Oscar L. Rueda-Ochoa
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Oscar L. Rueda-Ochoa has a total of 89 co-authors. Here are the top co-authors by number of shared publications:
- Maryam Kavousi
Department of Epidemiology, Erasmus MC, Rotterdam, The Netherlands
5 shared publications - Mohammad Arfan Ikram
Erasmus MC University Medical Center, Rotterdam, The Netherlands
5 shared publications - Oscar Horacio Franco
Utrecht University
4 shared publications - Hyeon Chang Kim
Department of Preventive Medicine, Yonsei University College of Medicine, Seoul, Korea
2 shared publications - Hyuk-Jae Chang
Division of Cardiology, Yonsei Cardiovascular Center, Yonsei University Health System, Seoul, South Korea
2 shared publications Khalid El Ghoul
From the Departments of Epidemiology (O.L.R.O., F.Z., M.K.I., M.W.V., O.H.F., D.B., M.K.), Cardiology (L.R.B., S.R., K.E.G., J.W.D., J.W.R.H.), Radiology and Nuclear Medicine (R.P.J.B., M.W.V., A.v.d.L., D.B.), and Neurology (M.K.I.), Erasmus MC, University Medical Center Rotterdam, PO Box 2040, 3000 CA Rotterdam, the Netherlands
2 shared publications
Latest Publications
- 1 July 2022
- Published by Radiological Society of North America (RSNA) in Radiology
- Vol. 304 (1) , 208-215
Thoracic Aortic Diameter and Cardiovascular Events and Mortality among Women and Men
Greater ascending and descending thoracic aortic diameters are risk factors for adverse cardiovascular outcomes and mortality in women and men. Background Thoracic aortic diameter may have a role as a biomarker for major adverse cardiovascular events. Purpose To evaluate the sex-specific association of the diameters of the ascending (AA) a...
- 1 July 2022
- Published by Radiological Society of North America (RSNA) in Radiology
- Vol. 304 (1) , 208-215
Thoracic Aortic Diameter and Cardiovascular Events and Mortality among Women and Men
Greater ascending and descending thoracic aortic diameters are risk factors for adverse cardiovascular outcomes and mortality in women and men. Background Thoracic aortic diameter may have a role as a biomarker for major adverse cardiovascular events. Purpose To evaluate the sex-specific association of the diameters of the ascending (AA) a...
- Open Access
- 5 March 2020
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 15 (3) , e0229576
The cardiovascular risk profile of middle age women previously diagnosed with premature ovarian insufficiency: A case-control study
Cardiovascular disease (CVD) is the leading cause of death in women worldwide. The cardiovascular risk profile deteriorates after women enter menopause. By definition, women diagnosed with premature ovarian insufficiency (POI) experience menopause before 40 years of age, which may render these women even more susceptible to develop CVD later in life. However, prospective long-term follow up data of well phenotyped women with POI are scarce. In the current study we compare the CVD profile and risk of middle aged women previously diagnosed with POI, to a population based reference group matched for age and BMI. We compared 123 women (age 49.0 (± 4.3) years) and diagnosed with POI 8.1 (IQR: 6.8–9.6) years earlier, with 123 population controls (age 49.4 (± 3.9) years). All women underwent an extensive standardized cardiovascular screening. We assessed CVD risk factors including waist circumference, BMI, blood pressure, lipid profile, pulse wave velocity (PWV), and the prevalence of diabetes mellitus, metabolic syndrome (MetS) and carotid intima media thickness (cIMT), in both women with POI and controls. We calculated the 10-year CVD Framingham Risk Score (FRS) and the American Heart Association’s suggested cardiovascular health score (CHS). Waist circumference (90.0 (IQR: 83.0–98.0) versus 80.7 (IQR: 75.1–86.8), p < 0.01), waist-to-hip ratio (0.90 (IQR: 0.85–0.93) versus 0.79 (IQR: 0.75–0.83), p < 0.01), systolic blood pressure (124 (IQR 112–135) versus 120 (IQR109-131), p < 0.04) and diastolic blood pressure (81 (IQR: 76–89) versus 78 (IQR: 71–86), p < 0.01), prevalence of hypertension (45 (37%) versus 21 (17%), p < 0.01) and MetS (19 (16%) versus 4 (3%), p < 0.01) were all significantly increased in women with POI compared to healthy controls. Other risk factors, however, such as lipids, glucose levels and prevalence of diabetes were similar comparing women with POI versus controls. The arterial stiffness assessed by PWV was also similar in both populations (8.1 (IQR: 7.1–9.4) versus 7.9 (IQR: 7.1–8.4), p = 0.21). In addition, cIMT was lower in women with POI compared to controls (550 μm (500–615) versus 684 μm (618–737), p < 0.01). The calculated 10-year CVD risk was 5.9% (IQR: 3.7–10.6) versus 6.0% (IQR: 3.9–9.0) (p = 0.31) and current CHS was 6.1 (1.9) versus 6.5 (1.6) (p = 0.07), respectively in POI versus controls. Middle age women with POI presented with more unfavorable cardiovascular risk factors (increased waist circumference and a higher prevalence of hypertension and MetS) compared to age and BMI matched population controls. In contrast, the current study reveals a lower cIMT and similar 10-year cardiovascular disease risk and cardiovascular health score. In summary, neither signs of premature atherosclerosis nor a worse cardiovascular disease risk or health score were observed among middle age women with POI compared to population controls. Longer-term follow-up studies of women of more advanced age are warranted to establish whether women with POI are truly at increased risk of developing CVD events later in life. ClinicalTrials.gov Identifier: NCT02616510.
