Receiver Operating Characteristic Curve Values Research Articles

Neural Networks Detect Inter-Turn Short Circuit Faults Using Inverter Switching Statistics for a Closed-Loop Controlled Motor Drive

Early detection of an inter-turn short circuit fault (ISCF) can reduce repair costs and downtime of an electrical machine. In an induction machine (IM) driven by an inverter with a model predictive control (MPC) algorithm, the controller outputs are influenced by a fault due to the fault-controller interaction. Based on this observation, this study developed neural network models using inverter switching statistics to detect the ISCF of an IM. The method was non-invasive, and it did not require any additional sensors. In the fault detection task, an area under receiver operating characteristics curve value of 0.9950 (95% Confidence IntervaI: 0.9949 - 0.9951) was obtained. At the rated operating conditions, the neural network model detected and located an ISCF of 2-turns (out of 104 turns per phase) under 0.1 seconds, a speedup of more than two times compared to the thresholding-based method. Moreover, we published the switching vector data collected at various load torque and shaft speed values for healthy and faulty states of the IM, becoming the first publicly available ISCF detection dataset. Together with the dataset, we provided performance baselines for three main neural network architectures, namely, multi-layer perceptron, convolutional neural network, and recurrent neural network.

Machine Learning to Predict Risk of Intracranial Hemorrhage in ICU Patients Receiving Direct Oral Anticoagulants

INTRODUCTION The direct oral anticoagulants (DOACs) given to patients in the intensive care unit (ICU) are associated with various clinical outcomes, such as major bleeding episodes or intracranial hemorrhages (ICH). These pose major challenges for managing patients in ICUs. There is a need for early detection and intervention among patients taking DOACs to prevent adverse clinical outcomes. We hypothesized that machine-learning algorithms could be applied to develop a more accurate and user-friendly diagnostic tool that integrates various clinical and laboratory data and considers complex interactions. METHODS This was an observational, retrospective study using patients from the MIMIC-IV database. The study included patients who were admitted to the ICU with a DOAC (apixaban, rivaroxaban, dabigatran, or edoxaban) listed as one of the active medications at the time of hospital admission. ICD-9 and ICD-10 codes were utilized to identify patients with ICH (I61, 431). Descriptive statistics were reported for all variables of interest. Various machine-learning models were utilized to predict the outcome variable of ICH. Patients who had expired as discharge dispositions were excluded. Models accounted for variables from the demographic, laboratory, comorbidity, etc. domains. Model performance was evaluated by comparing the accuracy and Area Under the Receiver Operating Characteristic Curve (AUROC) values. RESULTS Overall, there were 3597 cases that met the inclusion criteria. Among these, 1517 (42.2%) were female. The majority of the sample was White (70.5%), and 2093 (58.2%) cases had Medicare insurance. The median [Q1, Q3] age of the sample was 74 [65, 83] years. ICH was reported in 245 (6.8%) cases. ICU mortality was reported in 297 cases (8.2%). The proportion of ER admissions among those cases who did not have ICH was 40.7%, and 64.1% among those who had ICH in the ICU (p < 0.05). ICU mortality among patients with an ICH was 24.1% vs. 7.1% among those without an ICH (p < 0.001). There were more females in the ICH group (49.8% vs. 41.6%; p < 0.015), and the patients were older (median age 76 vs. 74 years; p < 0.001). The machine learning model results showed that the XGBoost classifier performed best in predicting mortality in the ICU, with an AUROC value of 91.4%. The top ten variables associated with ICH by feature importance using the ‘Gain’ metric, which measures the improvement in accuracy brought by a feature to the branches it is on, were ICU length of stay, age, heart rate, respiratory rate, oxygen saturation, anion gap, mean arterial blood pressure, temperature, blood urea nitrogen, and platelet count. CONCLUSION The XGBoost classifier algorithm improved discrimination and calibration compared to other ML models for predicting ICH in the ICU among patients taking DOACs. It has the potential to reduce delayed diagnosis and overtreatment in clinical practice. Future studies will validate this model in broader clinical settings.

Prediction of Hypoglycemia From Continuous Glucose Monitoring in Insulin-Treated Patients With Type 2 Diabetes Using Transfer Learning on Type 1 Diabetes Data: A Deep Transfer Learning Approach.

Hypoglycemia is common in insulin-treated type 2 diabetes (T2D) patients, which can lead to decreased quality of life or premature death. Deep learning models offer promise of accurate predictions, but data scarcity poses a challenge. This study aims to develop a deep learning model utilizing transfer learning to predict hypoglycemia. Continuous glucose monitoring (CGM) data from 226 patients with type 1 diabetes (T1D) and 180 patients with T2D were utilized. Data were structured into one-hour samples and labeled as hypoglycemia or not depending on whether three consecutive CGM values were below 3.9 [mmol/L] (70 mg/dL) one hour after the sample. A convolutional neural network (CNN) was pre-trained with the T1D data set and subsequently fitted using a T2D data set, all while being optimized toward maximizing the area under the receiver operating characteristics curve (AUC) value, and it was externally validated on a separate T2D data set. The developed model was externally validated with 334 711 one-hour CGM samples, of which 15 695 (4.69%) were labeled as hypoglycemic. The model achieved an AUC of 0.941 and a positive predictive value of 40.49% at a specificity of 95% and a sensitivity of 69.16%. The transfer learned CNN model showed promising performance in predicting hypoglycemic episodes and with slightly better results than a non-transfer learned CNN model.

