≥20 s, APTT≥40 s, FIB< 1.5 g/L, ALBI≥-1.76, combined pulmonary hemorrhage, DIC, central nervous system involvement, gastrointestinal bleeding, and not using blood purification may be the prognostic risk factors for children with sHLH (P < 0.05). Multivariate Cox regression results showed that FIB< 1.5 g/L (HR =2.119, 95% CI :1.028-4.368), ALBI≥-1.76 (HR =2.452, 95%CI :1.233-4.875), and central nervous system involvement (HR=4.674, 95%CI :2.486-8.789) were independent risk factors affecting prognosis, while blood purification (HR =0.306, 95%CI :0.153-0.612) was an independent protective factor for prognosis. The application of restricted cubic splines shows that the risk of death increases with the increase of ALBI score. The area under the ROC curve (AUC) of the ALBI score for predicting the risk of 1-week, 2-week, 4-week, and overall mortality were 0.825, 0.807, 0.700, and 0.693, respectively, indicating good predictive performance for early mortality risk. According to subgroup analysis results of clinical manifestations, compared with the ALBI < -1.76 group, ALBI≥-1.76 was associated with age ≤2 years, EBV infection, HLH-1994/2004 treatment, concomitant respiratory failure, and ANC≤1.0×10 9/L, HB< 90 g/L, PLT < 100×109/L, TG≥3.0 mmol/L, LDH≥1 000 U/L, APTT≥40 s, and FIB< 1.5 g/L (P < 0.05). The ALBI score is related to the clinical characteristics and laboratory indicators of sHLH, and can be used as a beneficial indicator for assessing the prognostic risk of sHLH in children. It has good accuracy and clinical application value in predicting the prognosis of sHLH in children.

Utility of Sonoelastography and Magnetic Resonance Imaging in Characterization of Thyroid Nodules.

To evaluate thyroid nodules with sonoelastography and magnetic resonance imaging (MRI). The study included 28 patients with 40 thyroid nodules. Clearance was obtained from the institute's ethical clearance committee. Patients with pure cystic nodules or nodules with eggshell calcification, diffuse thyroid pathology (such as Graves' disease, Hashimoto's thyroiditis, De Quervain thyroiditis, and Riedel's thyroiditis), inaccessible nodules via fine needle aspiration cytology (FNAC), or patients with a history of thyroid gland surgery were excluded from the study. Strain elastography was performed on a Phillips iU22 machine, producing qualitative color-coded strain maps (graded using the Rago 5-point system) and semiquantitative strain ratios. MRI was performed on a Phillips ACHIEVA 1.5T magnet with a head and neck coil. Rago scores statistically correlated (χ2 = 18.052, p < 0.001) with malignant nodules, and using the receiver operating characteristic (ROC) curve, the area under the ROC curve (AUROC) for the mean strain ratio predicting malignant outcomes was 0.88 [95% confidence interval (CI): 0.767-0.992], which was also statistically significant (p < 0.001). A cutoff of mean strain ratio ≥2.48 predicted malignant outcomes with 100% specificity. T2 signal intensity ratio (SIR) and apparent diffusion coefficient (ADC) values were not statistically significant in predicting malignant outcomes. Kinetic curves were statistically significant for Rago scores (χ2 = 11.356, p = 0.045); however, no significant difference was found in predicting malignant outcomes. We concluded that sonoelastography, along with grayscale ultrasound, is a useful noninvasive technique for predicting histological outcomes. However, MRI should largely be reserved as a problem-solving tool rather than a standalone imaging modality. The kinetic curves show some degree of overlap between histologically distinct diseases, and thus large-scale multicenter trials are needed for further standardization.

