Comparison of Machine Learning Methods With National Cardiovascular Data Registry Models for Prediction of Risk of Bleeding After Percutaneous Coronary Intervention

Mortazavi, Bobak J.; Bucholz, Emily M.; Desai, Nihar R.; Huang, Chenxi; Curtis, Jeptha P.; Masoudi, Frederick A.; Shaw, Richard E.; Negahban, Sahand; Krumholz, Harlan M.

doi:10.1001/jamanetworkopen.2019.6835

Cited by 68 publications

(52 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In the specific clinical context of predicting mortality from gastrointestinal bleeding, a systematic review demonstrated higher c ‐indices and predictive capacity with ML compared with clinical risk scores 38 . Another specific study aiming to predict bleeding risk following percutaneous coronary intervention reported that ML better characterized bleeding risk than a standard registry model 39 . An important distinction of our review was that we included disease‐specific studies that were published in the computer science literature, which were underrepresented in the aforementioned earlier comparisons of ML and CSMs.…”

Section: Discussionmentioning

confidence: 99%

Machine learning vs. conventional statistical models for predicting heart failure readmission and mortality

et al. 2020

View full text Add to dashboard Cite

Aims This study aimed to review the performance of machine learning (ML) methods compared with conventional statistical models (CSMs) for predicting readmission and mortality in patients with heart failure (HF) and to present an approach to formally evaluate the quality of studies using ML algorithms for prediction modelling. Methods and results Following Preferred Reporting Items for Systematic Reviews and Meta‐Analyses guidelines, we performed a systematic literature search using MEDLINE, EPUB, Cochrane CENTRAL, EMBASE, INSPEC, ACM Library, and Web of Science. Eligible studies included primary research articles published between January 2000 and July 2020 comparing ML and CSMs in mortality and readmission prognosis of initially hospitalized HF patients. Data were extracted and analysed by two independent reviewers. A modified CHARMS checklist was developed in consultation with ML and biostatistics experts for quality assessment and was utilized to evaluate studies for risk of bias. Of 4322 articles identified and screened by two independent reviewers, 172 were deemed eligible for a full‐text review. The final set comprised 20 articles and 686 842 patients. ML methods included random forests (n = 11), decision trees (n = 5), regression trees (n = 3), support vector machines (n = 9), neural networks (n = 12), and Bayesian techniques (n = 3). CSMs included logistic regression (n = 16), Cox regression (n = 3), or Poisson regression (n = 3). In 15 studies, readmission was examined at multiple time points ranging from 30 to 180 day readmission, with the majority of studies (n = 12) presenting prediction models for 30 day readmission outcomes. Of a total of 21 time‐point comparisons, ML‐derived c‐indices were higher than CSM‐derived c‐indices in 16 of the 21 comparisons. In seven studies, mortality was examined at 9 time points ranging from in‐hospital mortality to 1 year survival; of these nine, seven reported higher c‐indices using ML. Two of these seven studies reported survival analyses utilizing random survival forests in their ML prediction models. Both reported higher c‐indices when using ML compared with CSMs. A limitation of studies using ML techniques was that the majority were not externally validated, and calibration was rarely assessed. In the only study that was externally validated in a separate dataset, ML was superior to CSMs (c‐indices 0.913 vs. 0.835). Conclusions ML algorithms had better discrimination than CSMs in most studies aiming to predict risk of readmission and mortality in HF patients. Based on our review, there is a need for external validation of ML‐based studies of prediction modelling. We suggest that ML‐based studies should also be evaluated using clinical quality standards for prognosis research. Registration: PROSPERO CRD42020134867

show abstract

Section: Discussionmentioning

confidence: 99%

Machine learning vs. conventional statistical models for predicting heart failure readmission and mortality

et al. 2020

View full text Add to dashboard Cite

show abstract

“…We first applied the XGBoost method using the gradient boosting decision tree algorithm to construct a risk prediction model. The model generates the importance score of each feature automatically by calculating the average improvement for each feature of the model after it was introduced to a branch [21][22][23][24]. A higher importance score indicated that the variable had a higher predictive value for the model.…”

Section: Prediction Model Using the Extreme Gradient Boosting Methodsmentioning

confidence: 99%

Machine Learning for Prediction and Risk Stratification of Lupus Nephritis Renal Flare

