Construction of a Risk Prediction Model for Hospital-Acquired Pulmonary Embolism in Hospitalized Patients

Hou, Lengchen; Hu, Longjun; Gao, Wenxue; Sheng, Wenbo; Hao, Zedong; Chen, Yiwei; Li, Jiyu

doi:10.1177/10760296211040868

Cited by 11 publications

(13 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, there is still a major inconsistency among the potential predictors of VTE identified by previous studies [ 24 ]. In this context, machine learning (ML) techniques, which can identify complex (non-linear) correlations among potential predictors, may be useful tools [ 25 ]. However, to the best of our knowledge, the use of ML as an approach to assess VTE predictors in COVID-19 patients has not yet been reported.…”

Section: Introductionmentioning

confidence: 99%

Predictors of venous thromboembolism in COVID-19 patients: results of the COVID-19 Brazilian Registry

et al. 2022

View full text Add to dashboard Cite

Previous studies that assessed risk factors for venous thromboembolism (VTE) in COVID-19 patients have shown inconsistent results. Our aim was to investigate VTE predictors by both logistic regression (LR) and machine learning (ML) approaches, due to their potential complementarity. This cohort study of a large Brazilian COVID-19 Registry included 4120 COVID-19 adult patients from 16 hospitals. Symptomatic VTE was confirmed by objective imaging. LR analysis, tree-based boosting, and bagging were used to investigate the association of variables upon hospital presentation with VTE. Among 4,120 patients (55.5% men, 39.3% critical patients), VTE was confirmed in 6.7%. In multivariate LR analysis, obesity (OR 1.50, 95% CI 1.11–2.02); being an ex-smoker (OR 1.44, 95% CI 1.03–2.01); surgery ≤ 90 days (OR 2.20, 95% CI 1.14–4.23); axillary temperature (OR 1.41, 95% CI 1.22–1.63); D-dimer ≥ 4 times above the upper limit of reference value (OR 2.16, 95% CI 1.26–3.67), lactate (OR 1.10, 95% CI 1.02–1.19), C-reactive protein levels (CRP, OR 1.09, 95% CI 1.01–1.18); and neutrophil count (OR 1.04, 95% CI 1.005–1.075) were independent predictors of VTE. Atrial fibrillation, peripheral oxygen saturation/inspired oxygen fraction (SF) ratio and prophylactic use of anticoagulants were protective. Temperature at admission, SF ratio, neutrophil count, D-dimer, CRP and lactate levels were also identified as predictors by ML methods. By using ML and LR analyses, we showed that D-dimer, axillary temperature, neutrophil count, CRP and lactate levels are risk factors for VTE in COVID-19 patients. Supplementary Information The online version contains supplementary material available at 10.1007/s11739-022-03002-z.

show abstract

Section: Introductionmentioning

confidence: 99%

Predictors of venous thromboembolism in COVID-19 patients: results of the COVID-19 Brazilian Registry

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Non-ICU COVID-19 patients with a D-dimer level ≥2,000 ng/mL are considered to need further examination to rule out PE (Thoreau et al, 2021). Data from a 3,619 study of PE showed that high D-dimer levels had almost been proved to be significantly associated with the severity of PE (Hou et al, 2021). Another trial looked at the percentage decrease in D-dimer concentration between at diagnosis and within 1 month of diagnosis.…”

Section: Discussionmentioning

confidence: 99%

A new machine learning model for predicting severity prognosis in patients with pulmonary embolism: Study protocol from Wenzhou, China

Shou

et al. 2022

Front. Neuroinform.

View full text Add to dashboard Cite

IntroductionPulmonary embolism (PE) is a common thrombotic disease and potentially deadly cardiovascular disorder. The ratio of clinical misdiagnosis and missed diagnosis of PE is very large because patients with PE are asymptomatic or non-specific.MethodsUsing the clinical data from the First Affiliated Hospital of Wenzhou Medical University (Wenzhou, China), we proposed a swarm intelligence algorithm-based kernel extreme learning machine model (SSACS-KELM) to recognize and discriminate the severity of the PE by patient’s basic information and serum biomarkers. First, an enhanced method (SSACS) is presented by combining the salp swarm algorithm (SSA) with the cuckoo search (CS). Then, the SSACS algorithm is introduced into the KELM classifier to propose the SSACS-KELM model to improve the accuracy and stability of the traditional classifier.ResultsIn the experiments, the benchmark optimization performance of SSACS is confirmed by comparing SSACS with five original classical methods and five high-performance improved algorithms through benchmark function experiments. Then, the overall adaptability and accuracy of the SSACS-KELM model are tested using eight public data sets. Further, to highlight the superiority of SSACS-KELM on PE datasets, this paper conducts comparison experiments with other classical classifiers, swarm intelligence algorithms, and feature selection approaches.DiscussionThe experimental results show that high D-dimer concentration, hypoalbuminemia, and other indicators are important for the diagnosis of PE. The classification results showed that the accuracy of the prediction model was 99.33%. It is expected to be a new and accurate method to distinguish the severity of PE.

