Prediction of in-hospital adverse clinical outcomes in patients with pulmonary thromboembolism, machine learning based models
BackgroundPulmonary thromboembolism (PE) is the third leading cause of cardiovascular events. The conventional modeling methods and severity risk scores lack multiple laboratories, paraclinical and imaging data. Data science and machine learning (ML) based prediction models may help better predict o...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-03-01
|
| Series: | Frontiers in Cardiovascular Medicine |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fcvm.2023.1087702/full |
| _version_ | 1797871464580382720 |
|---|---|
| author | Yaser Jenab Kaveh Hosseini Zahra Esmaeili Saeed Tofighi Hamid Ariannejad Houman Sotoudeh |
| author_facet | Yaser Jenab Kaveh Hosseini Zahra Esmaeili Saeed Tofighi Hamid Ariannejad Houman Sotoudeh |
| author_sort | Yaser Jenab |
| collection | DOAJ |
| description | BackgroundPulmonary thromboembolism (PE) is the third leading cause of cardiovascular events. The conventional modeling methods and severity risk scores lack multiple laboratories, paraclinical and imaging data. Data science and machine learning (ML) based prediction models may help better predict outcomes.Materials and methodsIn this retrospective registry-based design, all consecutive hospitalized patients diagnosed with pulmonary thromboembolism (based on pulmonary CT angiography) from 2011 to 2019 were recruited. ML based algorithms [Gradient Boosting (GB) and Deep Learning (DL)] were applied and compared with logistic regression (LR) to predict hemodynamic instability and/or all-cause mortality.ResultsA total number of 1,017 patients were finally enrolled in the study, including 465 women and 552 men. Overall incidence of study main endpoint was 9.6%, (7.2% in men and 12.4% in women; p-value = 0.05). The overall performance of the GB model is better than the other two models (AUC: 0.94 for GB vs. 0.88 and 0.90 for DL and LR models respectively). Based on GB model, lower O2 saturation and right ventricle dilation and dysfunction were among the strongest adverse event predictors.ConclusionML-based models have notable prediction ability in PE patients. These algorithms may help physicians to detect high-risk patients earlier and take appropriate preventive measures. |
| first_indexed | 2024-04-10T00:44:34Z |
| format | Article |
| id | doaj.art-fa4761166d48473fb298e82db10ec602 |
| institution | Directory Open Access Journal |
| issn | 2297-055X |
| language | English |
| last_indexed | 2024-04-10T00:44:34Z |
| publishDate | 2023-03-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Cardiovascular Medicine |
| spelling | doaj.art-fa4761166d48473fb298e82db10ec6022023-03-14T04:48:44ZengFrontiers Media S.A.Frontiers in Cardiovascular Medicine2297-055X2023-03-011010.3389/fcvm.2023.10877021087702Prediction of in-hospital adverse clinical outcomes in patients with pulmonary thromboembolism, machine learning based modelsYaser Jenab0Kaveh Hosseini1Zahra Esmaeili2Saeed Tofighi3Hamid Ariannejad4Houman Sotoudeh5Tehran Heart Center, Cardiovascular Diseases Research Institute, Tehran University of Medical Sciences, Tehran, IranTehran Heart Center, Cardiovascular Diseases Research Institute, Tehran University of Medical Sciences, Tehran, IranFaculty of Medicine, Tehran University of Medical Sciences, Tehran, IranTehran Heart Center, Cardiovascular Diseases Research Institute, Tehran University of Medical Sciences, Tehran, IranTehran Heart Center, Cardiovascular Diseases Research Institute, Tehran University of Medical Sciences, Tehran, IranDepartment of Radiology, University of Alabama at Birmingham (UAB), Birmingham, AL, United StatesBackgroundPulmonary thromboembolism (PE) is the third leading cause of cardiovascular events. The conventional modeling methods and severity risk scores lack multiple laboratories, paraclinical and imaging data. Data science and machine learning (ML) based prediction models may help better predict outcomes.Materials and methodsIn this retrospective registry-based design, all consecutive hospitalized patients diagnosed with pulmonary thromboembolism (based on pulmonary CT angiography) from 2011 to 2019 were recruited. ML based algorithms [Gradient Boosting (GB) and Deep Learning (DL)] were applied and compared with logistic regression (LR) to predict hemodynamic instability and/or all-cause mortality.ResultsA total number of 1,017 patients were finally enrolled in the study, including 465 women and 552 men. Overall incidence of study main endpoint was 9.6%, (7.2% in men and 12.4% in women; p-value = 0.05). The overall performance of the GB model is better than the other two models (AUC: 0.94 for GB vs. 0.88 and 0.90 for DL and LR models respectively). Based on GB model, lower O2 saturation and right ventricle dilation and dysfunction were among the strongest adverse event predictors.ConclusionML-based models have notable prediction ability in PE patients. These algorithms may help physicians to detect high-risk patients earlier and take appropriate preventive measures.https://www.frontiersin.org/articles/10.3389/fcvm.2023.1087702/fullmachine learingoutcome analysisrisk factorslogistic modelsgradient boosting machinepulmonary embolism |
| spellingShingle | Yaser Jenab Kaveh Hosseini Zahra Esmaeili Saeed Tofighi Hamid Ariannejad Houman Sotoudeh Prediction of in-hospital adverse clinical outcomes in patients with pulmonary thromboembolism, machine learning based models Frontiers in Cardiovascular Medicine machine learing outcome analysis risk factors logistic models gradient boosting machine pulmonary embolism |
| title | Prediction of in-hospital adverse clinical outcomes in patients with pulmonary thromboembolism, machine learning based models |
| title_full | Prediction of in-hospital adverse clinical outcomes in patients with pulmonary thromboembolism, machine learning based models |
| title_fullStr | Prediction of in-hospital adverse clinical outcomes in patients with pulmonary thromboembolism, machine learning based models |
| title_full_unstemmed | Prediction of in-hospital adverse clinical outcomes in patients with pulmonary thromboembolism, machine learning based models |
| title_short | Prediction of in-hospital adverse clinical outcomes in patients with pulmonary thromboembolism, machine learning based models |
| title_sort | prediction of in hospital adverse clinical outcomes in patients with pulmonary thromboembolism machine learning based models |
| topic | machine learing outcome analysis risk factors logistic models gradient boosting machine pulmonary embolism |
| url | https://www.frontiersin.org/articles/10.3389/fcvm.2023.1087702/full |
| work_keys_str_mv | AT yaserjenab predictionofinhospitaladverseclinicaloutcomesinpatientswithpulmonarythromboembolismmachinelearningbasedmodels AT kavehhosseini predictionofinhospitaladverseclinicaloutcomesinpatientswithpulmonarythromboembolismmachinelearningbasedmodels AT zahraesmaeili predictionofinhospitaladverseclinicaloutcomesinpatientswithpulmonarythromboembolismmachinelearningbasedmodels AT saeedtofighi predictionofinhospitaladverseclinicaloutcomesinpatientswithpulmonarythromboembolismmachinelearningbasedmodels AT hamidariannejad predictionofinhospitaladverseclinicaloutcomesinpatientswithpulmonarythromboembolismmachinelearningbasedmodels AT houmansotoudeh predictionofinhospitaladverseclinicaloutcomesinpatientswithpulmonarythromboembolismmachinelearningbasedmodels |