In this paper, we first evaluated the distribution of blood pressure (BP) using a dataset containing 113,979 measurements in 48,022 individuals with the cooperation of one health insurance society in Japan from April,...In this paper, we first evaluated the distribution of blood pressure (BP) using a dataset containing 113,979 measurements in 48,022 individuals with the cooperation of one health insurance society in Japan from April, 2013 to March, 2016. The means of the systolic BP (SBP) and diastolic BP (DBP) were 125.4 and 77.6 mmHg with standard deviations of 16.5 and 11.7 mmHg, respectively. Under the 140/90 criterion, 21.6% of the measurements showed hypertension. According to the World Health Organization/International Society of Hypertension criterion, 16.4%, 4.2% and 0.96% were classified as grades 1, 2 and 3, respectively. The factors affecting BP were evaluated by a regression analysis and were found to include age, gender, some eating habits, daily activities, smoking, drinking alcohol, sleeping and wages. Age was a very important factor, and the age cohorts from the previous study might be revised based on these findings. Among factors that individuals can control, the influence of drinking alcohol is very large. Comparing to an individual who does not drink, SBP and DBP of a heavy drinker are more than 5.0 mmHg higher on the average.展开更多
In this study,single-channel photoplethysmography(PPG)signals were used to estimate the heart rate(HR),diastolic blood pressure(DBP),and systolic blood pressure(SBP).A deep learning model was proposed using a long-ter...In this study,single-channel photoplethysmography(PPG)signals were used to estimate the heart rate(HR),diastolic blood pressure(DBP),and systolic blood pressure(SBP).A deep learning model was proposed using a long-term recurrent convolutional network(LRCN)modified from a deep learning algorithm,the convolutional neural network model of the modified inception deep learning module,and a long short-term memory network(LSTM)to improve the model’s accuracy of BP and HR measurements.The PPG data of 1,551 patients were obtained from the University of California Irvine Machine Learning Repository.How to design a filter of PPG signals and how to choose the loss functions for deep learning model were also discussed in the study.Finally,the stability of the proposed model was tested using a 10-fold cross-validation,with an MAE±SD of 2.942±5.076 mmHg for SBP,1.747±3.042 mmHg for DBP,and 1.137±2.463 bpm for the HR.Compared with its existing counterparts,the model entailed less computational load and was more accurate in estimating SBP,DBP,and HR.These results established the validity of the model.展开更多
<strong>Background: </strong>The high blood pressure (BP) or hypertension is a widely prevalent disease and its costs are very high, and many studies about the relationships between BP and health condition...<strong>Background: </strong>The high blood pressure (BP) or hypertension is a widely prevalent disease and its costs are very high, and many studies about the relationships between BP and health conditions have been done. We need to know the precise distributions of BP and factors affecting BP. <strong>Data and Methods</strong><strong>:</strong> The distributions of BP are analyzed using 12,877,653 observations obtained from the JMDC Claims Database. The factors that may affect the BP are analyzed by the regression models using 4,615,346 observations. <strong>Results:</strong> The averages of systolic BP (SBP) and diastolic BP (DBP) are 120.4 and 74.2 mmHg with standard deviations of 15.9 and 11.3 mmHg, respectively. Among the nonmodifiable factors, age and gender are important factors. Among the modifiable factors, variables related to obesity are important risk factors. Taking antihypertensive drugs makes SBP and DBP 13.4 mmHg and 7.8 mmHg lower. <strong>Conclusion:</strong> The criteria of BP should be carefully determined considering age and gender. The effects of age may be a little different for SBP and DBP. It is necessary to use the proper model to evaluate the effect of antihypertensive drugs correctly. <strong>Limitations:</strong> The dataset is observatory. Although there are various types of treatment methods and antihypertension drugs, their effects are not evaluated.展开更多
目的探讨不同程度高血压和正常血压人群应用两种不同测量血压的方法,进行分析对比血压值的差异。方法2 000例患者分为4组:正常血压组,高血压1级组,高血压2级组,高血压3级组。把听诊器胸件分别放置于血压计袖带内和袖带外测量血压并记录...目的探讨不同程度高血压和正常血压人群应用两种不同测量血压的方法,进行分析对比血压值的差异。方法2 000例患者分为4组:正常血压组,高血压1级组,高血压2级组,高血压3级组。把听诊器胸件分别放置于血压计袖带内和袖带外测量血压并记录。结果正常血压组:血压平均值袖带外118.7/75 mm Hg,袖带内113.2/71.3 mm Hg,高血压1级组袖带外144.8/84.3 mm Hg,袖带内140.4/82.3 mm Hg,高血压2级组袖带外157.8/95.9 mm Hg,袖带内149.3/92.9 mm Hg,高血压3级组袖带外181.5/110.6 mm Hg,袖带内178.6/114.3 mm Hg。结论正常血压组,高血压1级组,高血压2级组,袖带外收缩压和舒张压均显著高于袖带内血压(P<0.01,P<0.05),高血压3级组袖带内外均差异无统计学意义(P>0.05)。展开更多
文摘In this paper, we first evaluated the distribution of blood pressure (BP) using a dataset containing 113,979 measurements in 48,022 individuals with the cooperation of one health insurance society in Japan from April, 2013 to March, 2016. The means of the systolic BP (SBP) and diastolic BP (DBP) were 125.4 and 77.6 mmHg with standard deviations of 16.5 and 11.7 mmHg, respectively. Under the 140/90 criterion, 21.6% of the measurements showed hypertension. According to the World Health Organization/International Society of Hypertension criterion, 16.4%, 4.2% and 0.96% were classified as grades 1, 2 and 3, respectively. The factors affecting BP were evaluated by a regression analysis and were found to include age, gender, some eating habits, daily activities, smoking, drinking alcohol, sleeping and wages. Age was a very important factor, and the age cohorts from the previous study might be revised based on these findings. Among factors that individuals can control, the influence of drinking alcohol is very large. Comparing to an individual who does not drink, SBP and DBP of a heavy drinker are more than 5.0 mmHg higher on the average.
