AI-integrated dye@MOF@MIPs odor sensors for monitoring of food freshness.

This study introduces a novel approach for the rapid monitoring and visualization of food freshness using AI-integrated colorimetric odor sensors. Targeting total volatile basic nitrogen (TVB-N), a key indicator of meat spoilage, the method utilizes a color-sensitive sensor array developed for the detection of trimethylamine (TMA), the primary component of TVB-N. To enhance selectivity for TMA, a molecularly imprinted polymer (MIPs) shell was in situ grown on a metal-organic framework (MOF), forming a core-shell MOF@MIPs composite. Nine dyes were then incorporated into the composite to create a sensing array. Density functional theory (DFT) calculations provided valuable insights into the enhanced adsorption mechanism. The sensor array displayed distinct color changes and achieved a detection limit of 124 ppb for TMA. For quantitative analysis, an AI-assisted FocusNet model was employed to process the sensor data, accurately predicting TVB-N levels in pork samples with a correlation coefficient of 0.9367 between predicted and reference values. This study demonstrates an effective strategy that combines selective nano-sensors with AI, offering a promising solution for on-site monitoring and quality assessment of food freshness.