Functional Covalent Organic Framework Nanosheets for Ratiometric Monitoring of Pseudomonas aeruginosa Biomarker with Deep Learning-Assisted Assay

The structure of covalent organic framework nanosheets (COF NSs) effectively mitigates fluorescence quenching caused by π–π stacking. Designing functionalized COF NSs as luminescent materials represents an innovative approach to expanding research on COF-based fluorescence sensing. Given that COF NSs expose a greater number of binding sites, they can be functionalized with appropriate luminescent elements to form a COF NS composite. The interaction between analytes and COF NS composites results in significant changes in luminescence, making these composites promising candidates for chemosensing applications. In this study, we introduce functionalized dual luminescent centers (silicon quantum dots (SiQDs) and Eu3+ complexes) into COF skeletons via a modification-facilitated exfoliation strategy. This process transforms the bulk COF (MPCOF) into a COF NS composite (1) with an enhanced luminescent performance. As a fluorescent sensor, 1 detects the biomarker 2-aminoacetophenone (2-AA) from Pseudomonas aeruginosa using ratiometric signals, facilitating visual sensing analysis. Leveraging its high sensitivity, an algorithmic method based on deep learning has been established to predict the fluorescence response of sensor 1 to specific concentrations of 2-AA. This facile determination of 2-AA enables effective monitoring of human or animal health and evaluation of water environments.