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

2026-04-21

AI Summary

This study addresses the challenge in automated bacterial colony counting where cardinality classification models struggle to distinguish three or more overlapping colonies. By applying explainable artificial intelligence (XAI) techniques to visualize and attribute predictions of MicrobiaNet, the work reveals— for the first time—that the primary bottleneck lies in high inter-class visual similarity among colony configurations, rather than insufficient model representational capacity. This insight revises the prevailing understanding of current model limitations and clarifies future directions: either explicitly modeling visual similarities between colonies or shifting toward a density estimation paradigm. The findings thus provide a new theoretical foundation and concrete technical pathways for advancing research in this domain.

Technology Category

• Computer Vision: Interpretability, Explainability, and Transparency
• Machine Learning: Calibration & Uncertainty Quantification
• Humans and AI: Explainable AI (XAI) for Human Understanding

Application Category

• Web Mining and Content Analysis: Models for Web evolution
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

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 more individuals. However, it is unclear if this is due to properties of the data together with inherent characteristics of the MicrobiaNet model. By analysing MicrobiaNet with explainable artificial intelligence (XAI), we demonstrate that XAI can provide insights into how data properties constrain cardinality classification performance in colony counting. Our results show that high visual similarity across classes is the key issue hindering further performance improvement, revising prior assertions about MicrobiaNet. These findings suggest future work should focus on models that explicitly incorporate visual similarity or explore density estimation approaches, with broader implications for neural network classifiers trained on imbalanced datasets.

Problem

• bacterial colony counting
• cardinality classification
• visual similarity
• explainable artificial intelligence
• imbalanced datasets

Innovation

• explainable artificial intelligence
• cardinality classification
• bacterial colony counting
• visual similarity
• imbalanced datasets