- 5 March 2020
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 15 (3) , e0229576
Open AccessThe cardiovascular risk profile of middle age women previously diagnosed with premature ovarian insufficiency: A case-control study
Cardiovascular disease (CVD) is the leading cause of death in women worldwide. The cardiovascular risk profile deteriorates after women enter menopause. By definition, women diagnosed with premature ovarian insufficiency (POI) experience menopause before 40 years of age, which may render these women even more susceptible to develop CVD later in life. However, prospective long-term follow up data of well phenotyped women with POI are scarce. In the current study we compare the CVD profile and risk of middle aged women previously diagnosed with POI, to a population based reference group matched for age and BMI. We compared 123 women (age 49.0 (± 4.3) years) and diagnosed with POI 8.1 (IQR: 6.8–9.6) years earlier, with 123 population controls (age 49.4 (± 3.9) years). All women underwent an extensive standardized cardiovascular screening. We assessed CVD risk factors including waist circumference, BMI, blood pressure, lipid profile, pulse wave velocity (PWV), and the prevalence of diabetes mellitus, metabolic syndrome (MetS) and carotid intima media thickness (cIMT), in both women with POI and controls. We calculated the 10-year CVD Framingham Risk Score (FRS) and the American Heart Association’s suggested cardiovascular health score (CHS). Waist circumference (90.0 (IQR: 83.0–98.0) versus 80.7 (IQR: 75.1–86.8), p < 0.01), waist-to-hip ratio (0.90 (IQR: 0.85–0.93) versus 0.79 (IQR: 0.75–0.83), p < 0.01), systolic blood pressure (124 (IQR 112–135) versus 120 (IQR109-131), p < 0.04) and diastolic blood pressure (81 (IQR: 76–89) versus 78 (IQR: 71–86), p < 0.01), prevalence of hypertension (45 (37%) versus 21 (17%), p < 0.01) and MetS (19 (16%) versus 4 (3%), p < 0.01) were all significantly increased in women with POI compared to healthy controls. Other risk factors, however, such as lipids, glucose levels and prevalence of diabetes were similar comparing women with POI versus controls. The arterial stiffness assessed by PWV was also similar in both populations (8.1 (IQR: 7.1–9.4) versus 7.9 (IQR: 7.1–8.4), p = 0.21). In addition, cIMT was lower in women with POI compared to controls (550 μm (500–615) versus 684 μm (618–737), p < 0.01). The calculated 10-year CVD risk was 5.9% (IQR: 3.7–10.6) versus 6.0% (IQR: 3.9–9.0) (p = 0.31) and current CHS was 6.1 (1.9) versus 6.5 (1.6) (p = 0.07), respectively in POI versus controls. Middle age women with POI presented with more unfavorable cardiovascular risk factors (increased waist circumference and a higher prevalence of hypertension and MetS) compared to age and BMI matched population controls. In contrast, the current study reveals a lower cIMT and similar 10-year cardiovascular disease risk and cardiovascular health score. In summary, neither signs of premature atherosclerosis nor a worse cardiovascular disease risk or health score were observed among middle age women with POI compared to population controls. Longer-term follow-up studies of women of more advanced age are warranted to establish whether women with POI are truly at increased risk of developing CVD events later in life. ClinicalTrials.gov Identifier: NCT02616510.
