Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2506.20649 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916810958307328 |
|---|---|
| author | Dapueto, Jacopo Pastore, Vito Paolo Noceti, Nicoletta Odone, Francesca |
| author_facet | Dapueto, Jacopo Pastore, Vito Paolo Noceti, Nicoletta Odone, Francesca |
| contents | Microscopy image analysis is fundamental for different applications, from diagnosis to synthetic engineering and environmental monitoring. Modern acquisition systems have granted the possibility to acquire an escalating amount of images, requiring a consequent development of a large collection of deep learning-based automatic image analysis methods. Although deep neural networks have demonstrated great performance in this field, interpretability, an essential requirement for microscopy image analysis, remains an open challenge.
This work proposes a Disentangled Representation Learning (DRL) methodology to enhance model interpretability for microscopy image classification. Exploiting benchmark datasets from three different microscopic image domains (plankton, yeast vacuoles, and human cells), we show how a DRL framework, based on transferring a representation learnt from synthetic data, can provide a good trade-off between accuracy and interpretability in this domain. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_20649 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Disentangled representations of microscopy images Dapueto, Jacopo Pastore, Vito Paolo Noceti, Nicoletta Odone, Francesca Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Microscopy image analysis is fundamental for different applications, from diagnosis to synthetic engineering and environmental monitoring. Modern acquisition systems have granted the possibility to acquire an escalating amount of images, requiring a consequent development of a large collection of deep learning-based automatic image analysis methods. Although deep neural networks have demonstrated great performance in this field, interpretability, an essential requirement for microscopy image analysis, remains an open challenge. This work proposes a Disentangled Representation Learning (DRL) methodology to enhance model interpretability for microscopy image classification. Exploiting benchmark datasets from three different microscopic image domains (plankton, yeast vacuoles, and human cells), we show how a DRL framework, based on transferring a representation learnt from synthetic data, can provide a good trade-off between accuracy and interpretability in this domain. |
| title | Disentangled representations of microscopy images |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2506.20649 |