Deep Learning Applied to SEM Images for Supporting Marine Coralline Algae Classification
The classification of coralline algae commonly relies on the morphology of cells and reproductive structures, along with thallus organization, observed through Scanning Electron Microscopy (SEM). Nevertheless, species identification based on morphology often leads to uncertainty, due to their genera...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-12-01
|
| Series: | Diversity |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-2818/13/12/640 |
| _version_ | 1797505391257452544 |
|---|---|
| author | Giulia Piazza Cecile Valsecchi Gabriele Sottocornola |
| author_facet | Giulia Piazza Cecile Valsecchi Gabriele Sottocornola |
| author_sort | Giulia Piazza |
| collection | DOAJ |
| description | The classification of coralline algae commonly relies on the morphology of cells and reproductive structures, along with thallus organization, observed through Scanning Electron Microscopy (SEM). Nevertheless, species identification based on morphology often leads to uncertainty, due to their general plasticity. Evolutionary and environmental studies featured coralline algae for their ecological significance in both recent and past Oceans and need to rely on robust taxonomy. Research efforts towards new putative diagnostic tools have recently been focused on cell wall ultrastructure. In this work, we explored a new classification tool for coralline algae, using fine-tuning pretrained Convolutional Neural Networks (CNNs) on SEM images paired to morphological categories, including cell wall ultrastructure. We considered four common Mediterranean species, classified at genus and at the species level (<i>Lithothamnion corallioides</i>, <i>Mesophyllum philippii</i>, <i>Lithophyllum racemus</i>, <i>Lithophyllum pseudoracemus</i>). Our model produced promising results in terms of image classification accuracy given the constraint of a limited dataset and was tested for the identification of two ambiguous samples referred to as <i>L.</i> cf. <i>racemus</i>. Overall, explanatory image analyses suggest a high diagnostic value of calcification patterns, which significantly contributed to class predictions. Thus, CNNs proved to be a valid support to the morphological approach to taxonomy in coralline algae. |
| first_indexed | 2024-03-10T04:17:58Z |
| format | Article |
| id | doaj.art-5c03a32f740e4533b5884f96afd904f1 |
| institution | Directory Open Access Journal |
| issn | 1424-2818 |
| language | English |
| last_indexed | 2024-03-10T04:17:58Z |
| publishDate | 2021-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diversity |
| spelling | doaj.art-5c03a32f740e4533b5884f96afd904f12023-11-23T07:56:30ZengMDPI AGDiversity1424-28182021-12-01131264010.3390/d13120640Deep Learning Applied to SEM Images for Supporting Marine Coralline Algae ClassificationGiulia Piazza0Cecile Valsecchi1Gabriele Sottocornola2Department of Earth and Environmental Sciences, University of Milano-Bicocca, 20126 Milano, ItalyDepartment of Earth and Environmental Sciences, University of Milano-Bicocca, 20126 Milano, ItalyDepartment of Information Science and Technology, Free University of Bozen-Bolzano, 39100 Bolzano, ItalyThe classification of coralline algae commonly relies on the morphology of cells and reproductive structures, along with thallus organization, observed through Scanning Electron Microscopy (SEM). Nevertheless, species identification based on morphology often leads to uncertainty, due to their general plasticity. Evolutionary and environmental studies featured coralline algae for their ecological significance in both recent and past Oceans and need to rely on robust taxonomy. Research efforts towards new putative diagnostic tools have recently been focused on cell wall ultrastructure. In this work, we explored a new classification tool for coralline algae, using fine-tuning pretrained Convolutional Neural Networks (CNNs) on SEM images paired to morphological categories, including cell wall ultrastructure. We considered four common Mediterranean species, classified at genus and at the species level (<i>Lithothamnion corallioides</i>, <i>Mesophyllum philippii</i>, <i>Lithophyllum racemus</i>, <i>Lithophyllum pseudoracemus</i>). Our model produced promising results in terms of image classification accuracy given the constraint of a limited dataset and was tested for the identification of two ambiguous samples referred to as <i>L.</i> cf. <i>racemus</i>. Overall, explanatory image analyses suggest a high diagnostic value of calcification patterns, which significantly contributed to class predictions. Thus, CNNs proved to be a valid support to the morphological approach to taxonomy in coralline algae.https://www.mdpi.com/1424-2818/13/12/640machine learningCNNsSEM imagescoralline algaetaxonomyultrastructure |
| spellingShingle | Giulia Piazza Cecile Valsecchi Gabriele Sottocornola Deep Learning Applied to SEM Images for Supporting Marine Coralline Algae Classification Diversity machine learning CNNs SEM images coralline algae taxonomy ultrastructure |
| title | Deep Learning Applied to SEM Images for Supporting Marine Coralline Algae Classification |
| title_full | Deep Learning Applied to SEM Images for Supporting Marine Coralline Algae Classification |
| title_fullStr | Deep Learning Applied to SEM Images for Supporting Marine Coralline Algae Classification |
| title_full_unstemmed | Deep Learning Applied to SEM Images for Supporting Marine Coralline Algae Classification |
| title_short | Deep Learning Applied to SEM Images for Supporting Marine Coralline Algae Classification |
| title_sort | deep learning applied to sem images for supporting marine coralline algae classification |
| topic | machine learning CNNs SEM images coralline algae taxonomy ultrastructure |
| url | https://www.mdpi.com/1424-2818/13/12/640 |
| work_keys_str_mv | AT giuliapiazza deeplearningappliedtosemimagesforsupportingmarinecorallinealgaeclassification AT cecilevalsecchi deeplearningappliedtosemimagesforsupportingmarinecorallinealgaeclassification AT gabrielesottocornola deeplearningappliedtosemimagesforsupportingmarinecorallinealgaeclassification |