- Open Access
- 7 February 2020
- Published by American Diabetes Association in Diabetes Care
- Vol. 43 (4) , 875-884
Epigenetic Link Between Statin Therapy and Type 2 Diabetes
OBJECTIVE To investigate the role of epigenetics in statins’ diabetogenic effect comparing DNA methylation (DNAm) between statin users and nonusers in an epigenome-wide association study in blood. RESEARCH DESIGN AND METHODS Five cohort studies’ participants (n = 8,270) were classified as statin users when they were on statin therapy at the time of DNAm assessment with Illumina 450K or EPIC array or noncurrent users otherwise. Associations of DNAm with various outcomes like incident type 2 diabetes, plasma glucose, insulin, and insulin resistance (HOMA of insulin resistance [HOMA-IR]) as well as with gene expression were investigated. RESULTS Discovery (n = 6,820) and replication (n = 1,450) phases associated five DNAm sites with statin use: cg17901584 (1.12 × 10−25 [DHCR24]), cg10177197 (3.94 × 10−08 [DHCR24]), cg06500161 (2.67 × 10−23 [ABCG1]), cg27243685 (6.01 × 10−09 [ABCG1]), and cg05119988 (7.26 × 10−12 [SC4MOL]). Two sites were associated with at least one glycemic trait or type 2 diabetes. Higher cg06500161 methylation was associated with higher fasting glucose, insulin, HOMA-IR, and type 2 diabetes (odds ratio 1.34 [95% CI 1.22, 1.47]). Mediation analyses suggested that ABCG1 methylation partially mediates the effect of statins on high insulin and HOMA-IR. Gene expression analyses showed that statin exposure and ABCG1 methylation were associated with ABCG1 downregulation, suggesting epigenetic regulation of ABCG1 expression. Further, outcomes insulin and HOMA-IR were significantly associated with ABCG1 expression. CONCLUSIONS This study sheds light on potential mechanisms linking statins with type 2 diabetes risk, providing evidence on DNAm partially mediating statins’ effects on insulin traits. Further efforts shall disentangle the molecular mechanisms through which statins may induce DNAm changes, potentially leading to ABCG1 epigenetic regulation.
- 7 February 2020
- Published by American Diabetes Association in Diabetes Care
- Vol. 43 (4) , 875-884
Open AccessEpigenetic Link Between Statin Therapy and Type 2 Diabetes
OBJECTIVE To investigate the role of epigenetics in statins’ diabetogenic effect comparing DNA methylation (DNAm) between statin users and nonusers in an epigenome-wide association study in blood. RESEARCH DESIGN AND METHODS Five cohort studies’ participants (n = 8,270) were classified as statin users when they were on statin therapy at the time of DNAm assessment with Illumina 450K or EPIC array or noncurrent users otherwise. Associations of DNAm with various outcomes like incident type 2 diabetes, plasma glucose, insulin, and insulin resistance (HOMA of insulin resistance [HOMA-IR]) as well as with gene expression were investigated. RESULTS Discovery (n = 6,820) and replication (n = 1,450) phases associated five DNAm sites with statin use: cg17901584 (1.12 × 10−25 [DHCR24]), cg10177197 (3.94 × 10−08 [DHCR24]), cg06500161 (2.67 × 10−23 [ABCG1]), cg27243685 (6.01 × 10−09 [ABCG1]), and cg05119988 (7.26 × 10−12 [SC4MOL]). Two sites were associated with at least one glycemic trait or type 2 diabetes. Higher cg06500161 methylation was associated with higher fasting glucose, insulin, HOMA-IR, and type 2 diabetes (odds ratio 1.34 [95% CI 1.22, 1.47]). Mediation analyses suggested that ABCG1 methylation partially mediates the effect of statins on high insulin and HOMA-IR. Gene expression analyses showed that statin exposure and ABCG1 methylation were associated with ABCG1 downregulation, suggesting epigenetic regulation of ABCG1 expression. Further, outcomes insulin and HOMA-IR were significantly associated with ABCG1 expression. CONCLUSIONS This study sheds light on potential mechanisms linking statins with type 2 diabetes risk, providing evidence on DNAm partially mediating statins’ effects on insulin traits. Further efforts shall disentangle the molecular mechanisms through which statins may induce DNAm changes, potentially leading to ABCG1 epigenetic regulation.
