Method for detecting microorganisms

A learning device comprising a processor configured to

• read, as learning data, fluorescence images obtained by imaging a membrane filter cultured with microorganisms collected from a liquid and fluorescently stained with a fluorescent staining agent for microorganisms and microorganism position information indicating the microorganism positions on the fluorescence images, the learning data obtained following a predetermined period of time after the microorganisms are cultured, and
• generate a trained model configured to identify whether a fluorescent signal of each pixel in inspection fluorescence images input as an inspection target is a fluorescent signal derived from the microorganisms or noise by performing machine learning based on the learning data.

An inspection device comprising a processor configured to determine the presence or absence of microorganisms by identifying whether a fluorescent signal of each pixel in inspection fluorescence images input as an inspection target is a fluorescent signal derived from the microorganisms or noise using a trained model generated by performing machine learning based on fluorescence images obtained by imaging a membrane filter cultured with microorganisms collected from a liquid and fluorescently stained with a fluorescent staining agent for microorganisms and microorganism position information indicating the microorganism positions on the fluorescence images, wherein the fluorescent signals are identified following a predetermined period of time after the microorganisms are cultured.