Structured sparse logistic regression with application to lung cancer prediction using breath volatile biomarkers

Abstract

This article is motivated by a study of lung cancer prediction using breath volatile organic compound (VOC) biomarkers, where the challenge is that the predictors include not only high‐dimensional time‐dependent or functional VOC features but also the time‐independent clinical variables. We consider a high‐dimensional logistic regression and propose two different penalties: group spline‐penalty or group smooth‐penalty to handle the group structures of the time‐dependent variables in the model. The new methods have the advantage for the situation where the model coefficients are sparse but change smoothly within the group, compared with other existing methods such as the group lasso and the group bridge approaches. Our methods are easy to implement since they can be turned into a group minimax concave penalty problem after certain transformations. We show that our fitting algorithm possesses the descent property and leads to attractive convergence properties. The simulation studies and the lung cancer application are performed to demonstrate the accuracy and stability of the proposed approaches.

Target Projection Feature Matching Based Deep ANN with LSTM for Lung Cancer Prediction

“…But the designed network failed to collect more training data with higher robustness. A high-dimensional logistic regression was developed in [5] for lung cancer prediction with higher accuracy. But, the feature selection process was not carried out with a lesser prediction time.…”

RAMRSGL: A Robust Adaptive Multinomial Regression Model for Multicancer Classification

“…Due to the characteristics of automatic variable selection, sparse regression methods [9][10][11][12][13] have attracted a surge of attention in cancer diagnosis and gene selection. To tackle the problem that the l 1 regularization has a biased gene selection and does not have the oracle property, Wu et al [13] in 2018 have investigated l 1 /l 2 regularized logistic regression for gene selection in high-dimensional cancer classification.…”

“…The experimental results demonstrate that the top selected genes are biologically related to the cancer type, which is useful for cancer classification using DNA gene expression data in real clinical practice. To handle the group structures of the time-dependent clinical variables in the model, Zhang et al [10] in 2020 have developed a high-dimensional logistic regression and introduced the group spline-penalty or group smooth-penalty. This method is easy to implement since it can be turned into a group minimax concave penalty problem after certain transformations.…”