- Open Access
- 3 February 2020
- Published by Ubiquity Press, Ltd. in Global Heart
- Vol. 15 (1) , 36
Myocardial Involvement in Chagas Disease and Insulin Resistance: A Non-Metabolic Model of Cardiomyopathy
Heart failure (HF) and type 2 Diabetes Mellitus (T2DM) represent two chronic interrelated conditions accounting for significant morbidity and mortality worldwide. Insulin resistance (IR) has been identified as a risk factor for HF; however, the risk of IR that HF confers has not been well elucidated. The present study aims to analyze the association between myocardial involvement in Chronic Chagas Cardiomyopathy (CCM) and IR, taking advantage of this non-metabolic model of the disease. Cross-sectional study performed during the period 2015–2016. Adults with a serological diagnosis of Chagas disease were included, being divided into two groups: CCM and non-CCM. IR was determined by HOMA-IR index. Bivariate analysis and multivariate logistic regression were performed to determine the association between IR as an outcome and CCM as primary exposure. 200 patients were included in the study, with a mean age of 54.7 years and a female predominance (53.5%). Seventy-four (37.0%) patients were found to have IR, with a median HOMA-IR index of 3.9 (Q1 = 3.1; Q3 = 5.1). Multiple metabolic variables were significantly associated with IR. In a model analyzing only individuals with an altered HWI, an evident association between CCM and IR was observed (OR 4.08; 95% CI 1.55–10.73, p = 0.004). CCM was significantly associated with IR in patients with an altered HWI. The presence of this association in a non-metabolic model of HF (in which the myocardial involvement is expected to be mediated mostly by the parasitic infection) may support the evidence of a direct unidirectional correlation between this last and IR.
- 3 February 2020
- Published by Ubiquity Press, Ltd. in Global Heart
- Vol. 15 (1) , 36
Open AccessMyocardial Involvement in Chagas Disease and Insulin Resistance: A Non-Metabolic Model of Cardiomyopathy
Heart failure (HF) and type 2 Diabetes Mellitus (T2DM) represent two chronic interrelated conditions accounting for significant morbidity and mortality worldwide. Insulin resistance (IR) has been identified as a risk factor for HF; however, the risk of IR that HF confers has not been well elucidated. The present study aims to analyze the association between myocardial involvement in Chronic Chagas Cardiomyopathy (CCM) and IR, taking advantage of this non-metabolic model of the disease. Cross-sectional study performed during the period 2015–2016. Adults with a serological diagnosis of Chagas disease were included, being divided into two groups: CCM and non-CCM. IR was determined by HOMA-IR index. Bivariate analysis and multivariate logistic regression were performed to determine the association between IR as an outcome and CCM as primary exposure. 200 patients were included in the study, with a mean age of 54.7 years and a female predominance (53.5%). Seventy-four (37.0%) patients were found to have IR, with a median HOMA-IR index of 3.9 (Q1 = 3.1; Q3 = 5.1). Multiple metabolic variables were significantly associated with IR. In a model analyzing only individuals with an altered HWI, an evident association between CCM and IR was observed (OR 4.08; 95% CI 1.55–10.73, p = 0.004). CCM was significantly associated with IR in patients with an altered HWI. The presence of this association in a non-metabolic model of HF (in which the myocardial involvement is expected to be mediated mostly by the parasitic infection) may support the evidence of a direct unidirectional correlation between this last and IR.
