PURPOSE: Hemophilia A and B (HA, HB) are the most common X-linked inherited bleeding disorders. The introduction of factor concentrates has allowed for control of the lifelong chronic disease. However, no studies have been published regarding the epidemiology of hemophilia in Taiwan. Our aim was to determine the prevalence, incidence, and mortality rate, as well as trends in the use of factor concentrates, in individuals with hemophilia in Taiwan. MATERIALS AND METHODS: A retrospective study was conducted using the National Health Insurance Research Database between 1997 and 2007. RESULTS: We identified 988 males with hemophilia (HA : HB ratio=5.4 : 1). The mean prevalence per 100000 males was 6.7+/-0.1 for HA and 1.2+/-0.1 for HB. The estimated mean annual incidence per live male birth was 1 in 10752 for HA and 1 in 47619 for HB. Standardized mortality ratios for males with hemophilia (all severities) or severe hemophilia were 1.3- and 2.1-fold higher than that of the general male population, respectively. Mean factor VIII (FVIII) and factor IX (FIX) usage was 1.5003+/-0.4029 and 0.3126+/-0.0904 international units (IUs) per capita, respectively. Mean FVIII and FIX usage per patient with hemophilia (all severities) or severe hemophilia was 44027+/-11532 and 72341+/-17298, respectively, and 49407+/-13015 and 74369+/-18411 IUs per person with HA or HB, respectively. CONCLUSION: Our data revealed epidemiologic and factor concentrate usage trends in males with hemophilia in Taiwan, highlighting a need for improvements in the mandatory National Health Insurance registry. A better-designed, patient-centered registry system would enable more detailed patient information collection and analysis, improving subsequent care.
Adolescent ; Adult ; Aged ; Child ; Child, Preschool ; Databases, Factual ; Factor IX/therapeutic use ; Factor VIII/therapeutic use ; Hemophilia A/*drug therapy/*epidemiology/ethnology ; Hemophilia B/*drug therapy/*epidemiology/ethnology ; Humans ; Incidence ; Infant ; Male ; Middle Aged ; Prevalence ; Registries ; Retrospective Studies ; Taiwan/epidemiology ; Young Adult

Background: This study aimed to develop a diabetic kidney disease (DKD) prediction model using long short term memory (LSTM) neural network and evaluate its performance using accuracy, precision, recall, and area under the curve (AUC) of the receiver operating characteristic (ROC) curve. Methods: The study identified DKD risk factors through literature review and physician focus group, and collected 7 years of data from 6,040 type 2 diabetes mellitus patients based on the risk factors. Pytorch was used to build the LSTM neural network, with 70% of the data used for training and the other 30% for testing. Three models were established to examine the impact of glycosylated hemoglobin (HbA1c), systolic blood pressure (SBP), and pulse pressure (PP) variabilities on the model’s performance. Results: The developed model achieved an accuracy of 83% and an AUC of 0.83. When the risk factor of HbA1c variability, SBP variability, or PP variability was removed one by one, the accuracy of each model was significantly lower than that of the optimal model, with an accuracy of 78% (P<0.001), 79% (P<0.001), and 81% (P<0.001), respectively. The AUC of ROC was also significantly lower for each model, with values of 0.72 (P<0.001), 0.75 (P<0.001), and 0.77 (P<0.05). Conclusion The developed DKD risk predictive model using LSTM neural networks demonstrated high accuracy and AUC value. When HbA1c, SBP, and PP variabilities were added to the model as featured characteristics, the model’s performance was greatly improved.

Background: This study aimed to develop a diabetic kidney disease (DKD) prediction model using long short term memory (LSTM) neural network and evaluate its performance using accuracy, precision, recall, and area under the curve (AUC) of the receiver operating characteristic (ROC) curve. Methods: The study identified DKD risk factors through literature review and physician focus group, and collected 7 years of data from 6,040 type 2 diabetes mellitus patients based on the risk factors. Pytorch was used to build the LSTM neural network, with 70% of the data used for training and the other 30% for testing. Three models were established to examine the impact of glycosylated hemoglobin (HbA1c), systolic blood pressure (SBP), and pulse pressure (PP) variabilities on the model’s performance. Results: The developed model achieved an accuracy of 83% and an AUC of 0.83. When the risk factor of HbA1c variability, SBP variability, or PP variability was removed one by one, the accuracy of each model was significantly lower than that of the optimal model, with an accuracy of 78% (P<0.001), 79% (P<0.001), and 81% (P<0.001), respectively. The AUC of ROC was also significantly lower for each model, with values of 0.72 (P<0.001), 0.75 (P<0.001), and 0.77 (P<0.05). Conclusion The developed DKD risk predictive model using LSTM neural networks demonstrated high accuracy and AUC value. When HbA1c, SBP, and PP variabilities were added to the model as featured characteristics, the model’s performance was greatly improved.

Background: This study aimed to develop a diabetic kidney disease (DKD) prediction model using long short term memory (LSTM) neural network and evaluate its performance using accuracy, precision, recall, and area under the curve (AUC) of the receiver operating characteristic (ROC) curve. Methods: The study identified DKD risk factors through literature review and physician focus group, and collected 7 years of data from 6,040 type 2 diabetes mellitus patients based on the risk factors. Pytorch was used to build the LSTM neural network, with 70% of the data used for training and the other 30% for testing. Three models were established to examine the impact of glycosylated hemoglobin (HbA1c), systolic blood pressure (SBP), and pulse pressure (PP) variabilities on the model’s performance. Results: The developed model achieved an accuracy of 83% and an AUC of 0.83. When the risk factor of HbA1c variability, SBP variability, or PP variability was removed one by one, the accuracy of each model was significantly lower than that of the optimal model, with an accuracy of 78% (P<0.001), 79% (P<0.001), and 81% (P<0.001), respectively. The AUC of ROC was also significantly lower for each model, with values of 0.72 (P<0.001), 0.75 (P<0.001), and 0.77 (P<0.05). Conclusion The developed DKD risk predictive model using LSTM neural networks demonstrated high accuracy and AUC value. When HbA1c, SBP, and PP variabilities were added to the model as featured characteristics, the model’s performance was greatly improved.