scholarly journals Machine Learning Applied to Omics Datasets Predicts Mortality in Patients with Alcoholic Hepatitis

Metabolites ◽  
2022 ◽  
Vol 12 (1) ◽  
pp. 41
Author(s):  
Bei Gao ◽  
Tsung-Chin Wu ◽  
Sonja Lang ◽  
Lu Jiang ◽  
Yi Duan ◽  
...  
Keyword(s):  
Alcoholic Hepatitis ◽  
The United States ◽  
Meld Score ◽  
Mortality Prediction ◽  
Gradient Boosting ◽  
Support Vector ◽  
Risk Of Death ◽  
Bacterial Microbiota ◽  
Health Care Burden ◽  
End Stage

Alcoholic hepatitis is a major health care burden in the United States due to significant morbidity and mortality. Early identification of patients with alcoholic hepatitis at greatest risk of death is extremely important for proper treatments and interventions to be instituted. In this study, we used gradient boosting, random forest, support vector machine and logistic regression analysis of laboratory parameters, fecal bacterial microbiota, fecal mycobiota, fecal virome, serum metabolome and serum lipidome to predict mortality in patients with alcoholic hepatitis. Gradient boosting achieved the highest AUC of 0.87 for both 30-day mortality prediction using the bacteria and metabolic pathways dataset and 90-day mortality prediction using the fungi dataset, which showed better performance than the currently used model for end-stage liver disease (MELD) score.

Cardiovascular Applications of Artificial Intelligence in Research, Diagnosis, and Disease Management

2022 ◽  
pp. 80-127
Author(s):  
Viswanathan Rajagopalan ◽  
Houwei Cao
Keyword(s):  
Artificial Intelligence ◽  
Disease Management ◽  
Heart Attack ◽  
Congenital Heart ◽  
The United States ◽  
Disease Classification ◽  
Gradient Boosting ◽  
Support Vector ◽  
Gradient Boosting Machine ◽  
Infarction Heart

Despite significant advancements in diagnosis and disease management, cardiovascular (CV) disorders remain the No. 1 killer both in the United States and across the world, and innovative and transformative technologies such as artificial intelligence (AI) are increasingly employed in CV medicine. In this chapter, the authors introduce different AI and machine learning (ML) tools including support vector machine (SVM), gradient boosting machine (GBM), and deep learning models (DL), and their applicability to advance CV diagnosis and disease classification, and risk prediction and patient management. The applications include, but are not limited to, electrocardiogram, imaging, genomics, and drug research in different CV pathologies such as myocardial infarction (heart attack), heart failure, congenital heart disease, arrhythmias, valvular abnormalities, etc.

Concomitant Use of Gabapentinoids with Opioids Is Associated with Increased Mortality and Morbidity among Dialysis Patients

2020 ◽  
Vol 51 (6) ◽  
pp. 424-432 ◽  
Author(s):  
Salina P. Waddy ◽  
Adan Z. Becerra ◽  
Julia B. Ward ◽  
Kevin E. Chan ◽  
Chyng-Wen Fwu ◽  
...  
Keyword(s):  
The United States ◽  
Future Research ◽  
Dialysis Patients ◽  
Mortality And Morbidity ◽  
Medication Prescription ◽  
Risk Of Death ◽  
Increased Risk ◽  
Stage Renal Disease ◽  
End Stage ◽  
Dialysis Discontinuation

