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.

BACKGROUND: We aimed to compare the diagnostic accuracy of the Alere Triage Cardio3 Tropinin I (TnI) assay (Alere, Inc., USA) and the PathFast cTnI-II (Mitsubishi Chemical Medience Corporation, Japan) against the central laboratory assay Singulex Erenna TnI assay (Singulex, USA). METHODS: Using the Markers in the Diagnosis of Acute Coronary Syndromes (MIDAS) study population, we evaluated the ability of three different assays to identify patients with acute myocardial infarction (AMI). The MIDAS dataset, described elsewhere, is a prospective multicenter dataset of emergency department (ED) patients with suspected acute coronary syndrome (ACS) and a planned objective myocardial perfusion evaluation. Myocardial infarction (MI) was diagnosed by central adjudication. RESULTS: The C-statistic with 95% confidence intervals (CI) for diagnosing MI by using a common population (n=241) was 0.95 (0.91-0.99), 0.95 (0.91-0.99), and 0.93 (0.89-0.97) for the Triage, Singulex, and PathFast assays, respectively. Of samples with detectable troponin, the absolute values had high Pearson (R(P)) and Spearman (R(S)) correlations and were R(P)=0.94 and R(S)=0.94 for Triage vs Singulex, R(P)=0.93 and R(S)=0.85 for Triage vs PathFast, and R(P)=0.89 and R(S)=0.73 for PathFast vs Singulex. CONCLUSIONS: In a single comparative population of ED patients with suspected ACS, the Triage Cardio3 TnI, PathFast, and Singulex TnI assays provided similar diagnostic performance for MI.
Acute Coronary Syndrome/*diagnosis ; Biomarkers/analysis ; Emergency Service, Hospital ; Humans ; Laboratories/standards ; Myocardial Infarction/diagnosis ; *Point-of-Care Systems ; Prospective Studies ; Reagent Kits, Diagnostic ; Sensitivity and Specificity ; Troponin I/*analysis