show abstract

“…that explored in-and post-hospital VTE at different time periods, were included (see Supplementary Table 1). Diagnosis of VTE was either confirmed via International Classification of Diseases (ICD) codes 31,32,36,37 , administration of anticoagulation treatment 36,37 or imaging studies 30,33,35,49 . Most of the published studies are retrospective and are derived from a single institution except for 36,50 which are derived from multiple centers.…”

Section: Prediction Of Vte Risk In Non-surgical Hospitalized Patientsmentioning

confidence: 99%

“…Most of the published studies are retrospective and are derived from a single institution except for 36,50 which are derived from multiple centers. Ma et al 35 and Hou et al 33 addressed the temporal characteristics of the laboratory tests, such as min, max, last measurement, and time windows for some dynamic laboratory features. Most of the studies did not provide clear information regarding feature selection, 31,37,49,50 preprocessing and missing values handling 31,36,37,49,50 and hyperparameter tuning 31,33,36,49 .…”

Section: Prediction Of Vte Risk In Non-surgical Hospitalized Patientsmentioning

confidence: 99%

“…Ma et al 35 and Hou et al 33 addressed the temporal characteristics of the laboratory tests, such as min, max, last measurement, and time windows for some dynamic laboratory features. Most of the studies did not provide clear information regarding feature selection, 31,37,49,50 preprocessing and missing values handling 31,36,37,49,50 and hyperparameter tuning 31,33,36,49 . Interestingly, methodological pitfalls in the ML architecture that can lead to data leakage have been identified, such as handling of missing values with imputation methods before This article is protected by copyright.…”

Section: Prediction Of Vte Risk In Non-surgical Hospitalized Patientsmentioning

confidence: 99%

See 1 more Smart Citation

Machine Learning-Based Predictive Models for Patients with Venous Thromboembolism: A Systematic Review

Danilatou,

Dimopoulos,

Kostoulas

et al. 2024

Thromb Haemost

View full text Add to dashboard Cite

Background: Venous thromboembolism (VTE) is a chronic disorder with a significant health and economic burden. Several VTE-specific Clinical Prediction Models (CPMs) have been used to assist physicians in decision-making but have several limitations. This systematic review explores if machine learning (ML) can enhance CPMs by analyzing extensive patient data derived from electronic health records (EHRs). We aimed to explore ML-CPMs applications in VTE for risk stratification, outcome prediction, diagnosis, and treatment. Methods: Three databases were searched, PubMed, Google Scholar, and IEEE electronic library. Inclusion criteria focused on studies using structured data, excluding non-English publications, studies on non-humans, and certain data types such as natural language processing and image processing. Studies involving pregnant women, cancer patients, and children were also excluded. After excluding irrelevant studies, a total of 77 studies were included. Results: Most studies report that ML-CPMs outperformed traditional CPMs in terms of receiver operating area under the curve in the four clinical domains that were explored. However, the majority of the studies were retrospective, monocentric, and lacked detailed model architecture description and external validation, which are essential for quality audit. This review identified research gaps and highlighted challenges related to standardized reporting, reproducibility, and model comparison. Conclusion: ML-CPMs show promise in improving risk assessment and individualized treatment recommendations in VTE. Apparently, there is an urgent need for standardized reporting and methodology for ML models, external validation, prospective and real-world data studies, as well as interventional studies to evaluate the impact of AI in VTE.

show abstract

Construction of a Risk Prediction Model for Hospital-Acquired Pulmonary Embolism in Hospitalized Patients

Cited by 11 publications

References 41 publications

Predictors of venous thromboembolism in COVID-19 patients: results of the COVID-19 Brazilian Registry

Predictors of venous thromboembolism in COVID-19 patients: results of the COVID-19 Brazilian Registry

A new machine learning model for predicting severity prognosis in patients with pulmonary embolism: Study protocol from Wenzhou, China

Machine Learning-Based Predictive Models for Patients with Venous Thromboembolism: A Systematic Review

Contact Info

Product

Resources

About