Background: The opioid epidemic is a public health emergency and appropriate medication prescription for pain remains challenging. Physicians have increasingly prescribed gabapentinoids for pain despite limited evidence supporting their use. We determined the prevalence of concomitant gabapentinoid and opioid prescriptions and evaluated their associations with outcomes among dialysis patients. Methods: We used the United States Renal Data System to identify patients treated with dialysis with Part A, B, and D coverage for all of 2010. Patients were grouped into 4 categories of drugs exposure status in 2010: (1) no prescriptions of either an opioid or gabapentinoid, (2) ≥1 prescription of an opioid and no prescriptions of gabapentinoids, (3) no prescriptions of an opioid and ≥1 prescription of gabapenbtinoids, (4) ≥1 prescription of both an opioid and gabapentinoid. Outcomes included 2-year all-cause death, dialysis discontinuation, and hospitalizations assessed in 2011 and 2012. Results: The study population included 153,758 dialysis patients. Concomitant prescription of an opioid and gabapentin (15%) was more common than concomitant prescription of an opioid and pregabalin (4%). In adjusted analyses, concomitant prescription of an opioid and gabapentin compared to no prescription of either was associated with increased risk of death (hazard ratio [HR] 1.16, 95% CI 1.12–1.19), dialysis discontinuation (HR 1.14, 95% CI 1.03–1.27), and hospitalization (HR 1.33, 95% CI 1.31–1.36). Concomitant prescription of an opioid and pregabalin compared to no prescription of either was associated with increased mortality (HR 1.22, 95% CI 1.16–1.28) and hospitalization (HR 1.37, 95% CI 1.33–1.41), but not dialysis discontinuation (HR 1.13, 95% CI 0.95–1.35). Prescription of opioids and gabepentinoids compared to only being prescribed opioids was associated with higher risk of hospitalizations, but not mortality, or dialysis discontinuation. Conclusions: Concomitant prescription of opioids and gabapentinoids among US dialysis patients is common, and both drugs have independent effects on outcomes. Future research should prospectively investigate the potential harms of such drugs and identify safer alternatives for treatment of pain in end-stage renal disease patients.

scholarly journals Improved survival with renal transplantation for end-stage renal disease due to granulomatosis with polyangiitis: data from the United States Renal Data System

2018 ◽  
Vol 77 (9) ◽  
pp. 1333-1338 ◽  
Author(s):  
Zachary S Wallace ◽  
Rachel Wallwork ◽  
Yuqing Zhang ◽  
Na Lu ◽  
Frank Cortazar ◽  
...  
Keyword(s):  
United States ◽  
Renal Transplantation ◽  
Renal Transplant ◽  
End Stage Renal Disease ◽  
Granulomatosis With Polyangiitis ◽  
The United States ◽  
Risk Of Death ◽  
All Cause Mortality ◽  
Stage Renal Disease ◽  
End Stage

BackgroundRenal transplantation is the optimal treatment for selected patients with end-stage renal disease (ESRD). However, the survival benefit of renal transplantation among patients with ESRD attributed to granulomatosis with polyangiitis (GPA) is unknown.MethodsWe identified patients from the United States Renal Data System with ESRD due to GPA (ESRD-GPA) between 1995 and 2014. We restricted our analysis to waitlisted subjects to evaluate the impact of transplantation on mortality. We followed patients until death or the end of follow-up. We compared the relative risk (RR) of all-cause and cause-specific mortality in patients who received a transplant versus non-transplanted patients using a pooled logistic regression model with transplantation as a time-varying exposure.ResultsDuring the study period, 1525 patients were waitlisted and 946 received a renal transplant. Receiving a renal transplant was associated with a 70% reduction in the risk of all-cause mortality in multivariable-adjusted analyses (RR=0.30, 95% CI 0.25 to 0.37), largely attributed to a 90% reduction in the risk of death due to cardiovascular disease (CVD) (RR=0.10, 95% 0.06–0.16).DiscussionRenal transplantation is associated with a significant decrease in all-cause mortality among patients with ESRD attributed to GPA, largely due to a decrease in the risk of death to CVD. Prompt referral for transplantation is critical to optimise outcomes for this patient population.

scholarly journals Validation of a new prognostic model to predict short and medium-term survival in patients with liver cirrhosis

2020 ◽  
Vol 20 (1) ◽  
Author(s):  
Tomasz Dziodzio ◽  
Robert Öllinger ◽  
Wenzel Schöning ◽  
Antonia Rothkäppel ◽  
Radoslav Nikolov ◽  
...  
Keyword(s):  
Liver Cirrhosis ◽  
Liver Disease ◽  
Risk Score ◽  
Meld Score ◽  
Medium Term ◽  
Term Survival ◽  
Risk Of Death ◽  
Class B ◽  
Pugh Class ◽  
End Stage

