Journal
COMPUTERS IN BIOLOGY AND MEDICINE
Volume 163, Issue -, Pages -Publisher
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.compbiomed.2023.107216
Keywords
Crohn's disease; Prediction; Feature selection; Ant colony optimization; Kernel extreme learning machine
Ask authors/readers for more resources
Changes in human lifestyles have led to a dramatic increase in the incidence of Crohn's disease worldwide. Predicting the activity and remission of Crohn's disease has become an urgent research problem. In this paper, a wrapper feature selection classification model called bIACOR-KELM-FS was proposed, which combines the improved ant colony optimization algorithm and the kernel extreme learning machine. The model showed a high prediction accuracy of 98.98% for predicting the activity and remission of Crohn's disease, and the analysis of important attributes improved the interpretability of the model and provided a reference for the diagnosis of Crohn's disease.
Changes in human lifestyles have led to a dramatic increase in the incidence of Crohn's disease worldwide. Predicting the activity and remission of Crohn's disease has become an urgent research problem. In addition, the influence of each attribute in the test sample on the prediction results and the interpretability of the model still deserves further investigation. Therefore, in this paper, we proposed a wrapper feature selection classification model based on a combination of the improved ant colony optimization algorithm and the kernel extreme learning machine, called bIACOR-KELM-FS. IACOR introduces an evasive strategy and astrophysics strategy to balance the exploration and exploitation phases of the algorithm and enhance its optimization capabilities. The optimization capability of the proposed IACOR was validated on the IEEE CEC2017 benchmark test function. And the prediction was performed on Crohn's disease dataset. The results of the quantitative analysis showed that the prediction accuracy of bIACOR-KELM-FS for predicting the activity and remission of Crohn's disease reached 98.98%. The analysis of important attributes improved the interpretability of the model and provided a reference for the diagnosis of Crohn's disease. Therefore, the proposed model is considered a promising adjunctive diag-nostic method for Crohn's disease.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available