Machine Learning to Predict Risk of Venous Thromboembolism Among Patients with Myeloproliferative Neoplasms in the Intensive Care Unit

INTRODUCTION Myeloproliferative neoplasms (MPN) increase the risk of venous thromboembolism (VTE) and constitute important thrombotic risk factors. VTE risk is especially high if these patients require the intensive care unit (ICU). Precise and user-friendly diagnostic tools that incorporate complex interactions between clinical data and laboratory results to predict VTE among patients with MPN in the ICU are lacking. We hypothesize that machine-learning algorithms could be used to predict VTE and guide thromboprophylaxis in this population. METHODS This was an observational, retrospective study using routine clinical practice data from MIMIC-IV, a freely accessible electronic health record database. We analyzed patients with an MPN diagnosis who were admitted to the ICU. ICD-9 and ICD-10 codes were utilized to identify incident VTE events during ICU stay (I82X, 453X). Descriptive statistics were reported for all variables of interest. Various machine-learning models were utilized to predict the incident VTE as the outcome variable. Patients who had expired as discharge dispositions were excluded. Models accounted for demographic characteristics, laboratory results, comorbidity burden, and other domains. Model performance was evaluated by comparing the accuracy and Area Under the Receiver Operating Characteristic Curve (AUROC) values. RESULTS We analyzed 378 cases that met the inclusion criteria. Among these, 160 (42.3%) were female. The majority were White (65.9%), and 151 (39.9%) had Medicare insurance. The median [Q1, Q3] age of the sample was 64.0 [52.2, 76.8] years. VTE occurred in 27 patients (7.1% incidence). ICU mortality was reported in 26 (6.9%) cases. The proportion of ER admissions was 48.1% among those cases who had VTE and 47.6% among those who didn't have VTE in the ICU (p > 0.05). Patients who developed VTE were more likely males (66.7% vs. 33.3%), but there was no difference in age (median age 61 vs. 64 years; p > 0.05). The machine learning model results showed that the Naive Bayes classifier performed best in predicting VTE in the ICU among patients with MPNs, with an AUROC value of 84.36%, followed by the deep learning model, with an AUROC value of 68.3%. The deep learning model showed that the top ten variables associated with VTE by feature importance were oxygen saturation, length of stay, blood urea nitrogen, other insurance type, anion gap, respiratory rate, bicarbonate, pressure of oxygen, pH value, and race other than White. CONCLUSION Predicting VTE in the ICU among patients with MPNs may be achieved by machine learning methods. The deep learning algorithm outperformed other models in terms of discrimination and calibration. If validated and applied in clinical practice, these methods may optimize thromboprophylaxis strategies. Future research is needed to test this model in more extensive clinical settings.

Time-domain heart rate variability features for automatic congestive heart failure prediction.

Heart failure is a serious condition that often goes undiagnosed in primary care due to the lack of reliable diagnostic tools and the similarity of its symptoms with other diseases. Non-invasive monitoring of heart rate variability (HRV), which reflects the activity of the autonomic nervous system, could offer a novel and accurate way to detect and manage heart failure patients. This study aimed to assess the feasibility of using machine learning techniques on HRV data as a non-invasive biomarker to classify healthy adults and those with heart failure. We used digitized electrocardiogram recordings from 54 adults with normal sinus rhythm and 44 adults categorized into New York Heart Association classes 1, 2, and 3, suffering from congestive heart failure. All recordings were sourced from the PhysioNet database. Following data pre-processing, we performed time-domain HRV analysis on all individual recordings, including root mean square of the successive difference in adjacent RR interval (RRi) (RMSSD), the standard deviation of RRi (SDNN, the NN stands for natural or sinus intervals), the standard deviation of the successive differences between successive RRi (SDSD), the number or percentage of RRi longer than 50ms (NN50 and pNN50), and the average value of RRi [mean RR interval (mRRi)]. In our experimental classification performance evaluation, on the computed HRV parameters, we optimized hyperparameters and performed five-fold cross-validation using four machine learning classification algorithms: support vector machine, k-nearest neighbour (KNN), naïve Bayes, and decision tree (DT). We evaluated the prediction accuracy of these models using performance criteria, namely, precision, recall, specificity, F1 score, and overall accuracy. For added insight, we also presented receiver operating characteristic (ROC) plots and area under the ROC curve (AUC) values. The overall best performance accuracy of 77% was achieved when KNN and DT were trained on computed HRV parameters with a 5min time window. KNN obtained an AUC of 0.77, while DT attained 0.78. Additionally, in the classification of severe congestive heart failure, KNN and DT had the best accuracy of 91%, with KNN achieving an AUC of 0.88 and DT obtaining 0.92. The results show that HRV can accurately predict severe congestive heart failure. The findings of this study could inform the use of machine learning approaches on non-invasive HRV, to screen congestive heart failure individuals in primary care.

