Are model-based tools still relevant for bioimage analysis?
Poster presented as part of the Crick BioImage Analysis Symposium 2023.
Artificial Intelligence (AI) methods have taken the field of bioimage analysis by storm, eclipsing model-based methods with their promises of learning the model from the data. But model-based methods offer several advantages:
- avoid the training process, which requires big datasets
- provide better generalization for out-of-distribution samples
- are explainable, based on hypotheses to understand a priori their applicability
Permission has been given by authors to upload to Crick Figshare. Copyright remains with the original authors.