Hybrid learning method based on feature clustering and scoring for enhanced COVID-19 breath analysis by an electronic nose

作者：

Highlights：

• A portable electronic nose (GeNose C19) has been developed for COVID-19 detection.

• Effective and optimum feature selection is obtained using hybrid learning method.

• Hierarchical agglomerative clustering is combined with permutation feature importance.

• Accuracy, sensitivity, and specificity of >86% have been achieved by GeNose C19.

• The proposed learning algorithm can eliminate the redundant sensors in the system.

摘要

•A portable electronic nose (GeNose C19) has been developed for COVID-19 detection.•Effective and optimum feature selection is obtained using hybrid learning method.•Hierarchical agglomerative clustering is combined with permutation feature importance.•Accuracy, sensitivity, and specificity of >86% have been achieved by GeNose C19.•The proposed learning algorithm can eliminate the redundant sensors in the system.

论文关键词：Breath analysis,Electronic nose,Machine learning,Feature permutation importance,Hierarchical agglomerative clustering,GeNose C19

论文评审过程：Received 12 December 2021, Revised 5 May 2022, Accepted 12 May 2022, Available online 17 May 2022, Version of Record 19 May 2022.

论文官网地址：https://doi.org/10.1016/j.artmed.2022.102323