Sudomotor dysfunction is associated with impaired left ventricular diastolic function in persons with type 2 diabetes: a cross-sectional study.

The incidence of preserved ejection fraction heart failure has significantly increased in persons with type 2 diabetes mellitus (T2DM). Left ventricular (LV) diastolic dysfunction is an early and important manifestation of preserved ejection fraction heart failure. The onset of heart failure in persons with diabetes is associated with diabetic neuropathy. However, the relationship among sudomotor function, which is an early manifestation of small fiber neuropathy, and LV diastolic function remains unclear. This study aimed to explore the association between sudomotor function and LV diastolic function in persons with T2DM. In total, 699 persons with T2DM were enrolled and divided into three groups according to electrochemical skin conductance (ESC) assessed using the SUDOSCAN device: "no dysfunction" group (NSF), "moderate dysfunction" group (MDF), and "severe dysfunction" group (SDF). LV diastolic function was assessed using Doppler echocardiography. To evaluate the relationship between ESC and echocardiographic parameters, Pearson's correlation analysis was performed. Additionally, logistic regression analysis was used to determine the association between LV diastolic function and ESC. A receiver operating characteristic (ROC) curve was constructed to evaluate the performance of sudomotor function indicators in detecting impaired cardiac diastolic function. There were 301 persons (43.06%) in the NSF group, 232 (33.19%) in the MDF group, and 166 (23.75%) in the SDF group. Compared to the NSF group, the MDF and SDF groups had higher A and E/e' and lower e' values (all p < 0.05). Pearson's correlation analysis showed that A and E/e' were negatively associated with foot ESC (FESC) and hand ESC (HESC), whereas e' was positively associated with FESC and HESC (all p < 0.05). After adjusting for confounding factors, binary logistic regression analysis showed that ESC was independently associated with impaired LV diastolic function (p = 0.003). The area under the ROC curve values for FESC and HESC were 0.621 and 0.635, respectively (both p < 0.05). Deteriorating sudomotor function was associated with reduced diastolic function indicators. ESC can be used as a biomarker for detecting LV diastolic impairment.

Online interpretable dynamic prediction models for postoperative delirium after cardiac surgery under cardiopulmonary bypass developed based on machine learning algorithms: A retrospective cohort study

ObjectivePostoperative delirium (POD) is strongly associated with poor early and long-term prognosis in cardiac surgery patients with cardiopulmonary bypass (CPB). This study aimed to develop dynamic prediction models for POD after cardiac surgery under CPB using machine learning (ML) algorithms. MethodsFrom July 2021 to June 2022, clinical data were collected from patients undergoing cardiac surgery under CPB at Nanjing First Hospital. A dataset from the same center (October 2022 to November 2022) was also used for temporal external validation. We used ML and deep learning to build models in the training set, optimized parameters in the test set, and finally validated the best model in the validation set. The SHapley Additive exPlanations (SHAP) method was introduced to explain the best models. ResultsOf the 885 patients enrolled, 221 (25.0%) developed POD. 22 (22.0%) of 100 validation cohort patients developed POD. The preoperative and postoperative artificial neural network (ANN) models exhibited optimal performance. The validation results demonstrated satisfactory predictive performance of the ANN model, with area under the receiver operator characteristic curve (AUROC) values of 0.776 and 0.684 for the preoperative and postoperative models, respectively. Based on the ANN algorithm, we constructed dynamic, highly accurate, and interpretable web risk calculators for POD. ConclusionsWe successfully developed online interpretable dynamic ANN models as clinical decision aids to identify patients at high risk of POD before and after cardiac surgery to facilitate early intervention or care.

Variation trend prediction of ground-level ozone concentrations with high-resolution using landscape pattern data.

