scholarly journals A machine learning method for subgroup analysis of randomized controlled trials

2018 ◽  
Author(s):  
Ljubomir Buturović
Keyword(s):  
Machine Learning ◽  
Randomized Controlled Trials ◽  
Controlled Trials ◽  
Machine Learning Method ◽  
Learning Method ◽  
Randomized Controlled ◽  
Subgroup Analyses ◽  
Treatment Of Hypertension ◽  
Multiple Predictor ◽  
General Method

AbstractWe developed a machine learning method for subgroup analyses of randomized controlled trials (RCT), and applied it to the results of the SPRINT RCT for treatment of hypertension. To date, the subgroup analyses mostly focused on detecting associations between certain factors and outcome, in the hope that the results will point out biologically (for example, carriers of a certain mutation) or clinically (for example, smokers) distinct subgroups with different outcomes. This seldom worked in the sense of re-launching the intervention for the detected subgroup only and successfully treating it. In contrast, we propose an empirical and general method to develop a predictive multivariate classifier using the RCT outcomes and baseline data. The classifier identifies patients likely to benefit from the intervention, is not limited to a single factor of interest, and is ready for validation in a subsequent pivotal trial. We believe this approach has a better chance of succeeding in identifying the relevant subgroups because of increased accuracy made possible by the use of multiple predictor variables, and opportunity to use advanced machine learning. The method effectiveness is demonstrated by the analysis of the SPRINT trial.

scholarly journals Remote Ischemic Preconditioning to Prevent Acute Kidney Injury After Cardiac Surgery: A Meta-Analysis of Randomized Controlled Trials

2021 ◽  
Vol 8 ◽  
Author(s):  
Zigang Liu ◽  
Yongmei Zhao ◽  
Ming Lei ◽  
Guancong Zhao ◽  
Dongcheng Li ◽  
...  
Keyword(s):  
Acute Kidney Injury ◽  
Cardiac Surgery ◽  
Randomized Controlled Trials ◽  
Ischemic Preconditioning ◽  
Meta Analysis ◽  
Kidney Injury ◽  
Controlled Trials ◽  
Randomized Controlled ◽  
Subgroup Analyses ◽  
Meta Regression

Objective: Randomized controlled trials (RCTs) evaluating the influence of remote ischemic preconditioning (RIPC) on acute kidney injury (AKI) after cardiac surgery showed inconsistent results. We performed a meta-analysis to evaluate the efficacy of RIPC on AKI after cardiac surgery.Methods: Relevant studies were obtained by search of PubMed, Embase, and Cochrane's Library databases. A random-effect model was used to pool the results. Meta-regression and subgroup analyses were used to determine the source of heterogeneity.Results: Twenty-two RCTs with 5,389 patients who received cardiac surgery −2,702 patients in the RIPC group and 2,687 patients in the control group—were included. Moderate heterogeneity was detected (p for Cochrane's Q test = 0.03, I2 = 40%). Pooled results showed that RIPC significantly reduced the incidence of AKI compared with control [odds ratio (OR): 0.76, 95% confidence intervals (CI): 0.61–0.94, p = 0.01]. Results limited to on-pump surgery (OR: 0.78, 95% CI: 0.64–0.95, p = 0.01) or studies with acute RIPC (OR: 0.78, 95% CI: 0.63–0.97, p = 0.03) showed consistent results. Meta-regression and subgroup analyses indicated that study characteristics, including study design, country, age, gender, diabetic status, surgery type, use of propofol or volatile anesthetics, cross-clamp time, RIPC protocol, definition of AKI, and sample size did not significantly affect the outcome of AKI. Results of stratified analysis showed that RIPC significantly reduced the risk of mild-to-moderate AKI that did not require renal replacement therapy (RRT, OR: 0.76, 95% CI: 0.60–0.96, p = 0.02) but did not significantly reduce the risk of severe AKI that required RRT in patients after cardiac surgery (OR: 0.73, 95% CI: 0.50–1.07, p = 0.11).Conclusions: Current evidence supports RIPC as an effective strategy to prevent AKI after cardiac surgery, which seems to be mainly driven by the reduced mild-to-moderate AKI events that did not require RRT. Efforts are needed to determine the influences of patient characteristics, procedure, perioperative drugs, and RIPC protocol on the outcome.

scholarly journals Machine learning for identifying Randomized Controlled Trials: An evaluation and practitioner's guide

2018 ◽  
Vol 9 (4) ◽  
pp. 602-614 ◽  
Author(s):  
Iain J. Marshall ◽  
Anna Noel-Storr ◽  
Joël Kuiper ◽  
James Thomas ◽  
Byron C. Wallace
Keyword(s):  
Machine Learning ◽  
Randomized Controlled Trials ◽  
Controlled Trials ◽  
Randomized Controlled

scholarly journals Treatment of hypertension in patients 80 years and older: the lower the better? A meta-analysis of randomized controlled trials

