Accurate prediction of acute pancreatitis severity with integrative blood molecular measurements.
Ontology highlight
ABSTRACT: Background
Early diagnosis of severe acute pancreatitis (SAP) is essential to minimize its mortality and improve prognosis. We aimed to develop an accurate and applicable machine learning predictive model based on routine clinical testing results for stratifying acute pancreatitis (AP) severity.Results
We identified 11 markers predictive of AP severity and trained an AP stratification model called APSAVE, which classified AP cases within 24 hours at an average area under the curve (AUC) of 0.74 +/- 0.04. It was further validated in 568 validation cases, achieving an AUC of 0.73, which is similar to that of Ranson's criteria (AUC = 0.74) and higher than APACHE II and BISAP (AUC = 0.69 and 0.66, respectively).Conclusions
We developed and validated a venous blood marker-based AP severity stratification model with higher accuracy and broader applicability, which holds promises for reducing SAP mortality and improving its clinical outcomes.Materials and methods
Nine hundred and forty-five AP patients were enrolled into this study. Clinical venous blood tests covering 65 biomarkers were performed on AP patients within 24 hours of admission. An SAP prediction model was built with statistical learning to select biomarkers that are most predictive for AP severity.
SUBMITTER: Sun HW
PROVIDER: S-EPMC8034948 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:BackgroundPatients with severe acute pancreatitis (SAP) have a high mortality, thus early diagnosis and interventions are critical for improving survival. However, conventional tests are limited in acute pancreatitis (AP) stratification. We aimed to assess AP severity by integrating the informative clinical measurements with cell free DNA (cfDNA) methylation markers.MethodsOne hundred and seventy-five blood samples were collected from 61 AP patients at multiple time points, plus 24 samples from healthy individuals. Genome-wide cfDNA methylation profiles of all samples were characterized with reduced representative bisulfite sequencing. Clinical blood tests covering 93 biomarkers were performed on AP patients within 24 h. SAP predication models were built based on cfDNA methylation and conventional blood biomarkers separately and in combination.ResultsWe identified 565 and 59 cfDNA methylation markers informative for acute pancreatitis and its severity. These markers were used to develop prediction models for AP and SAP with area under the receiver operating characteristic of 0.92 and 0.81, respectively. Twelve blood biomarkers were systematically screened for a predictor of SAP with a sensitivity of 87.5% for SAP, and a specificity of 100% in mild acute pancreatitis, significantly higher than existing blood tests. An expanded model integrating 12 conventional blood biomarkers with 59 cfDNA methylation markers further improved the SAP prediction sensitivity to 92.2%.ConclusionsThese findings have demonstrated that accurate prediction of SAP by the integration of conventional and novel blood molecular markers, paving the way for early and effective SAP intervention through a non-invasive rapid diagnostic test.
| S-EPMC8680202 | biostudies-literature
Project description:ObjectivesThe aim of the study was to validate and optimize a severity prediction model for acute pancreatitis (AP) and to examine blood urea nitrogen (BUN) level changes from admission as a severity predictor.Study designPatients from 2 hospitals were included for the validation model (Children's Hospital of the King's Daughters and Children's National Hospital). Children's Hospital of the King's Daughters and Cincinnati Children's Hospital Medical Center data were used for analysis of BUN at 24 to 48 hours.ResultsThe validation cohort included 73 patients; 22 (30%) with either severe or moderately severe AP, combined into the all severe AP (SAP) group. Patients with SAP had higher BUN (P = 0.002) and lower albumin (P = 0.005). Admission BUN was confirmed as a significant predictor (P = 0.005) of SAP (area under the receiver operating characteristic [AUROC] 0.73, 95% confidence interval [CI] 0.60-0.86). Combining BUN (P = 0.005) and albumin (P = 0.004) resulted in better prediction for SAP (AUROC 0.83, 95% CI 0.72-0.94). A total of 176 AP patients were analyzed at 24-48 hours; 39 (22%) met criteria for SAP. Patients who developed SAP had a significantly higher BUN (P < 0.001) after 24 hours. Elevated BUN levels within 24 to 48 hours were independently predictive of developing SAP (AUROC: 0.76, 95% CI: 0.66-0.85). Patients who developed SAP had a significantly smaller percentage decrease in BUN from admission to 24 to 48 hours (P = 0.002).ConclusionWe externally validated the prior model with admission BUN levels and further optimized it by incorporating albumin. We also found that persistent elevation of BUN is associated with development of SAP. Our model can be used to risk stratify patients with AP on admission and again at 24 to 48 hours.