Abstract Background MELD score and MELD score derivates are used to objectify and grade the risk of liver-related death in patients with liver cirrhosis. We recently proposed a new predictive model that combines serum creatinine levels and maximum liver function capacity (LiMAx®), namely the CreLiMAx risk score. In this validation study we have aimed to reproduce its diagnostic accuracy in patients with end-stage liver disease. Methods Liver function of 113 patients with liver cirrhosis was prospectively investigated. Primary end-point of the study was liver-related death within 12 months of follow-up. Results Alcoholic liver disease was the main cause of liver disease (n = 51; 45%). Within 12 months of follow-up 11 patients (9.7%) underwent liver transplantation and 17 (15.1%) died (13 deaths were related to liver disease, two not). Measures of diagnostic accuracy were comparable for MELD, MELD-Na and the CreLiMAx risk score as to power in predicting short and medium-term mortality risk in the overall cohort: AUROCS for liver related risk of death were for MELD [6 months 0.89 (95% CI 0.80–0.98) p < 0.001; 12 months 0.89 (95% CI 0.81–0.96) p < 0.001]; MELD-Na [6 months 0.93 (95% CI 0.85–1.00) p < 0.001 and 12 months 0.89 (95% CI 0.80–0.98) p < 0.001]; CPS 6 months 0.91 (95% CI 0.85–0.97) p < 0.01 and 12 months 0.88 (95% CI 0.80–0.96) p < 0.001] and CreLiMAx score [6 months 0.80 (95% CI 0.67–0.96) p < 0.01 and 12 months 0.79 (95% CI 0.64–0.94) p = 0.001]. In a subgroup analysis of patients with Child-Pugh Class B cirrhosis, the CreLiMAx risk score remained the only parameter significantly differing in non-survivors and survivors. Furthermore, in these patients the proposed score had a good predictive performance. Conclusion The CreLiMAx risk score appears to be a competitive and valid tool for estimating not only short- but also medium-term survival of patients with end-stage liver disease. Particularly in patients with Child-Pugh Class B cirrhosis the new score showed a good ability to identify patients not at risk of death.

A Critical Review of MELD as a Reliable Tool for Transplant Prioritization

2019 ◽  
Vol 39 (04) ◽  
pp. 403-413 ◽  
Author(s):  
Sophie-Caroline Sacleux ◽  
Didier Samuel
Keyword(s):  
United States ◽  
Serum Creatinine ◽  
International Normalized Ratio ◽  
The United States ◽  
Meld Score ◽  
Waiting List ◽  
Organ Shortage ◽  
Short Term ◽  
Short Term Mortality ◽  
End Stage

AbstractIn a context of global organ shortage, the Model for End-Stage Liver Disease (MELD) score seems to be a fair prioritization tool, with a paradigm: “sickest first.” Since its introduction in the United States in 2002, it has been rapidly adopted by transplant centers and organ sharing agencies around the world. The MELD score showed its effectiveness with a 12% reduction in waiting list mortality in the United States. Its success is linked to its simplicity, the use of basic variables (serum creatinine, serum bilirubin, and international normalized ratio [INR]), and its ability to predict short-term mortality, particularly on the transplant waiting list. However, this score is not perfect: its variables may have disadvantages for some patients, especially women, with serum creatinine and interlaboratory variability of the INR. The MELD score does not take into account some variables associated with poor short-term prognosis in cirrhotic patients. In addition, it is currently capped at 40, which results in the exclusion of sicker patients who could greatly benefit from transplantation. Finally, the MELD score does not accurately reflect the prognosis of several conditions, requiring a MELD exception system. Some solutions have been suggested such as MELD-Na or MELD uncapping, but it has not yet been fully accepted by all transplant centers.

Clostridium difficile infection in dialysis patients

2016 ◽  
Vol 65 (2) ◽  
pp. 353-357 ◽  
Author(s):  
Ankita Tirath ◽  
Sandra Tadros ◽  
Samuel L Coffin ◽  
Kristina W Kintziger ◽  
Jennifer L Waller ◽  
...  
Keyword(s):  
Clostridium Difficile ◽  
The United States ◽  
Hiv Positive ◽  
Adjusted Relative Risk ◽  
Dialysis Patients ◽  
Risk Of Death ◽  
Nosocomial Diarrhea ◽  
Increased Risk ◽  
Stage Renal Disease ◽  
End Stage

