The Phantom of the Elytra -- Phylogenetic Trait Extraction from Images of Rove Beetles Using Deep Learning -- Is the Mask Enough?

2025-02-06

AI Summary

Traditional morphological phylogenetic analyses rely on manual feature extraction, resulting in low efficiency and poor scalability. Method: This study systematically compares three morphological representations—full segmentation, binary masks, and Fourier descriptors—for phylogenetic signal retention in deep learning–based feature extraction from rove beetle images. Using the Rove-Tree-11 dataset and a ResNet50 backbone, phylogenetic signal strength is quantified via normalized Align Score. Contribution/Results: We demonstrate for the first time that binary masks significantly outperform full segmentation (0.33 ± 0.02 vs. 0.39 ± 0.07) and Fourier descriptors (0.33 ± 0.02 vs. 0.45 ± 0.01), indicating that compact shape representations better align with deep networks and phylogenetic inference tasks. In contrast, elytral texture features exhibit weak phylogenetic signal. These findings establish a generalizable representation paradigm for automated morphological trait extraction in phylogenomics.

Technology Category

• Machine Learning: Feature Construction/Reformulation
• Computer Vision: Segmentation
• Natural Language Processing: Lexical Semantics and Morphology

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding

Abstract

Phylogenetic analysis traditionally relies on labor-intensive manual extraction of morphological traits, limiting its scalability for large datasets. Recent advances in deep learning offer the potential to automate this process, but the effectiveness of different morphological representations for phylogenetic trait extraction remains poorly understood. In this study, we compare the performance of deep learning models using three distinct morphological representations - full segmentations, binary masks, and Fourier descriptors of beetle outlines. We test this on the Rove-Tree-11 dataset, a curated collection of images from 215 rove beetle species. Our results demonstrate that the mask-based model outperformed the others, achieving a normalized Align Score of 0.33 plus/minus 0.02 on the test set, compared to 0.45 plus/minus 0.01 for the Fourier-based model and 0.39 plus/minus 0.07 for the segmentation-based model. The performance of the mask-based model likely reflects its ability to capture shape features while taking advantage of the depth and capacity of the ResNet50 architecture. These results also indicate that dorsal textural features, at least in this group of beetles, may be of lowered phylogenetic relevance, though further investigation is necessary to confirm this. In contrast, the Fourier-based model suffered from reduced capacity and occasional inaccuracies in outline approximations, particularly in fine structures like legs. These findings highlight the importance of selecting appropriate morphological representations for automated phylogenetic studies and the need for further research into explainability in automatic morphological trait extraction.

Problem

• Automating phylogenetic trait extraction using deep learning.
• Comparing morphological representations for phylogenetic analysis.
• Evaluating performance of mask-based models on beetle images.

Innovation

• Deep Learning for Phylogenetic Trait Extraction
• Mask-based Model with ResNet50 Architecture
• Comparison of Morphological Representations in Phylogeny