| S-EPMC8020899 | biostudies-literature
Project description:The need to calibrate to correct for sensor-to-sensor fabrication variation and sensor drift has proven a significant hurdle in the widespread use of biosensors. To maintain clinically relevant (±20% for this application) accuracy, for example, commercial continuous glucose monitors require recalibration several times a day, decreasing convenience and increasing the chance of user errors. Here, however, we demonstrate a "dual-frequency" approach for achieving the calibration-free operation of electrochemical biosensors that generate an output by using square-wave voltammetry to monitor binding-induced changes in electron transfer kinetics. Specifically, we use the square-wave frequency dependence of their response to produce a ratiometric signal, the ratio of peak currents collected at responsive and non- (or low) responsive square-wave frequencies, which is largely insensitive to drift and sensor-to-sensor fabrication variations. Using electrochemical aptamer-based (E-AB) biosensors as our test bed, we demonstrate the accurate and precise operation of sensors against multiple drugs, achieving accuracy in the measurement of their targets of within better than 20% across dynamic ranges of up to 2 orders of magnitude without the need to calibrate each individual sensor.
| S-EPMC6519131 | biostudies-literature
Project description:BackgroundMachine learning (ML) algorithms are widely applied in building models of medicine due to their powerful studying and generalizing ability. This study aims to explore different ML models for early identification of severe acute pancreatitis (SAP) among patients hospitalized for acute pancreatitis.MethodsThis retrospective study enrolled patients with acute pancreatitis (AP) from multiple centers. Data from the First Affiliated Hospital and Changshu No. 1 Hospital of Soochow University were adopted for training and internal validation, and data from the Second Affiliated Hospital of Soochow University were adopted for external validation from January 2017 to December 2021. The diagnosis of AP and SAP was based on the 2012 revised Atlanta classification of acute pancreatitis. Models were built using traditional logistic regression (LR) and automated machine learning (AutoML) analysis with five types of algorithms. The performance of models was evaluated by the receiver operating characteristic (ROC) curve, the calibration curve, and the decision curve analysis (DCA) based on LR and feature importance, SHapley Additive exPlanation (SHAP) Plot, and Local Interpretable Model Agnostic Explanation (LIME) based on AutoML.ResultsA total of 1,012 patients were included in this study to develop the AutoML models in the training/validation dataset. An independent dataset of 212 patients was used to test the models. The model developed by the gradient boost machine (GBM) outperformed other models with an area under the ROC curve (AUC) of 0.937 in the validation set and an AUC of 0.945 in the test set. Furthermore, the GBM model achieved the highest sensitivity value of 0.583 among these AutoML models. The model developed by eXtreme Gradient Boosting (XGBoost) achieved the highest specificity value of 0.980 and the highest accuracy of 0.958 in the test set.ConclusionsThe AutoML model based on the GBM algorithm for early prediction of SAP showed evident clinical practicability.
| S-EPMC9226483 | biostudies-literature
Project description:Imaging-based blood flow measurement techniques, such as particle image velocimetry, have become an important tool in cardiovascular research. They provide quantitative information about blood flow, which benefits applications ranging from developmental biology to tumor perfusion studies. Studies using these methods can be classified based on whether they use artificial tracers or red blood cells to visualize the fluid motion. We here present the first direct comparison in vivo of both methods. For high magnification cases, the experiments using red blood cells strongly underestimate the flow (up to 50% in the present case), as compared to the tracer results. For medium magnification cases, the results from both methods are indistinguishable as they give the same underestimation of the real velocities (approximately 33%, based on in vitro reference measurements). These results suggest that flow characteristics reported in literature cannot be compared without a careful evaluation of the imaging characteristics. A method to predict the expected flow averaging behavior for a particular facility is presented.
