Developing a Genetic Fuzzy System for Risk Assessment of Mortality After Cardiac Surgery

Abstract

Cardiac events could be taken into account as the leading causes of death throughout the globe. Such events also trigger an undesirable increase in what treatment procedures cost. Despite the giant leaps in technological development in heart surgery, coronary surgery still carries the high risk of the mortality. Besides, there is still a long way ahead to accurately predict and assess the mortality risk. This study is an attempt to develop an expert system for the risk assessment of mortality following the cardiac surgery. The developed system involves three main steps. In the first step, a filtering feature selection method is applied to select the best features. In the second step, an ad hoc data-driven method is utilized to generate the preliminary fuzzy inference system. Finally, a hybrid optimization method is presented to select the optimum subset of the rules. The study relies on 1,811 samples to evaluate the diagnosis performance of the proposed system. The obtained classification accuracy is very promising with regard to other benchmark classification methods including binary logistic regression (LR) and multilayer perceptron neural network (MLP) with the same attributes. The developed system leads to 100 % sensitivity and 84.7 % specificity, while LR and MLP methods statistically come up with lower figures (65, 78.6 and 65 %, 75.8 %), respectively. Now, a fuzzy supportive tool can be potentially taken as an alternative for the current mortality risk assessment system that are applied in coronary surgeries, and are chiefly based on crisp database.

Appendix

Begin Lookup

i = 0 {Iteration counter for number of input/output variable (N)}

 For i = 0 to N do

   Divide the space of variable (i) into k _i fuzzy regions.

     i = i + 1

 End

j = 0 {Iteration counter for number of rule (n-Rule = ∏ M _i )}

k = 0{Iteration counter for number of variable fuzzy regions}

l = 0 {Iteration counter for number of training data (T)}

 For j = 0 to n-Rule do

   Generate IF-part of Rule (j).

   For k = 0 to M _end do

    Generate THEN-part of Rule (j).

    For l = 0 to T

     Compute Compatibility grade of Rule (j).

      l = l + 1

    End

    k = k + 1

   End

   j = j + 1

  End

For j = 0 to n-Rule do

 Add the Rule with the maximum degree of consequence part in Rule base.

   IF Degree (Rule (j)) < =Threshold

   Delete Rule (j)

  End

 j = j + 1

End

Create primary fuzzy inference system (FIS).

End Lookup

Begin GFA

g = 0 {Iteration counter for number of generation (MaxIt)}

Initialize Rule base pop (g).

Evaluate pop (g) by Fitness function (defined in equation 3).

For g = 1 to MaxIt

 Select pop (g) from pop (g-1)

 Crossover {pop (g)}

 Mutate {pop (g)}

 Replace pop (g) with pop (g-1) in FIS.

 Evaluate pop (g) by fitness function.

 Choose the best pop (g)

 Search local space around the best pop (g)

 Update the mutation rate by the annealing equation.

  g = g + 1

End

End GFA