基金This study was supported in part by the Ministry of Science and Technology MOST108-2221-E-150-022-MY3 and Taiwan Ocean University.
文摘In this study,single-channel photoplethysmography(PPG)signals were used to estimate the heart rate(HR),diastolic blood pressure(DBP),and systolic blood pressure(SBP).A deep learning model was proposed using a long-term recurrent convolutional network(LRCN)modified from a deep learning algorithm,the convolutional neural network model of the modified inception deep learning module,and a long short-term memory network(LSTM)to improve the model’s accuracy of BP and HR measurements.The PPG data of 1,551 patients were obtained from the University of California Irvine Machine Learning Repository.How to design a filter of PPG signals and how to choose the loss functions for deep learning model were also discussed in the study.Finally,the stability of the proposed model was tested using a 10-fold cross-validation,with an MAE±SD of 2.942±5.076 mmHg for SBP,1.747±3.042 mmHg for DBP,and 1.137±2.463 bpm for the HR.Compared with its existing counterparts,the model entailed less computational load and was more accurate in estimating SBP,DBP,and HR.These results established the validity of the model.
文摘<strong>Background: </strong>The high blood pressure (BP) or hypertension is a widely prevalent disease and its costs are very high, and many studies about the relationships between BP and health conditions have been done. We need to know the precise distributions of BP and factors affecting BP. <strong>Data and Methods</strong><strong>:</strong> The distributions of BP are analyzed using 12,877,653 observations obtained from the JMDC Claims Database. The factors that may affect the BP are analyzed by the regression models using 4,615,346 observations. <strong>Results:</strong> The averages of systolic BP (SBP) and diastolic BP (DBP) are 120.4 and 74.2 mmHg with standard deviations of 15.9 and 11.3 mmHg, respectively. Among the nonmodifiable factors, age and gender are important factors. Among the modifiable factors, variables related to obesity are important risk factors. Taking antihypertensive drugs makes SBP and DBP 13.4 mmHg and 7.8 mmHg lower. <strong>Conclusion:</strong> The criteria of BP should be carefully determined considering age and gender. The effects of age may be a little different for SBP and DBP. It is necessary to use the proper model to evaluate the effect of antihypertensive drugs correctly. <strong>Limitations:</strong> The dataset is observatory. Although there are various types of treatment methods and antihypertension drugs, their effects are not evaluated.
文摘目的探讨不同程度高血压和正常血压人群应用两种不同测量血压的方法,进行分析对比血压值的差异。方法2 000例患者分为4组:正常血压组,高血压1级组,高血压2级组,高血压3级组。把听诊器胸件分别放置于血压计袖带内和袖带外测量血压并记录。结果正常血压组:血压平均值袖带外118.7/75 mm Hg,袖带内113.2/71.3 mm Hg,高血压1级组袖带外144.8/84.3 mm Hg,袖带内140.4/82.3 mm Hg,高血压2级组袖带外157.8/95.9 mm Hg,袖带内149.3/92.9 mm Hg,高血压3级组袖带外181.5/110.6 mm Hg,袖带内178.6/114.3 mm Hg。结论正常血压组,高血压1级组,高血压2级组,袖带外收缩压和舒张压均显著高于袖带内血压(P<0.01,P<0.05),高血压3级组袖带内外均差异无统计学意义(P>0.05)。