| S-EPMC3447936 | biostudies-literature
Project description:Background Early identification of severe acute pancreatitis (SAP) is key to reducing mortality and improving prognosis. We aimed to establish a radiomics model and nomogram for early prediction of acute pancreatitis (AP) severity based on contrast-enhanced computed tomography (CT) images. Methods We retrospectively analyzed 215 patients with first-episode AP, including 141 in the training cohort (87 men and 54 women, mean age 51.37±16.09 years) and 74 in the test cohort (40 men and 34 women, mean age 55.49±17.83 years). Radiomics features were extracted from portal venous phase images based on pancreatic and peripancreatic regions. The light gradient boosting machine (LightGBM) algorithm was used for feature selection, a logistic regression (LR) model was established and trained by 10-fold cross-validation, and a nomogram was established based on the best features. The model’s predictive performance was evaluated according to the area under the curve (AUC) of the receiver operating characteristic (ROC) curve, sensitivity, specificity, and accuracy. Results A total of 13 optimal radiomics features were selected by LightGBM for LR model building. The AUC of the radiomics (LR) model was 0.992 [95% confidence interval (CI): 0.963–0.996] in the training cohort, 0.965 (95% CI: 0.924–0.981) in the validation cohort, and 0.894 (95% CI: 0.789–0.966) in the test cohort. The sensitivity was 0.862 (95% CI: 0.674–0.954), the specificity was 0.800 (95% CI: 0.649–0.899), and the accuracy was 0.824 (95% CI: 0.720–0.919). The nomogram based on the 13 radiomics features showed that SAP would be predicted when the total score was greater than 124. Conclusions The radiomics model based on enhanced-CT images of pancreatic and peripancreatic regions performed well in the early prediction of AP severity. The nomogram based on selected radiomics features could provide a reference for AP clinical assessment.
| S-EPMC10006146 | biostudies-literature
Project description:Background: The management of the moderate and severe forms of acute pancreatitis (AP) with necrosis and multiorgan failure remains a challenge. To predict the severity and mortality of AP multiple clinical, laboratory-, and imaging-based scoring systems are available. Aim: To investigate, if the computed tomography severity index (CTSI) can predict the outcomes of AP better than other scoring systems. Methods: A systematic search was performed in three databases: Pubmed, Embase, and the Cochrane Library. Eligible records provided data from consecutive AP cases and used CTSI or modified CTSI (mCTSI) alone or in combination with other prognostic scores [Ranson, bedside index of severity in acute pancreatitis (BISAP), Acute Physiology, and Chronic Health Examination II (APACHE II), C-reactive protein (CRP)] for the evaluation of severity or mortality of AP. Area under the curves (AUCs) with 95% confidence intervals (CIs) were calculated and aggregated with STATA 14 software using the metandi module. Results: Altogether, 30 studies were included in our meta-analysis, which contained the data of 5,988 AP cases. The pooled AUC for the prediction of mortality was 0.79 (CI 0.73-0.86) for CTSI; 0.87 (CI 0.83-0.90) for BISAP; 0.80 (CI 0.72-0.89) for mCTSI; 0.73 (CI 0.66-0.81) for CRP level; 0.87 (CI 0.81-0.92) for the Ranson score; and 0.91 (CI 0.88-0.93) for the APACHE II score. The APACHE II scoring system had significantly higher predictive value for mortality than CTSI and CRP (p = 0.001 and p < 0.001, respectively), while the predictive value of CTSI was not statistically different from that of BISAP, mCTSI, CRP, or Ranson criteria. The AUC for the prediction of severity of AP were 0.80 (CI 0.76-0.85) for CTSI; 0.79, (CI 0.72-0.86) for BISAP; 0.83 (CI 0.75-0.91) for mCTSI; 0.73 (CI 0.64-0.83) for CRP level; 0.81 (CI 0.75-0.87) for Ranson score and 0.80 (CI 0.77-0.83) for APACHE II score. Regarding severity, all tools performed equally. Conclusion: Though APACHE II is the most accurate predictor of mortality, CTSI is a good predictor of both mortality and AP severity. When the CT scan has been performed, CTSI is an easily calculable and informative tool, which should be used more often in routine clinical practice.
