Hierarchical Specialised Ensembles for Classification of Zebrafish Phenotypes Using the Selected Image Recognition Methods
Computer Vision and Pattern RecognitionMachine Learning
Summary
The authors tested three step-by-step methods to classify zebrafish embryo images by their condition. First, all images go through a model to identify one of four categories: Normal, Chorion, Dead, or Other. Then, images labeled as Other are further analyzed using different approaches depending on the method. They tried different neural network models and found that ConvNeXt worked best overall, while one setup with specialized classifiers gave the best F1-score balance. Their work shows that these specialized stepwise systems are good at recognizing zebrafish embryo conditions.
Authors
Piotr S. Maciąg, Monika Maciąg, Magdalena Majdan
Abstract
We propose and evaluate three hierarchical ensemble setups for zebrafish phenotype classification from embryo images. In all setups, stage 1 uses a single four-class classifier to assign images to one of the exclusive phenotypes: Normal, Chorion, Dead, or Other. Images classified as Other are then processed in stage 2, where the ensemble design differs across setups: a single multi-label classifier, two specialized multi-label classifiers, or an ensemble of binary classifiers. We compare these setups using three backbone architectures: ResNet18, ViT, and ConvNeXt. Overall, ConvNeXt achieves the best performance across setups, while the specialized hierarchical ensemble in setup 2 provides the best balance in terms of F1-score. The results show that the proposed specialised hierarchical ensembles are effective for zebrafish phenotype recognition, and suggest that ConvNeXt is particularly useful backbone model.