Scientifically configuring landscape patterns based on their relationship with ground-level ozone concentrations (GOCs) is an effective way to prevent and control ground-level ozone pollution. In this paper, a GOC variation trend prediction model (hybrid model) combining a generalized linear model (GLM) and a logistic regression model (LRM) was established to analyze the spatiotemporal variation patterns in GOCs as well as their responses to landscape patterns. The model exhibited satisfactory performance, with percent of samples correctly predicted (PCP) value of 82.33% and area under receiver operating characteristics curve (AUC) value of 0.70. Using the hybrid model, the per-pixel rise probability of annual average GOCs at a spatial resolution of 1 km in Shenzhen were generated. The results showed that (1) annual average GOCs were increasing in Shenzhen from 2015 to 2020, and had obvious spatial differences, with a higher value in the west and a lower value in the east; (2) variation trend in GOCs was significant positively correlated with landscape heterogeneity (HET), while significant negatively correlated with dominance (DMG) and contagion (CON); (3) GOCs in Shenzhen has a great risk of rising, especially in GuangMing, PingShan, LongGang, LuoHu and BaoAn. The results provide not only a preliminary index for estimating the GOC variation trend in the absence of air quality monitoring data but also guidance for landscape optimizing design from the perspective of controlling ground-level ozone pollution.

Machine Learning to Detect Disparities in ICU Mortality Among Patients Receiving Direct Oral Anticoagulants

INTRODUCTION The current prescribing practices of direct oral anticoagulants (DOACs) in intensive care unit (ICU) patients and the associated clinical outcomes, including the incidence of major bleeding episodes and mortality, are significant clinical challenges. There is a need for early detection and intervention among patients taking DOACs to prevent adverse clinical outcomes. We hypothesized that machine-learning algorithms could be used to create a more precise and user-friendly diagnostic tool that incorporates a variety of clinical and laboratory data and takes complex interactions into account. METHODS This was an observational, retrospective study using routine clinical practice data from MIMIC-IV, a freely accessible electronic health record database. The study included patients who were admitted to the ICU with a DOAC (apixaban, rivaroxaban, dabigatran, or edoxaban) listed as one of the active medications at the time of hospital admission. ICD-9 and ICD-10 codes were utilized to identify cases (Z7901, V5861). Descriptive statistics were reported for all variables of interest. Various machine-learning models were utilized to predict the outcome variable of ICU mortality after accounting for variables from the demographic, laboratory, comorbidity, etc. domains. Model performance was evaluated by comparing the accuracy and Area Under the Receiver Operating Characteristic Curve (AUROC) values. RESULTS Overall, there were 3597 cases that met the inclusion criteria. Among these, 1517 (42.2%) were female. The majority of the sample was White (70.5%), and 2093 (58.2%) cases had Medicare insurance. The median [Q1, Q3] age of the sample was 74 [65, 83] years. ICU mortality was reported in 297 cases (8.2%). The mean ( SD) ICU length of stay was 3.5 (4.6) days among those cases who were discharged alive and 4.2 (4.7) days among those who expired (p &amp;lt; 0.012). The median blood urea nitrogen among those discharged alive and expired was 20.0 and 29.5 (p&amp;lt;0.001), respectively. Intracranial hemorrhage was reported in 186 (5.6%) and 59 (19.9%) of those discharged alive and expired, respectively (p&amp;lt;0.001). The machine learning model results showed that the Random forest classifier performed best in predicting mortality in the ICU, with an AUROC value of 85.3%. The top ten variables associated with ICU mortality by feature importance were pH, WBC count, platelet count, blood urea nitrogen, heart rate, anion gap, partial pressure of carbon dioxide, oxygen saturation, age, and respiratory rate. CONCLUSION The Random forest classifier algorithm improved discrimination and calibration compared to other ML models for predicting mortality in the ICU among patients taking DOACs. It has the potential to reduce delayed diagnosis and overtreatment in clinical practice. Future studies will validate this model in broader clinical settings.

Development of a prognostic nomogram for lymph node positive HR+/HER2- breast cancer patients: a study of SEER database and a Chinese cohort.

The hormone receptor+/human epidermal growth factor receptor 2- (HR+/HER2-) breast cancer (BC) patients account for the largest proportion in all patients and are still at high risk of long-range recurrence. This current study aimed to construct a prognostic nomogram to predict 3-year and 5-year BC-specific survival (BCSS) in HR+/HER2- BC patients with axillary lymph node metastasis. A total of 25,338 HR+/HER2- patients with axillary lymph node-positive BC were enrolled from the Surveillance, Epidemiology and End Results (SEER) database and randomly divided into the training (n=17,738) and validation (n=7,600) cohorts using a ratio of 7:3. Univariate and multivariable Cox regression hazards were used to build a prognostic nomogram based on the training cohort. The nomogram was validated with two independent cohorts. Receiver operating characteristic (ROC) curves and calibration plots were used to evaluate the performance of the model, and Kaplan-Meier survival analyses were applied to test the clinical utility of the risk stratification system. Twelve factors including age, race, marital status, grade, T stage, N stage, radiotherapy, chemotherapy, and metastasis to the bone, brain, lung and liver were identified and incorporated to construct the nomogram (P<0.001). The area under the ROC curve (AUC) values at 3- and 5-year in the training and internal validation sets were 0.800, 0.800, 0.831 and 0.819, respectively, while those of the external set were 0.765 and 0.735, indicating a satisfactory discrimination with our nomogram. The calibration curves showed highly consistent results for the actual and predicted survival probabilities. Furthermore, patients were divided into three risk groups according to the total scores of the nomogram. The risk stratification system accurately differentiated between patients with different BCSS rates. We constructed the first nomogram and corresponding risk stratification system to predict the 3-year and 5-year BCSS for HR+/HER2- patients with lymph node-positive BC, indicating a satisfactory accuracy and clinical application.

