An evidential reasoning extension of model-based failure diagnosis

AbstractA novel decision-making method based on evidential reasoning is proposed for solving the two-sided matching problem with uncertain information under multiple states in this paper. Firstly, the discernment frame of evidence is constructed for two-sided matching. Secondly, the preference ordinal values given by two-sided decision-makers are transformed into rank belief degrees. On this basis, and with two-sided satisfaction as the goal, two-sided rank belief degrees are taken as pieces of evidence, and satisfaction degrees of two-sided matching are obtained through evidence fusion. Then, a decision-making model based on satisfaction degrees is constructed in order to obtain the matching solution. Finally, an illustrative example demonstrates the application of the proposed approach.

Download Full-text

Machine learning analysis of metabolomic biomarkers for diagnosis of heart failure

European Heart Journal ◽

10.1093/eurheartj/ehab724.0864 ◽

2021 ◽

Vol 42 (Supplement_1) ◽

Author(s):

C Coorey ◽

O Tang ◽

J.Y.H Yang ◽

G Figtree

Keyword(s):

Machine Learning ◽

Heart Failure ◽

Prediction Model ◽

Failure Diagnosis ◽

Cardiac Biomarkers ◽

Support Vector ◽

Private Company ◽

Model Based ◽

Cardiovascular Assessment ◽

Heart Failure Diagnosis

Abstract Background There is emerging evidence that the pathophysiological mechanisms of heart failure are associated with alterations in serum metabolites. Such metabolomic signatures may be useful for heart failure diagnosis, stratification and prognosis. Purpose To evaluate the utility of including metabolomic biomarkers in addition to traditional cardiac biomarkers in a machine learning prediction model of heart failure diagnosis in the well-characterised Canagliflozin Cardiovascular Assessment Study (CANVAS) cohort. Methods A subgroup of the CANVAS/CANVAS-R study cohort was analysed. 101 metabolites in plasma were measured by HPLC (HILIC)-mass spectrometry. A 10-times 5-fold cross-validated support vector machine model with radial basis kernel function was constructed to predict heart failure diagnosis using traditional biomarkers alone and using the combination of traditional biomarkers and metabolomic biomarkers. Model performance and variable importance were both evaluated by area under the curve (AUC) of the receiver operating characteristics (ROC) curve. Results are shown as mean ± standard deviation. Results 967 patients (of which 402 patients had heart failure) were included in the analysis with 341 females, mean age 63±8 years and mean body mass index (BMI) 33±5 kg/m2. All patients had diabetes mellitus with mean HbA1c 8.2±0.9%. The prediction model based on only traditional biomarkers had mean AUC 72±3% and the prediction model based on both traditional biomarkers and metabolomic biomarkers had mean AUC 80±3%. The top metabolomic biomarkers for predicting heart failure were threonine, L-homoserine, creatine and deoxyadenosine. Conclusion Metabolomic biomarkers improved diagnostic performance of a heart failure prediction model and captured variation not encompassed by traditional cardiac biomarkers. FUNDunding Acknowledgement Type of funding sources: Private company. Main funding source(s): Janssen Research and Development

Download Full-text