Deep learning model with pathological knowledge for detection of colorectal neuroendocrine tumor

Colorectal neuroendocrine tumors (NETs) differ significantly from colorectal carcinoma (CRC) in terms of treatment strategy and prognosis, necessitating a cost-effective approach for accurate discrimination. Here, we propose an approach for distinguishing between colorectal NET and CRC based on pathological images by utilizing pathological prior information to facilitate the generation of robust slide-level features. By calculating the similarity between morphological descriptions and patches, our approach selects only 2% of the diagnostically relevant patches for both training and inference, achieving an area under the receiver operating characteristic curve (AUROC) of 0.9974 on the internal dataset, and AUROCs of 0.9724 and 0.9513 on two external datasets. Our model effectively identifies NETs from CRCs, reducing unnecessary immunohistochemical tests and enhancing the precise treatment for patients with colorectal tumors. Our approach also enables researchers to investigate methods with high accuracy and low computational complexity, thereby advancing the application of artificial intelligence in clinical settings.

A heterogeneous graph transformer framework for accurate cancer driver gene prediction and downstream analysis

Accurately predicting cancer driver genes remains a formidable challenge amidst the burgeoning volume and intricacy of cancer genomic data. In this investigation, we propose HGTDG, an innovative heterogeneous graph transformer framework tailored for precisely predicting cancer driver genes and exploring downstream tasks. A heterogeneous graph construction module is central to the framework, which assembles a gene-protein heterogeneous network leveraging the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways and protein-protein interactions sourced from the STRING (search tool for recurring instances of neighboring genes) database. Moreover, our framework introduces a pioneering heterogeneous graph transformer module, harnessing multi-head attention mechanisms for nuanced node embedding. This transformative module proficiently captures distinct representations for both nodes and edges, thereby enriching the model's predictive capacity. Subsequently, the generated node embeddings are seamlessly integrated into a classification module, facilitating the discrimination between driver and non-driver genes. Our experimental findings evince the superiority of HGTDG over existing methodologies, as evidenced by the enhanced performance metrics, including the area under the receiver operating characteristic curves (AUROC) and the area under the precision-recall curves (AUPRC). Furthermore, the downstream analysis utilizing the newly identified cancer driver genes underscores the efficacy and versatility of our proposed framework.

Development and Validation of a Novel Nomogram for Predicting Mechanical Ventilation After Cervical Spinal Cord Injury

ObjectiveTo investigate the risk factors relating to the need for mechanical ventilation (MV) in isolated patients with cervical spinal cord injury (cSCI) and to construct a nomogram prediction model. DesignRetrospective analysis study. SettingNational Spinal Cord Injury Model System Database (NSCID) observation data were initially collected during rehabilitation hospitalization. ParticipantsA total of 5784 patients (N=5784) who had a cSCI were admitted to the NSCID between 2006 and 2021. InterventionsNot applicable. Main Outcome Measure(s)A univariate and multivariate logistic regression analysis was used to identify the independent factors affecting the use of MV in patients with cSCI, and these independent influencing factors were used to develop a nomogram prediction model. The area under the receiver operating characteristic curve (AUROC), calibration curve, and decision curve analysis (DCA) were used to evaluate the efficiency and the clinical application value of the model, respectively. ResultsIn a series of 5784 included patients, 926 cases (16.0%) were admitted to spinal cord model system inpatient rehabilitation with the need for MV. Logistic regression analysis demonstrated that associated injury, American Spinal Cord Injury Association Impairment Scale (AIS), the sum of unilateral optimal motor scores for each muscle segment of upper extremities (sUEM), and neurologic level of injury (NLI) were independent predictors for the use of MV (P<.05). The prediction nomogram of MV usage in patients with cSCI was established based on the above independent predictors. The AUROC of the training set, internal verification set, and external verification set were 0.871 (0.857-0.886), 0.867 (0.843-0.891), and 0.850 (0.824-0.875), respectively. The calibration curve and DCA results showed that the model had good calibration and clinical practicability. ConclusionsThe nomograph prediction model based on sUEM, NLI, associated injury, and AIS can accurately and effectively predict the risk of MV in patients with cSCI, to help clinicians screen high-risk patients and formulate targeted intervention measures.