Dynamic Nomogram for Subsyndromal Delirium in Adult Intensive Care Unit: A Prospective Cohort Study.

To develop a dynamic nomogram of subsyndromal delirium (SSD) in intensive care unit (ICU) patients and internally validate its efficacy in predicting SSD. Patients who met the inclusion and exclusion criteria in the ICU of a tertiary hospital in Zhejiang from September 2021 to June 2022 were selected as the research objects. The patient data were randomly divided into the training set and validation set according to the ratio of 7:3. The least absolute shrinkage and selection operator (LASSO) and multivariate logistic regression were used to screen the predictors of SSD, and R software was used to construct a dynamic nomogram. Receiver operating characteristic (ROC) curve, calibration band and decision curve were used to evaluate the discrimination, calibration and clinical effectiveness of the model. A total of 1000 eligible patients were included, including 700 in the training set and 300 in the validation set. Age, drinking history, C reactive protein level, APACHE II, indwelling urinary catheter, mechanical ventilation, cerebrovascular disease, respiratory failure, constraint, dexmedetomidine, and propofol were predictors of SSD in ICU patients. The ROC curve values of the training set was 0.902 (95% confidence interval: 0.879-0.925), the best cutoff value was 0.264, the specificity was 78.4%, and the sensitivity was 88.0%. The ROC curve values of the validation set was 0.888 (95% confidence interval: 0.850-0.930), the best cutoff value was 0.543, the specificity was 94.9%, and the sensitivity was 70.9%. The calibration band showed good calibration in the training and validation set. Decision curve analysis showed that the net benefit in the model was significantly high. The dynamic nomogram has good predictive performance, so it is a precise and effective tool for medical staff to predict and manage SSD in the early stage.

Simplified urine-based method to detect rifampin underexposure in adults with tuberculosis: a prospective diagnostic accuracy study.

Variable pharmacokinetics of rifampin in tuberculosis (TB) treatment can lead to poor outcomes. Urine spectrophotometry is simpler and more accessible than recommended serum-based drug monitoring, but its optimal efficacy in predicting serum rifampin underexposure in adults with TB remains uncertain. Adult TB patients in New Jersey and Virginia receiving rifampin-containing regimens were enrolled. Serum and urine samples were collected over 24 h. Rifampin serum concentrations were measured using validated liquid chromatography-tandem mass spectrometry, and total exposure (area under the concentration-time curve) over 24 h (AUC0-24) was determined through noncompartmental analysis. The Sunahara method was used to extract total rifamycins, and rifampin urine excretion was measured by spectrophotometry. An analysis of 58 eligible participants, including 15 (26%) with type 2 diabetes mellitus, demonstrated that urine spectrophotometry accurately identified subtarget rifampin AUC0-24 at 0-4, 0-8, and 0-24 h. The area under the receiver operator characteristic curve (AUC ROC) values were 0.80 (95% CI 0.67-0.90), 0.84 (95% CI 0.72-0.94), and 0.83 (95% CI 0.72-0.93), respectively. These values were comparable to the AUC ROC of 2 h serum concentrations commonly used for therapeutic monitoring (0.82 [95% CI 0.71-0.92], P = 0.6). Diabetes status did not significantly affect the AUC ROCs for urine in predicting subtarget rifampin serum exposure (P = 0.67-0.92). Spectrophotometric measurement of urine rifampin excretion within the first 4 or 8 h after dosing is a simple and cost-effective test that accurately predicts rifampin underexposure. This test provides critical information for optimizing tuberculosis treatment outcomes by facilitating appropriate dose adjustments.

Development and application of a competitive ELISA for the detection of antibodies against Salmonella Abortusequi in equids.