| S-EPMC6718714 | biostudies-literature
Project description:Predicting the subcellular localization of proteins conquers the major drawbacks of high-throughput localization experiments that are costly and time-consuming. However, current subcellular localization predictors are limited in scope and accuracy. In particular, most predictors perform well on certain locations or with certain data sets while poorly on others. Here, we present PSI, a novel high accuracy web server for plant subcellular localization prediction. PSI derives the wisdom of multiple specialized predictors via a joint-approach of group decision making strategy and machine learning methods to give an integrated best result. The overall accuracy obtained (up to 93.4%) was higher than best individual (CELLO) by ~10.7%. The precision of each predicable subcellular location (more than 80%) far exceeds that of the individual predictors. It can also deal with multi-localization proteins. PSI is expected to be a powerful tool in protein location engineering as well as in plant sciences, while the strategy employed could be applied to other integrative problems. A user-friendly web server, PSI, has been developed for free access at http://bis.zju.edu.cn/psi/.
| S-EPMC3806775 | biostudies-other
Project description:Cannabis is the most commonly and widely used illicit drug in the world and is also the most commonly used drug of abuse in alcohol drinkers. Experimental studies have shown conflicting results of the effects of cannabis on the severity of acute pancreatitis (AP). The purpose of this study is to ascertain the clinical effects of simultaneous alcohol and cannabis use on severity at presentation and outcomes of acute alcoholic pancreatitis (AAP).A retrospective review was conducted on the patients discharged with principle or secondary diagnosis of AP using ICD-9 & ICD-10 codes during the time period from January 2006 to December 2015 at a large community-based hospital in Central Georgia. Patients with alcoholic pancreatitis with cannabis (CB+) and without cannabis (CB-) use were identified and were matched with sex and age.Our study findings showed that a greater percentage of CB+ patients did not have a systemic inflammatory response syndrome (SIRS) score (P=0.043), had a lower BISAP score (P=0.031), and had a significantly lower BUN level (P=0.033), but there was no difference in the Balthazar Index and revised Atlanta classification severity between the two groups. CB+ patients tended to need less ICU care than CB- patients (P=0.059).Based on our findings, we found that CB+ patients had less severe presentation of AAP indicating that cannabis could modulate the inflammatory effect of alcohol on the pancreas. Further large scale prospective studies are needed to confirm our results.
| S-EPMC5539398 | biostudies-other
Project description:BackgroundThe effect of diet on risk of acute pancreatitis (AP) has been suggested by prior studies, but the association of dietary habits with severity of AP has not been previously evaluated.ObjectiveThe objective of the study was to assess differences in reported dietary habits in patients with severe AP compared with those with mild or moderate AP.MethodsA prospectively maintained cohort of patients with AP was utilized. A brief questionnaire on dietary habits was implemented. Dietary habits were categorized based on the overall type of diet, fruit/vegetable servings, fat content, dairy consumption, dessert/sweets consumption, and fluid intake. Patients were grouped into mild/moderate and severe AP. Multivariate analysis was used to determine whether dietary habits have an independent association with AP severity.Results407 patients with AP were studied. Mean patient age was 51 y, and 202 (50%) were men. 29% of patients were smokers and 46% actively consumed alcohol. 225 patients had mild AP, 103 moderate AP, and 79 developed severe AP. The 3 groups were comparable in race, body mass index, etiology of AP, and comorbidities. Dietary factors were overall comparable between the groups except for diet type: subjects with severe AP had a higher percentage of consuming a meat-rich diet (84%) than patients with mild AP (72%) and moderate AP (67%) (P = 0.04). Based on multivariable logistic regression, the OR of developing severe AP was 2.5 (95% CI: 1.24-5.32, P = 0.01) between patients who eat a meat-rich diet and those who consume a vegetable-based diet.ConclusionsA meat-rich diet is independently associated with the development of persistent organ failure (severe disease) in patients with AP. These findings require further evaluation and could be useful for patient counseling, risk stratification, and disease prevention. This study is registered at clinicaltrials.gov as NCT03075605.
| S-EPMC6295618 | biostudies-literature