Chen

Huang²,

Chen³

et al. 2021

Am J Nephrol

View full text Add to dashboard Cite

Background: Renal flare of lupus nephritis (LN) is strongly associated with poor kidney outcomes, and predicting renal flare and stratifying its risk are important for clinical decision-making and individualized management to reduce LN flare. Methods: We randomly divided 1,694 patients with biopsy-proven LN, who had achieved remission after treatment, into a derivation cohort (n = 1,186) and an internal validation cohort (n = 508), at a ratio of 7:3. The risk of renal flare 5 years after remission was predicted using an eXtreme Gradient Boosting (XGBoost) method model, developed from 59 variables, including demographic, clinical, immunological, pathological, and therapeutic characteristics. A simplified risk score prediction model (SRSPM) was developed from important variables selected by XGBoost model using stepwise Cox regression for practical convenience. Results: The 5-year relapse rates were 39.5% and 38.2% in the derivation and internal validation cohorts, respectively. Both the XGBoost model and the SRSPM had good predictive performance, with a C-index of 0.819 (95% confidence interval [CI]: 0.774–0.857) and 0.746 (95% CI: 0.697–0.795), respectively, in the validation cohort. The SRSPM comprised 6 variables, including partial remission and endocapillary hypercellularity at baseline, age, serum Alb, anti-dsDNA, and serum complement C3 at the point of remission. Using Kaplan-Meier analysis, the SRSPM identified significant risk stratification for renal flares (p < 0.001). Conclusions: Renal flare of LN can be readily predicted using the XGBoost model and the SRSPM, and the SRSPM can also stratify flare risk. Both models are useful for clinical decision-making and individualized management in LN.

show abstract

“…When using the data, it was randomly divided into two parts, with 80% used for developing window behavior models and 20% for model validation. This division has been popularly adopted in existing studies [45,46,49,51,61]. Table 3 has listed the calculated Pearson Correlation Coefficient for each potential influential factor considered in this study.…”

Section: Resultsmentioning

confidence: 99%

“…As a new machine learning method, the XGBoost (eXtreme Gradient Boost) method was firstly introduced by Chen [37] in 2016, and has been used in many other applications, such as automotive manufacturing [38], predicting building cooling load [39] and fault detection for HVAC systems [40]. In existing studies, much evidence was available about its advantages (stability, accuracy and efficiency) in modeling complex process over other conventional machine learning methods, such as SVM algorithm [41,42], logistic regression method [43][44][45][46][47][48][49] and KNN/decision tree [50,51]. This study, therefore, was designed to justify its contribution to modeling accuracy of occupant window behavior in buildings, mainly against the most conventional modeling approach, i.e.…”

Section: Introductionmentioning

confidence: 99%

Developing window behavior models for residential buildings using XGBoost algorithm

Hao

Sun

Liu

et al. 2019

Energy and Buildings

120

View full text Add to dashboard Cite

Buildings account for over 32% of total society energy consumption, and to make buildings more energy efficient dynamic building performance simulation has been widely adopted during the buildings' design to help select most appropriate HVAC (Heating Ventilation and Air Conditioning) systems. Due to the lack of good behavioral models in current simulation packages, many researchers have tried to develop useful behavioral models to improve simulation accuracy, including window behavior models, using field data collected from real buildings. During this work, many mathematical and machine learning methods have been used, and some level of prediction accuracy has been achieved. XGBoost is a recently introduced machine learning algorithm, which has been proven as very powerful in modeling complicated processes in other research fields. In this study, this algorithm has been adopted to model and predict occupant window behavior, aiming to further improve the modeling accuracy from a globally accepted modeling approach, namely, Logistic Regression Analysis. Field data in terms of both occupant window behavior and relevant influential factors were collected from real residential buildings during transitional seasons. Both XGBoost and Logistic Regression Analysis were used to build window behavior models, after a feature selection work, and their prediction performances on an independent dataset were compared. The comparison revealed that XGBoost has solid advantages in modeling occupant window behavior, over Logistic Regression Analysis, and it is expecting that the same finding would be obtained for other behavioral types, such as blind control and air-conditioner operation.

show abstract

Comparison of Machine Learning Methods With National Cardiovascular Data Registry Models for Prediction of Risk of Bleeding After Percutaneous Coronary Intervention

Cited by 68 publications

References 20 publications

Machine learning vs. conventional statistical models for predicting heart failure readmission and mortality

Machine learning vs. conventional statistical models for predicting heart failure readmission and mortality

Machine Learning for Prediction and Risk Stratification of Lupus Nephritis Renal Flare

Developing window behavior models for residential buildings using XGBoost algorithm

Contact Info

Product

Resources

About