Clostridium difficile infection (CDI) is the most common cause of nosocomial diarrhea. Patients with end-stage renal disease (ESRD) may be at increased risk for CDI. Patients with ESRD with CDI have increased mortality, longer length of stay, and higher costs. The present studies extend these observations and address associated comorbidities, incidence of recurrence, and risk factors for mortality. We queried the United States Renal Data System (USRDS) for patients with ESRD diagnosed with CDI, and assessed for the incidence of infection, comorbidities, and mortality. The records of 419,875 incident dialysis patients from 2005 to 2008 were reviewed. 4.25% had a diagnosis of a first CDI. In the majority of patients with CDI positive, a hospitalization or ICU stay was documented within 90 days prior to the diagnosis of CDI. The greatest adjusted relative risk (aRR) of CDI was present in patients with HIV (aRR 2.68), age ≥65 years (aRR 1.76), and bacteremia (aRR 1.74). The adjusted HR (aHR) for death was 1.80 in patients with CDI. The comorbidities demonstrating the greatest risk for death in dialysis patients with CDI included age ≥65 years and cirrhosis (aHR 2.28 and 1.76, respectively). Recurrent CDI occurred in 23.6%, was more common in Caucasians, and in those who were older. CDI is a common occurrence in patients with ESRD, with elderly patients, patients with HIV positive, and bacteremic patients at highest risk for infection. Patients with CDI had nearly a twofold increased risk of death.

scholarly journals Predicting in-Hospital Mortality of Patients with COVID-19 Using Machine Learning Techniques

2021 ◽  
Vol 11 (5) ◽  
pp. 343
Author(s):  
Fabiana Tezza ◽  
Giulia Lorenzoni ◽  
Danila Azzolina ◽  
Sofia Barbar ◽  
Lucia Anna Carmela Leone ◽  
...  
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Hospital Mortality ◽  
Learning Algorithm ◽  
Vital Signs ◽  
Mortality Prediction ◽  
Machine Learning Techniques ◽  
Gradient Boosting ◽  
Support Vector ◽  
Learning Techniques

The present work aims to identify the predictors of COVID-19 in-hospital mortality testing a set of Machine Learning Techniques (MLTs), comparing their ability to predict the outcome of interest. The model with the best performance will be used to identify in-hospital mortality predictors and to build an in-hospital mortality prediction tool. The study involved patients with COVID-19, proved by PCR test, admitted to the “Ospedali Riuniti Padova Sud” COVID-19 referral center in the Veneto region, Italy. The algorithms considered were the Recursive Partition Tree (RPART), the Support Vector Machine (SVM), the Gradient Boosting Machine (GBM), and Random Forest. The resampled performances were reported for each MLT, considering the sensitivity, specificity, and the Receiving Operative Characteristic (ROC) curve measures. The study enrolled 341 patients. The median age was 74 years, and the male gender was the most prevalent. The Random Forest algorithm outperformed the other MLTs in predicting in-hospital mortality, with a ROC of 0.84 (95% C.I. 0.78–0.9). Age, together with vital signs (oxygen saturation and the quick SOFA) and lab parameters (creatinine, AST, lymphocytes, platelets, and hemoglobin), were found to be the strongest predictors of in-hospital mortality. The present work provides insights for the prediction of in-hospital mortality of COVID-19 patients using a machine-learning algorithm.

scholarly journals Factors Associated With Detection and Survival of T1 Hepatocellular Carcinoma in the United States: National Cancer Database Analysis

2020 ◽  
Vol 18 (9) ◽  
pp. 1210-1220
Author(s):  
Ju Dong Yang ◽  
Michael Luu ◽  
Amit G. Singal ◽  
Mazen Noureddin ◽  
Alexander Kuo ◽  
...  
Keyword(s):  
United States ◽  
The United States ◽  
Meld Score ◽  
Academic Institution ◽  
National Cancer Database ◽  
Factors Associated ◽  
Academic Center ◽  
Pertinent Data ◽  
T1 Stage ◽  
End Stage

