Investigation of cardinality classification for bacterial colony counting using explainable artificial intelligence

ArXi:2604.20026v1 Announce Type: new Automatic bacterial colony counting is a highly sought-after technology in modern biological laboratories because it eliminates manual counting effort. Previous work has observed that MicrobiaNet, currently the best-performing cardinality classification model for colony counting, has difficulty distinguishing colonies of three or individuals. However, it is unclear if this is due to properties of the data together with inherent characteristics of the MicrobiaNet model.