- Open Access
- 1 January 2020
- Published by XMLink in Korean Circulation Journal
- Vol. 50 (1) , 72-84
Development and External Validation of a Deep Learning Algorithm for Prognostication of Cardiovascular Outcomes
We aim to explore the additional discriminative accuracy of a deep learning (DL) algorithm using repeated-measures data for identifying people at high risk for cardiovascular disease (CVD), compared to Cox hazard regression. Two CVD prediction models were developed from National Health Insurance Service-Health Screening Cohort (NHIS-HEALS): a Cox regression model and a DL model. Performance of each model was assessed in the internal and 2 external validation cohorts in Koreans (National Health Insurance Service-National Sample Cohort; NHIS-NSC) and in Europeans (Rotterdam Study). A total of 412,030 adults in the NHIS-HEALS; 178,875 adults in the NHIS-NSC; and the 4,296 adults in Rotterdam Study were included. Mean ages was 52 years (46% women) and there were 25,777 events (6.3%) in NHIS-HEALS during the follow-up. In internal validation, the DL approach demonstrated a C-statistic of 0.896 (95% confidence interval, 0.886-0.907) in men and 0.921 (0.908-0.934) in women and improved reclassification compared with Cox regression (net reclassification index [NRI], 24.8% in men, 29.0% in women). In external validation with NHIS-NSC, DL demonstrated a C-statistic of 0.868 (0.860-0.876) in men and 0.889 (0.876-0.898) in women, and improved reclassification compared with Cox regression (NRI, 24.9% in men, 26.2% in women). In external validation applied to the Rotterdam Study, DL demonstrated a C-statistic of 0.860 (0.824-0.897) in men and 0.867 (0.830-0.903) in women, and improved reclassification compared with Cox regression (NRI, 36.9% in men, 31.8% in women). A DL algorithm exhibited greater discriminative accuracy than Cox model approaches. ClinicalTrials.gov Identifier: NCT02931500.
- 1 January 2020
- Published by XMLink in Korean Circulation Journal
- Vol. 50 (1) , 72-84
Open AccessDevelopment and External Validation of a Deep Learning Algorithm for Prognostication of Cardiovascular Outcomes
We aim to explore the additional discriminative accuracy of a deep learning (DL) algorithm using repeated-measures data for identifying people at high risk for cardiovascular disease (CVD), compared to Cox hazard regression. Two CVD prediction models were developed from National Health Insurance Service-Health Screening Cohort (NHIS-HEALS): a Cox regression model and a DL model. Performance of each model was assessed in the internal and 2 external validation cohorts in Koreans (National Health Insurance Service-National Sample Cohort; NHIS-NSC) and in Europeans (Rotterdam Study). A total of 412,030 adults in the NHIS-HEALS; 178,875 adults in the NHIS-NSC; and the 4,296 adults in Rotterdam Study were included. Mean ages was 52 years (46% women) and there were 25,777 events (6.3%) in NHIS-HEALS during the follow-up. In internal validation, the DL approach demonstrated a C-statistic of 0.896 (95% confidence interval, 0.886-0.907) in men and 0.921 (0.908-0.934) in women and improved reclassification compared with Cox regression (net reclassification index [NRI], 24.8% in men, 29.0% in women). In external validation with NHIS-NSC, DL demonstrated a C-statistic of 0.868 (0.860-0.876) in men and 0.889 (0.876-0.898) in women, and improved reclassification compared with Cox regression (NRI, 24.9% in men, 26.2% in women). In external validation applied to the Rotterdam Study, DL demonstrated a C-statistic of 0.860 (0.824-0.897) in men and 0.867 (0.830-0.903) in women, and improved reclassification compared with Cox regression (NRI, 36.9% in men, 31.8% in women). A DL algorithm exhibited greater discriminative accuracy than Cox model approaches. ClinicalTrials.gov Identifier: NCT02931500.
Sex-specific distributions and determinants of thoracic aortic diameters in the elderly
Objective To provide population-based distributions of thoracic aortic diameters in men and women aged 55 years or older and to identify determinants of thoracic aortic diameters. Methods From 2003 to 2006, 2505 participants (1208 men, mean age 69.1±6.8 years) from the prospective population-based Rotterdam Study underwent non-enhanced cardiac CT. The diameter of the ascending (AA) and descending aorta (DA) was measured at the level of the pulmonary bifurcation. Results The mean diameter of the ascending and descending aorta was substantially larger in men (38±4 mm and 30±2 mm) than in women (35±3 mm and 27±2 mm). An ascending aortic diameter of larger than 40 mm was found in 228 (18.9%) men and 76 (5.9%) women and a descending aortic diameter larger than 40 mm was found in two men and no women. Male sex was found to be independently associated with larger DA diameter (standardised β 0.24, 95% CI 0.19 to 0.30), while a statistically non-significant trend was found for the AA diameter (standardised β 0.06, 95% CI 0.00 to 0.12). Age, height, weight and traditional cardiovascular risk factors were also associated with larger AA and/or DA diameters. Diabetes was associated with smaller AA and DA diameters. We found no evidence for effect modification by sex. Conclusions In persons aged 55 years or older, an ascending aortic diameter of 40 mm or larger was found in 18.9% of men and 5.9% of women. Given the importance of sex, sex-specific distribution values may prove useful in clinical practice, even when correcting for body surface area or height.
