Pressure monitoring inside hermetically sealed containers generally involves the use of devices in direct contact with the product inside them. However, this requires a hole to be made in the wall of the container, with the risk of causing leaks or degradation of the product itself.

In this context, the aim of a recent study carried out by a group of US researchers (Prisbrey et al., 2024) was to develop a non-invasive technology for measuring pressure inside closed containers based on the use of acoustic resonance spectroscopy (ARS) and machine learning (ML) algorithms. In particular, during the experiment, the KNN (k-nearest neighbour) regression model was trained using ARS spectra as input.

The results show that the proposed system allows non-invasive pressure measurement inside closed containers without the use of sensors in permanent contact with the product.