Journal: Cognitive Computation

Article Title: An Integrated Deep Learning and Belief Rule Base Intelligent System to Predict Survival of COVID-19 Patient under Uncertainty

doi: 10.1007/s12559-021-09978-8

Figure Lengend Snippet: Trained Belief Rule Base Using fmincon for Patient Survival Probability (Y) where input antecedents are X1: Patient Condition, X2: Blood Pressure, X3: Chronic Obstructive Pulmonary Disease, X4: Blood Sugar, X5: Asthma, X6: Chronic Kidney Disease, X7: Obesity, X8: Acute Respiratory Distress Syndrome, X9: Pulse Oximetry, Critical: C, Non Critical: NC, Hypertensive Crisis: HC, Very Severely Abnormal: VSA, Diabetic: D, Severe Persistent: SP, Very Severe: VS, Level III: L-III, Severe: S, Moderate: M, Normal: N, Mild: Mi, VH: Very High, H: High, M: Medium, L: Low, VL: Very Low)

Article Snippet: In order to ensure the reliability of the rules in the initial BRB, the BRBES is trained using the non-linear optimization solver fmincon in MATLAB optimization toolbox [ ], Belief Rule-Based Adaptive Particle Swarm Optimization (BRBAPSO) [ ], and the enhanced Belief Rule-Based adaptive Differential Evolution (eBRBaDE) algorithm [ ].

Techniques:

Journal: Cognitive Computation

Article Title: An Integrated Deep Learning and Belief Rule Base Intelligent System to Predict Survival of COVID-19 Patient under Uncertainty

doi: 10.1007/s12559-021-09978-8

Figure Lengend Snippet: Comparison of AUC of CNN-BRBES and Expert Opinion

Article Snippet: In order to ensure the reliability of the rules in the initial BRB, the BRBES is trained using the non-linear optimization solver fmincon in MATLAB optimization toolbox [ ], Belief Rule-Based Adaptive Particle Swarm Optimization (BRBAPSO) [ ], and the enhanced Belief Rule-Based adaptive Differential Evolution (eBRBaDE) algorithm [ ].

Techniques: Comparison