Sex-specific distributions and determinants of thoracic aortic diameters in the elderly
Objective To provide population-based distributions of thoracic aortic diameters in men and women aged 55 years or older and to identify determinants of thoracic aortic diameters. Methods From 2003 to 2006, 2505 participants (1208 men, mean age 69.1±6.8 years) from the prospective population-based Rotterdam Study underwent non-enhanced cardiac CT. The diameter of the ascending (AA) and descending aorta (DA) was measured at the level of the pulmonary bifurcation. Results The mean diameter of the ascending and descending aorta was substantially larger in men (38±4 mm and 30±2 mm) than in women (35±3 mm and 27±2 mm). An ascending aortic diameter of larger than 40 mm was found in 228 (18.9%) men and 76 (5.9%) women and a descending aortic diameter larger than 40 mm was found in two men and no women. Male sex was found to be independently associated with larger DA diameter (standardised β 0.24, 95% CI 0.19 to 0.30), while a statistically non-significant trend was found for the AA diameter (standardised β 0.06, 95% CI 0.00 to 0.12). Age, height, weight and traditional cardiovascular risk factors were also associated with larger AA and/or DA diameters. Diabetes was associated with smaller AA and DA diameters. We found no evidence for effect modification by sex. Conclusions In persons aged 55 years or older, an ascending aortic diameter of 40 mm or larger was found in 18.9% of men and 5.9% of women. Given the importance of sex, sex-specific distribution values may prove useful in clinical practice, even when correcting for body surface area or height.
- Open Access
- 19 September 2019
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 14 (9) , e0222809
Development and verification of prediction models for preventing cardiovascular diseases
Cardiovascular disease (CVD) is one of the major causes of death worldwide. For improved accuracy of CVD prediction, risk classification was performed using national time-series health examination data. The data offers an opportunity to access deep learning (RNN-LSTM), which is widely known as an outstanding algorithm for analyzing time-series datasets. The objective of this study was to show the improved accuracy of deep learning by comparing the performance of a Cox hazard regression and RNN-LSTM based on survival analysis. We selected 361,239 subjects (age 40 to 79 years) with more than two health examination records from 2002–2006 using the National Health Insurance System-National Health Screening Cohort (NHIS-HEALS). The average number of health screenings (from 2002–2013) used in the analysis was 2.9 ± 1.0. Two CVD prediction models were developed from the NHIS-HEALS data: a Cox hazard regression model and a deep learning model. In an internal validation of the NHIS-HEALS dataset, the Cox regression model showed a highest time-dependent area under the curve (AUC) of 0.79 (95% CI 0.70 to 0.87) for in females and 0.75 (95% CI 0.70 to 0.80) in males at 2 years. The deep learning model showed a highest time-dependent AUC of 0.94 (95% CI 0.91 to 0.97) for in females and 0.96 (95% CI 0.95 to 0.97) in males at 2 years. Layer-wise Relevance Propagation (LRP) revealed that age was the variable that had the greatest effect on CVD, followed by systolic blood pressure (SBP) and diastolic blood pressure (DBP), in that order. The performance of the deep learning model for predicting CVD occurrences was better than that of the Cox regression model. In addition, it was confirmed that the known risk factors shown to be important by previous clinical studies were extracted from the study results using LRP.
- 19 September 2019
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 14 (9) , e0222809
Open AccessDevelopment and verification of prediction models for preventing cardiovascular diseases
Cardiovascular disease (CVD) is one of the major causes of death worldwide. For improved accuracy of CVD prediction, risk classification was performed using national time-series health examination data. The data offers an opportunity to access deep learning (RNN-LSTM), which is widely known as an outstanding algorithm for analyzing time-series datasets. The objective of this study was to show the improved accuracy of deep learning by comparing the performance of a Cox hazard regression and RNN-LSTM based on survival analysis. We selected 361,239 subjects (age 40 to 79 years) with more than two health examination records from 2002–2006 using the National Health Insurance System-National Health Screening Cohort (NHIS-HEALS). The average number of health screenings (from 2002–2013) used in the analysis was 2.9 ± 1.0. Two CVD prediction models were developed from the NHIS-HEALS data: a Cox hazard regression model and a deep learning model. In an internal validation of the NHIS-HEALS dataset, the Cox regression model showed a highest time-dependent area under the curve (AUC) of 0.79 (95% CI 0.70 to 0.87) for in females and 0.75 (95% CI 0.70 to 0.80) in males at 2 years. The deep learning model showed a highest time-dependent AUC of 0.94 (95% CI 0.91 to 0.97) for in females and 0.96 (95% CI 0.95 to 0.97) in males at 2 years. Layer-wise Relevance Propagation (LRP) revealed that age was the variable that had the greatest effect on CVD, followed by systolic blood pressure (SBP) and diastolic blood pressure (DBP), in that order. The performance of the deep learning model for predicting CVD occurrences was better than that of the Cox regression model. In addition, it was confirmed that the known risk factors shown to be important by previous clinical studies were extracted from the study results using LRP.