Background: It remains unknown to what extent hepatocellular carcinomas (HCCs) are detected very early (T1 stage; ie, unifocal <2 cm) in the United States. The aim of this study was to investigate the trends and factors associated with very early detection of HCC and resultant outcomes. Methods: Patients with HCC diagnosed from 2004 through 2014 were identified from the National Cancer Database. Logistic regression was used to identify factors associated with T1 HCC detection, and Cox proportional hazard analyses identified factors associated with overall survival among patients with T1 HCC. Results: Of 110,182 eligible patients, the proportion with T1 HCC increased from 2.6% in 2004 to 6.8% in 2014 (P<.01). The strongest correlate of T1 HCC detection was receipt of care at an academic institution (odds ratio, 3.51; 95% CI, 2.31–5.34). Older age, lack of insurance, high Model for End-Stage Liver Disease (MELD) score, high alpha-fetoprotein, increased Charlson-Deyo comorbidity score, and nonsurgical treatment were associated with increased mortality, and care at an academic center (hazard ratio [HR], 0.27; 95% CI, 0.15–0.48) was associated with reduced mortality in patients with T1 HCC. Liver transplantation (HR, 0.27; 95% CI, 0.20–0.37) and surgical resection (HR, 0.67; 95% CI, 0.48–0.93) were independently associated with improved survival compared with ablation. This is the first study to examine the trend of T1 HCC using the National Cancer Database, which covers approximately 70% of all cancer diagnoses in the United States, using robust statistical analyses. Limitations of the study include a retrospective study design using administrative data and some pertinent data that were not available. Conclusions: Despite increases over time, <10% of HCCs are detected at T1 stage. The strongest correlates of survival among patients with T1 HCC are receiving care at an academic institution and surgical treatment.

scholarly journals Assessment of the Model for End-stage Liver Disease (MELD) Score in Predicting Prognosis of Patients with Alcoholic Hepatitis

2014 ◽  
Vol 4 (1) ◽  
pp. 19-24 ◽  
Author(s):  
Sundeep K. Goyal ◽  
Vinod K. Dixit ◽  
Ashok K. Jain ◽  
Pradeep K. Mohapatra ◽  
Jayant K. Ghosh
Keyword(s):  
Liver Disease ◽  
Alcoholic Hepatitis ◽  
Meld Score ◽  
End Stage Liver Disease ◽  
End Stage

scholarly journals USEFULNESS OF A NEW PROGNOSTIC INDEX FOR ALCOHOLIC HEPATITIS

2015 ◽  
Vol 52 (1) ◽  
pp. 22-26
Author(s):  
Jazon Romilson de Souza ALMEIDA ◽  
Roberta Chaves ARAÚJO ◽  
Giane Vieira de CASTILHO ◽  
Letícia STAHELIN ◽  
Lívia dos Reis PANDOLFI ◽  
...  
Keyword(s):  
Liver Disease ◽  
Hepatic Encephalopathy ◽  
High Risk ◽  
Alcoholic Hepatitis ◽  
Prognostic Index ◽  
Risk Of Death ◽  
End Stage Liver Disease ◽  
Average Survival ◽  
Short Term Mortality ◽  
End Stage

Background Alcoholic liver disease is a major cause of end-stage liver disease worldwide and severe forms of alcoholic hepatitis are associated with a high short-term mortality. Objectives To analyze the importance of age-bilirubin-INR-creatinine (ABIC) score as an index of mortality and predictor for complications in patients with alcoholic hepatitis. To evaluate its correlation with those complications, with risk of death, as well as the scores model for end stage liver disease (MELD) and Maddrey’s discriminat function. Methods A total of 46 medical records of patients who had been hospitalized with alcoholic hepatitis were assessed retrospectively with lab tests on admission and after seven days. Score calculations were carried out and analyzed as well. Results The scores showed positive reciprocal correlation and were associated with both hepatic encephalopathy and ascites. ABIC index, which was classified as high risk, presented as a risk factor for these complications and for death. In univariate logistic regression analysis of mortality, the ABIC index at hospital admission odds ratio was 19.27, whereas after 7 days, it was 41.29. The average survival of patients with ABIC of low and intermediate risk was 61.1 days, and for those with high risk, 26.2 days. Conclusions ABIC index is a predictor factor for complications such as ascites and hepatic encephalopathy, as well as for risk of death. Thus, it is a useful tool for clinical practice.

Sign in / Sign up

Export Citation Format

Share Document