The high abortion rate associated with Salmonella Abortusequi (S. Abortusequi) infection in equids has re-emerged over the past 10 years and has caused serious economic losses to China. Our previous studies showed that the flagellin FljB gene could distinguish S. Abortusequi from most Salmonella serotypes. In this study, the flagellin antigen was used to develop a competitive enzyme-linked immunosorbent assay (cELISA) that could be used to detect both horse and donkey serum samples using a monoclonal antibody (MAb) that was found to bind to FljB. A cELISA was established using the purified MAb coating of the plate and incubation of the mixture of horseradish peroxidase (HRP)-conjugated FljB antigen with the undiluted serum sample. The performance of the cELISA and the tube agglutination test (TAT) assay was compared with respect to sensitivity and specificity, by testing a panel containing 660 S. Abortusequi-positive and 515 S. Abortusequi-negative serum samples, all of which had been characterized by Western blotting. Receiver operator characteristic (ROC) analyses were performed to determine the cutoff value and estimate the detection specificity (Sp) and sensitivity (Se). ROC analysis showed that the area under the ROC curve (AUC) values of cELISA [AUC = 0.9941; 95% confidence interval (CI), 0.9898-0.9984] were higher than those of TAT (AUC = 0.7705; 95% Cl, 0.7437-0.7972). A cutoff value of 39.5% was selected with Sp and Se values of 100 (95% Cl, 99.26-100.00) and 97.58 (95% Cl, 96.10-98.50), respectively. The cELISA has excellent futures compared with TAT, such as shortened detection time, no need for pre-treatment of sera, and easy interpretation of the results, and is more suitable for disease surveillance.

Minimal clinically important differences in walking capacity and physical activity after nonsurgical treatment in patients with lumbar spinal stenosis: a secondary analysis of a randomized controlled trial

BACKGROUND CONTEXTLittle information is available about the minimal clinically important differences (MCIDs) for objective physical measurements in people with lumbar spinal stenosis (LSS). PURPOSETo use disorder-specific anchor and, multiple anchor-, and distribution-based approaches to determine the MCIDs for walking capacity and physical activity in patients with LSS receiving nonsurgical treatment. STUDY DESIGN/SETTINGSecondary analysis of a randomized controlled trial. PATIENT SAMPLESixty-nine patients with neurogenic claudication caused by LSS receiving outpatient physical therapy. OUTCOME MEASURESZurich claudication questionnaire (ZCQ), self-paced walking test (SPWT), and number of daily steps measured by pedometry. METHODSAll patients completed the ZCQ, SPWT, and pedometry at the baseline and after 6 weeks. For the anchor-based approach, ZCQ symptom severity, physical function, and satisfaction subscales were used as the external anchors. Using the receiver-operating characteristic (ROC) curve, the MCIDs were determined based on the optimal cutoff points for changes in the SPWT or daily steps. For the distribution-based approach, the MCIDs were estimated from the standard deviations (SDs) of the baseline scores of the SPWT and daily steps. RESULTSIn the anchor-based approach, only the ZCQ satisfaction subscale for the SPWT (0.73), and ZCQ symptom severity subscale for daily steps (0.71) exceeded the area under the ROC curve value of 0.7, which is considered acceptable. When using these subscales as anchors, the ROC curves and optimal cutoff points indicated MCIDs of 151 m for the SPWT and 1,149 steps for daily steps. The distribution-based approach estimated the MCIDs as 280 m for the SPWT and 1,274 steps for daily steps, and had a moderate effect size (0.5 SD). CONCLUSIONSThe anchor-based approach had limited external responsiveness when the ZCQ was used as the anchor. However, this information may be helpful for interpreting walking capacity and physical activity in patients with LSS receiving nonsurgical treatment and for estimating power and sample size when planning new trials.

Preeclampsia in pregnant women with polycystic ovary syndrome: risk factor analysis based on a retrospective cohort study.

To compare the clinical characteristics of pregnant women with polycystic ovary syndrome (PCOS) and perinatal outcomes with or without preeclampsia (PE) and to factors that are potentially associated with the onset of PE. This was a retrospective study of pregnant women diagnosed with PCOS from January 2017 to December 2021. Eligible patients were divided into two groups based on the presence or absence of preeclampsia: a PE group and a non-PE group. Demographics, clinical characteristics, maternal and perinatal outcomes, and potential factors linked to disease recurrence were analyzed. In total, 616 patients were enrolled and respectively classified into the PE group (n = 51) and the non-PE group (n = 565). The incidence of PE in pregnant women with PCOS was 8.28%; this was significantly higher than that in non-PCOS pregnant women (3.22%, p < 0.001). Logistic regression analysis of the predictive factors for PE in women with PCOS revealed that the combination of maternal hyperandrogenism, a pre-pregnancy BMI ≥ 24 kg/m², and a family history of cardiovascular disease (CVD) and assisted reproductive techniques (ART) exhibited the steepest receiver-operating characteristic (ROC) curve value at 0.797 [95% confidence interval (CI): 0.733-0.862]. Patients with PCOS have a higher incidence of PE. We identified a series of significant and independent factors associated with PE in PCOS: maternal hyperandrogenism, a pre-pregnancy BMI ≥ 24 kg/m², and a family history of CVD and ART.

