Machine Learning-Based Multi-Level Fusion Framework for a Hybrid Voltammetric and Impedimetric Metal Ions Electronic Tongue

Electronic tongues and artificial gustation for crucial analytes in the environment, such as metal ions, are becoming increasingly important. In this contribution, we propose a multi-level fusion framework hybrid impedimetric voltammetric electronic tongue to enhance accuracy of K+, Mg2+, Ca2+ detection an extensive concentration range (100.0 nM–1.0 mM). The proposed extracts electrochemical-based features separately fuses, first step, features, which characteristic points fixed frequency current potential data reduction by LDA classification kNN. Then, second decision is carried out combine results both measurement methods based on Dempster–Shafer (DS) evidence theory. reach 80.98% 81.48% measurements measurements, respectively. DS theory improves total recognition 91.60%, thus realizing significantly high comparison state-of-the-art. comparison, feature one step reaches only 89.13%. hierarchical considers time from multiple electrochemical sensor arrays. developed approach can be implemented several further applications pattern fusion, e.g., noses, environmental contaminants heavy pesticides, explosives, biomarkers, cancers diabetes.