- Open Access
Risk factors for longitudinal changes in left ventricular diastolic function among women and men
Objective To evaluate changes in left ventricular diastolic function (LVDF) parameters and their associated risk factors over a period of 11 years among community-dwelling women and men. Methods Echocardiography was performed three times among 870 women and 630 men (age 67±3 years) from the prospective population-based Rotterdam Study during a period of 11-year follow-up. Changes in six continuous LVDF parameters were correlated with cardiovascular risk factors using a linear-mixed effect model (LMM). Results In women, smoking was associated with deleterious longitudinal changes in deceleration time (DT) (Beta (β): 7.73; 95% CI 2.56 to 12.9) and high-density lipoprotein cholesterol was associated with improvement of septal e′ (β: 0.37; 95% CI 0.13 to 0.62) and E/e′ ratio (β: −0.46; 95% CI −0.84 to –0.08) trajectories. Among men, diabetes was associated with deleterious longitudinal changes in A wave (β: 3.83; 95% CI 0.06 to 7.60), septal e′ (β: −0.40; 95% CI −0.70 to –0.09) and E/e′ ratio (β: 0.60; 95% CI 0.14 to 1.06) and body mass index was associated with deleterious longitudinal changes in A wave (β: 1.25; 95% CI 0.84 to 1.66), E/A ratio (β: −0.007; 95% CI −0.01 to –0.003), DT (β: 0.86; 95% CI 0.017 to 1.71) and E/e′ ratio (β: 0.12; 95% CI 0.06 to 0.19). Conclusions Smoking among women and metabolic factors (diabetes mellitus and body mass index) among men showed larger deleterious associations with longitudinal changes in LVDF parameters. The favourable association of HDL was mainly observed among women. This study, for the first time, evaluates risk factors associated with changes over time in continuous LVDF parameters among women and men and generates new hypothesis for further medical research.
Open AccessRisk factors for longitudinal changes in left ventricular diastolic function among women and men
Objective To evaluate changes in left ventricular diastolic function (LVDF) parameters and their associated risk factors over a period of 11 years among community-dwelling women and men. Methods Echocardiography was performed three times among 870 women and 630 men (age 67±3 years) from the prospective population-based Rotterdam Study during a period of 11-year follow-up. Changes in six continuous LVDF parameters were correlated with cardiovascular risk factors using a linear-mixed effect model (LMM). Results In women, smoking was associated with deleterious longitudinal changes in deceleration time (DT) (Beta (β): 7.73; 95% CI 2.56 to 12.9) and high-density lipoprotein cholesterol was associated with improvement of septal e′ (β: 0.37; 95% CI 0.13 to 0.62) and E/e′ ratio (β: −0.46; 95% CI −0.84 to –0.08) trajectories. Among men, diabetes was associated with deleterious longitudinal changes in A wave (β: 3.83; 95% CI 0.06 to 7.60), septal e′ (β: −0.40; 95% CI −0.70 to –0.09) and E/e′ ratio (β: 0.60; 95% CI 0.14 to 1.06) and body mass index was associated with deleterious longitudinal changes in A wave (β: 1.25; 95% CI 0.84 to 1.66), E/A ratio (β: −0.007; 95% CI −0.01 to –0.003), DT (β: 0.86; 95% CI 0.017 to 1.71) and E/e′ ratio (β: 0.12; 95% CI 0.06 to 0.19). Conclusions Smoking among women and metabolic factors (diabetes mellitus and body mass index) among men showed larger deleterious associations with longitudinal changes in LVDF parameters. The favourable association of HDL was mainly observed among women. This study, for the first time, evaluates risk factors associated with changes over time in continuous LVDF parameters among women and men and generates new hypothesis for further medical research.