A new prognostic risk model for acute myeloid leukemia patients based on telomere-related genes

Telomere maintenance is critical to ensure unlimited cancer cell proliferation, but the role of telomere-related genes in acute myeloid leukemia (AML) has not yet been thoroughly discussed. This study aims to develop a new prognostic risk model based on telomere-related genes and analyze potential mechanisms and targets. Cox regression analyses were used to build the prognostic risk model. Kaplan-Meier (KM) survival analysis and receiver operating characteristic (ROC) curve were used to assess the model performance. At the same time, we analyzed the relationship between the risk score and chemotherapy and immunotherapy and preliminarily explored possible mechanisms of immune resistance. The real-time polymerase chain reaction (PCR) was used to detect the prognosis gene expression levels. Finally, a prognostic signature of six telomere-related genes (TGPS6) including ALDH2, CDK18, DNMT3B, FRAT2, LGALSL, and RBL2 was constructed. The TGPS6 score was confirmed as an independent prognostic factor (HR 2.74, CI [2.13–3.53], p < 0.001) in AML and the five-year area under the ROC curve (AUC) value of the score in the training and validation set reached 0.74, 0.81 respectively. In addition, the TGPS6 perfected the European LeukemiaNet (ELN) 2017 prognosis risk stratification and performed well in both AML and cytogenetically normal AML (CN-AML) cohorts. The TGPS6 score also provided a reference for chemotherapy and immunotherapy in patients with AML.

Comparison of heart, grace and TIMI scores to predict major adverse cardiac events from chest pain in a Spanish health care region

Acute non-traumatic chest pain (ANTCP) is the second cause of consultation in the Emergency department (ED). About 70% of all Acute Myocardial Infarctions present as non persistent ST-elevation acute coronary syndrome (NSTE-ACS) in the electrocardiogram. Our aim was to compare whether the HEART risk score is more effective than the GRACE and TIMI scores for the diagnosis and prognosis of Major Adverse Cardiac Events (MACE) at six weeks in patients with ANTCP and NSTE-ACS. A prospective cohort study was conducted with patients with ANTCP that attended an ED and a Primary Care Emergency Center (PCEC) from April 2018 to December 2020. The primary outcome was MACE at six weeks. Diagnostic performance was calculated for each scale as the Area under the Receiver Operating Characteristic (ROC) curve (AUC), sensitivity (SE), specificity (SP), and predictive values (PV). Qualitative variables were compared using the Chi-square test, and continuous variables were compared using the nonparametric Kruskal–Wallis test. We adjusted a logistic regression for risk groups, age, and gender to determine the effect of the HEART, GRACE, and TIMI scores on MACE. The degree of agreement (kappa index) was calculated in the categorical classification of patients according to the three risk scales. Cox proportional hazards regressions were performed for each scale and were compared using partial likelihood ratio tests for non-nested models. From a sample of 317 patients with ANTCP, 14.82% had MACE at six weeks. The AUC was 0.743 (95% CI 0.67–0.81) for the HEART score, 0.717 (95% CI 0.64–0.79) for the TIMI score, and 0.649 (95% CI 0.561–0.738) for the GRACE score. The HEART scale identified low-risk patients with a higher SE and negative PV than the GRACE and TIMI scores. The HEART scale was better than the GRACE and TIMI scores at diagnosing and predicting MACE at six weeks in patients with ANTCP and probable NSTE-ACS. It was also a reliable tool for risk stratification in low-risk patients. Its application is feasible in EDs and PCECs, avoiding the need for complementary tests and their associated costs without compromising patient health.

HOMA-IR is an effective biomarker of non-alcoholic fatty liver disease in non-diabetic population.

This study aimed to investigate the correlation between homeostasis model assessment of insulin resistance (HOMA-IR) and non-alcoholic fatty liver disease (NAFLD) in the non-diabetic population and establish its diagnostic efficacy. This observational study involved participants divided into NAFLD and non-NAFLD groups, and baseline data were analyzed. Univariate and multivariate logistic regression analyses were used to correlate HOMA-IR with the risk of NAFLD. Receiver operating characteristic (ROC) curves were used to evaluate the diagnostic efficacy of HOMA-IR for NAFLD. Subgroup analyses of non-obese individuals were performed. Overall, 2234 non-diabetic participants were included. The HOMA-IR was significantly higher in the NAFLD group than in the non-NAFLD group. Multivariate logistic regression analysis showed that HOMA-IR was a strong and independent risk factor for NAFLD after correcting for confounding factors. The area under the ROC curve (AUC) value of HOMA-IR for predicting NAFLD was 0.792. In the non-obese non-diabetic population, HOMA-IR was an independent risk factor for increased risk of lean NAFLD after correcting for confounding factors. The AUC value of HOMA-IR for predicting lean NAFLD was 0.770. HOMA-IR is independently associated with the risk of NAFLD in the non-diabetic and non-obese non-diabetic populations and has good diagnostic value.

