Dho-gml Optimization Approach for Thyroid Classification Using Artificial Intelligence

Disease classification using Artificial Intelligence (AI) is one of the emerging areas for medical professionals to diagnose the disease. Thyroid is a common disease which is faced by most of the people that produce severe health problems. As a second opinion to medical professionals, various classification models are used such as Logistic Regression, Decision Tree, Random Forest, K Nearest Neighbor and gradient boosting algorithms.An Artificial Neural Network (ANN) is a part of AI mainly used to identify the nonlinear relationship among the features for better prediction. However, there are some common problems like under fitting and overfitting that occur during the analysis and prediction. The present study suggests the Deep Hyper Optimization (DHO) - Genetic Meta Learning (GML) technique to increase the accuracy and reduce the elapsed time of execution without overfitting. In the first phase, DHO is used to fine tune the weight, bias and identify the optimal number of parameters in the hidden layer using DHO. The performance of various algorithms are observed and compared with Deep Hyper Optimization (DHO) technique. It chooses the best parameter for classification based on the highest probability of particular value in the parameters after performing the divide and conquer approach. In the second phase, GML technique is used to identify the optimal model for classification. It uses the crossover to select the best model and generate offspring population. The performance of each method is observed based on the accuracy, elapsed time, precision, recall, and F1 score, FPR, FNR, TNR and Area under Curve (AUC).