2010 ◽  
Vol 28 (7) ◽  
pp. 1366-1372 ◽  
Author(s):  
Theodora Bejan-Angoulvant ◽  
Mitra Saadatian-Elahi ◽  
James M Wright ◽  
Eleanor B Schron ◽  
Lars H Lindholm ◽  
...  
Keyword(s):  
Randomized Controlled Trials ◽  
Meta Analysis ◽  
Controlled Trials ◽  
Randomized Controlled ◽  
Treatment Of Hypertension

scholarly journals Deep Neural Network for Reducing the Screening Workload in Systematic Reviews for Clinical Guidelines: Algorithm Validation Study

10.2196/22422 ◽  
2020 ◽  
Vol 22 (12) ◽  
pp. e22422
Author(s):  
Tomohide Yamada ◽  
Daisuke Yoneoka ◽  
Yuta Hiraike ◽  
Kimihiro Hino ◽  
Hiroyoshi Toyoshiba ◽  
...  
Keyword(s):  
Machine Learning ◽  
Randomized Controlled Trials ◽  
Systematic Reviews ◽  
Clinical Guidelines ◽  
Learning System ◽  
Controlled Trials ◽  
Data Set ◽  
Randomized Controlled ◽  
Fold Reduction ◽  
Meta Analyses

Background Performing systematic reviews is a time-consuming and resource-intensive process. Objective We investigated whether a machine learning system could perform systematic reviews more efficiently. Methods All systematic reviews and meta-analyses of interventional randomized controlled trials cited in recent clinical guidelines from the American Diabetes Association, American College of Cardiology, American Heart Association (2 guidelines), and American Stroke Association were assessed. After reproducing the primary screening data set according to the published search strategy of each, we extracted correct articles (those actually reviewed) and incorrect articles (those not reviewed) from the data set. These 2 sets of articles were used to train a neural network–based artificial intelligence engine (Concept Encoder, Fronteo Inc). The primary endpoint was work saved over sampling at 95% recall (WSS@95%). Results Among 145 candidate reviews of randomized controlled trials, 8 reviews fulfilled the inclusion criteria. For these 8 reviews, the machine learning system significantly reduced the literature screening workload by at least 6-fold versus that of manual screening based on WSS@95%. When machine learning was initiated using 2 correct articles that were randomly selected by a researcher, a 10-fold reduction in workload was achieved versus that of manual screening based on the WSS@95% value, with high sensitivity for eligible studies. The area under the receiver operating characteristic curve increased dramatically every time the algorithm learned a correct article. Conclusions Concept Encoder achieved a 10-fold reduction of the screening workload for systematic review after learning from 2 randomly selected studies on the target topic. However, few meta-analyses of randomized controlled trials were included. Concept Encoder could facilitate the acquisition of evidence for clinical guidelines.

Effect of quercetin supplementation on plasma lipid profiles, blood pressure, and glucose levels: a systematic review and meta-analysis

2020 ◽  
Vol 78 (8) ◽  
pp. 615-626 ◽  
Author(s):  
Haohai Huang ◽  
Dan Liao ◽  
Yong Dong ◽  
Rong Pu
Keyword(s):  
Randomized Controlled Trials ◽  
Meta Analysis ◽  
Plasma Lipid ◽  
Density Lipoprotein ◽  
Lipid Profiles ◽  
Controlled Trials ◽  
Parallel Design ◽  
Glucose Levels ◽  
Randomized Controlled ◽  
Subgroup Analyses

Abstract Context Clinical trials examining the cardiovascular protective effects of quercetin in humans have reported conflicting results. Objective The aim of this systematic review was to summarize evidence of the effects of quercetin supplementation on plasma lipid profiles, blood pressure (BP), and glucose levels in humans by performing a meta-analysis of randomized controlled trials. Data Sources MEDLINE, Embase, and Scopus databases were searched electronically from their inception to July 2018 to identify randomized controlled trials that assessed the impact of quercetin on lipid profiles, BP, and glucose levels. Study Selection Randomized controlled trials assessing the effects of quercetin or a standardized quercetin-enriched extract on plasma lipid profiles, BP, and glucose levels in humans were eligible for inclusion. Data Extraction A random-effects model was used for data analysis. Continuous variables were expressed as weighted mean differences (WMDs) and 95%CIs. Subgroup analyses were conducted to explore possible influences of study characteristics. Sensitivity analyses were also performed, as were analyses of publication bias. Results Seventeen trials (n = 896 participants total) were included in the overall analysis. Pooled results showed that quercetin significantly lowered both systolic BP (WMD, −3.09 mmHg; 95%CI, −4.59 to −1.59; P = 0.0001) and diastolic BP (WMD, −2.86 mmHg; 95%CI, −5.09 to −0.63; P = 0.01). Neither lipid profiles nor glucose concentrations changed significantly. In subgroup analyses, significant changes in high-density lipoprotein cholesterol and triglycerides were observed in trials with a parallel design and in which participants consumed quercetin for 8 weeks or more. Conclusion Quercetin intake resulted in significantly decreased BP in humans. Moreover, participants who consumed quercetin for 8 weeks or more showed significantly changed levels of high-density lipoprotein cholesterol and triglycerides in trials with a parallel design.

Sign in / Sign up

Export Citation Format

Share Document