Application of neural network and nomogram for the prediction of risk factors for bone mineral density abnormalities: A cross-sectional NHANES-based survey

BackgroundThe risk of bone mineral density abnormalities is inconsistent between eastern and western regions owing to differences in ethnicity and dietary habits. A diet comprising carbohydrates and dietary fiber is not the common daily diet of the American population. Thus far, no studies have assessed the risk of bone mineral density abnormalities in the American population, and no predictive model has considered the intake of carbohydrates, dietary fiber, and coffee, as well as levels of various electrolytes for assessing bone mineral density abnormalities, especially in the elderly. This study conducted a neural network analysis and established a predictive nomogram considering an unusual diet to determine risk factors for bone mineral density abnormalities in the American population, mainly to provide a reference for the prevention and treatment of related bone mineral density abnormalities. MethodsOverall, 9871 patients who had complete data were selected from the National Health and Nutrition Examination Survey database during 2017–2020 as the research object, and patients’ general clinical characteristics were compared. Neural networks and nomograms were analyzed to screen for and quantify risk factors for bone mineral density abnormalities. Finally, the receiver operating characteristic (ROC) curve, calibration curve, decision curve analysis (DCA), and community indifference curve (CIC) were constructed to comprehensively verify the accuracy, differential ability, and clinical practicability of the neural network and nomogram. ResultsThe important risk factors for bone mineral density abnormalities were caffeine intake, carbohydrate consumption, body mass index (BMI), height, blood sodium, blood calcium, blood phosphorus, blood potassium, dietary fiber, vitamin D, participant age, weight, race, family history, and sex. The nomogram revealed that caffeine intake, carbohydrate consumption, blood potassium, and age were positively correlated with bone mineral density abnormalities, whereas BMI, height, blood phosphate, dietary fiber, and blood sodium were negatively correlated with bone mineral density abnormalities. Women were more prone to these abnormalities than men. The area under the ROC curve values of the neural network and nomogram were 85.8 % and 77.7 %, respectively. The Youden index was 58.04 % and 41.87 %, respectively. The detection sensitivity was 75.73 % and 65.06 %, respectively, and the specificity was 82.31 % and 76.81 %, respectively. Calibration curves of the neural network and nomogram showed better discrimination ability from the standard curve (P > 0.05). DCA and CIC analyses showed that the application of the neural network and nomogram to explore risk factors for bone mineral density abnormalities had certain clinical practicability, and the overall predictive effect of the model was good. ConclusionThe outcomes of the neural network and nomogram analyses suggested that diet structure and electrolyte changes are important significant risk factors for bone mineral density abnormalities, especially with increasing carbohydrate and caffeine intake and decreasing dietary fiber intake. The established model can also provide a reference for future risk prediction.

Antimicrobial Susceptibility Profiles and Key Determinants for Mortality in Burkholderia cepacia Complex Infections.

We aimed to define the clinical features and antimicrobial susceptibility profiles of Burkholderia cepacia complex infections and to determine the predictors for mortality. Our single-center retrospective study included patients with nosocomial B. cepacia complex infection between 2018 and 2022. We evaluated the predictors of 14-day and 28-day mortality by analyzing clinical and microbiological data. A total of 87 patients were included. Most infections (79.3%) occurred in the intensive care units (ICUs). Among B. cepacia complex isolates, 74.7% were susceptible to trimethoprim-sulfamethoxazole, 70.3% to levofloxacin, 50% to meropenem, and 23.4% to ceftazidime. The rates of 14-day mortality, 28-day mortality, and in-hospital mortality were 41.3% (n=36), 52.8% (n=46), and 64.3% (n=56), respectively. Multivariate analysis revealed neutrophil/lymphocyte ratio (NLR) (odds ratio [OR]=1.05, p=0.024), platelet count (OR=1.00, p=0.011), creatinine (OR=2.14, p=0.006), and aspartate aminotransferase (AST) (OR=1.02, p=0.028) as predictors for 14-day mortality. In addition to NLR (OR=1.07, p=0.014), platelet count (OR=1.00, p=0.039), creatinine (OR=2.05, p=0.008), and AST (OR=1.02, p=0.035), procalcitonin (OR=1.05, p=0.049) was also found as an independent predictor for 28-day mortality. In receiver operating characteristic (ROC) curve analysis for predicting 14-day mortality, area under the ROC curve (AUC) values were 0.684 (p=0.003) in NLR, 0.719 (p<0.001) in platelet count, 0.673 (p=0.003) in procalcitonin, 0.743 (p<0.001) in creatinine, and 0.700 (p<0.001) in AST. In ROC curve analysis for predicting 28-day mortality, AUC values were 0.674 (p=0.002) in NLR, 0.651 (p=0.010) in platelet count, 0.638 (p=0.020) in procalcitonin, 0.730 (p<0.001) in creatinine, and 0.692 (p=0.001) in AST. Increasing antibiotic resistance and higher mortality rates justify that B. cepacia complex is a significant threat to hospitalized patients, especially in ICUs. Elevated levels of NLR, AST, creatinine, procalcitonin, and decreased platelet may predict poor clinical outcomes and could help clinicians in the management of this notorious bacterial pathogen.