Deep feature selection for cervical cancer

Abstract

Cervical cancer is second only to breast cancer, early screening is beneficial to the prevention and treatment of cervical cancer as soon as possible. However, due to the high dimensional and complex characteristics of multi-omics data, the model generalization ability is still low in cancer prediction. To solve this problem, this paper proposes a new deep feature selection algorithm, which is based on Lasso penalty estimation, feature selection is performed and a neural network model is embedded to improve the generalization ability of cervical cancer omics data. Finally, through the comparison and survival analysis with different classifiers, it is proved that the algorithm has good performance in feature selection and improving data generalization ability.