Classification of Mycena and <i>Marasmius</i> Species Using Deep Learning Models: An Ecological and Taxonomic Approach
Fungi play a critical role in ecosystems, contributing to biodiversity and providing economic and biotechnological value. In this study, we developed a novel deep learning-based framework for the classification of seven macrofungi species from the genera <i>Mycena</i> and <i>Marasm...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-03-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/25/6/1642 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850088261519671296 |
|---|---|
| author | Fatih Ekinci Guney Ugurlu Giray Sercan Ozcan Koray Acici Tunc Asuroglu Eda Kumru Mehmet Serdar Guzel Ilgaz Akata |
| author_facet | Fatih Ekinci Guney Ugurlu Giray Sercan Ozcan Koray Acici Tunc Asuroglu Eda Kumru Mehmet Serdar Guzel Ilgaz Akata |
| author_sort | Fatih Ekinci |
| collection | DOAJ |
| description | Fungi play a critical role in ecosystems, contributing to biodiversity and providing economic and biotechnological value. In this study, we developed a novel deep learning-based framework for the classification of seven macrofungi species from the genera <i>Mycena</i> and <i>Marasmius</i>, leveraging their unique ecological and morphological characteristics. The proposed approach integrates a custom convolutional neural network (CNN) with a self-organizing map (SOM) adapted for supervised learning and a Kolmogorov–Arnold Network (KAN) layer to enhance classification performance. The experimental results demonstrate significant improvements in classification metrics when using the CNN-SOM and CNN-KAN architectures. Additionally, advanced pretrained models such as MaxViT-S and ResNetV2-50 achieved high accuracy rates, with MaxViT-S achieving 98.9% accuracy. Statistical analyses using the chi-square test confirmed the reliability of the results, emphasizing the importance of validating evaluation metrics statistically. This research represents the first application of SOM in fungal classification and highlights the potential of deep learning in advancing fungal taxonomy. Future work will focus on optimizing the KAN architecture and expanding the dataset to include more fungal classes, further enhancing classification accuracy and ecological understanding. |
| format | Article |
| id | doaj-art-74764ebd28e742f3aabd6ef1767e394f |
| institution | DOAJ |
| issn | 1424-8220 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-74764ebd28e742f3aabd6ef1767e394f2025-08-20T02:43:03ZengMDPI AGSensors1424-82202025-03-01256164210.3390/s25061642Classification of Mycena and <i>Marasmius</i> Species Using Deep Learning Models: An Ecological and Taxonomic ApproachFatih Ekinci0Guney Ugurlu1Giray Sercan Ozcan2Koray Acici3Tunc Asuroglu4Eda Kumru5Mehmet Serdar Guzel6Ilgaz Akata7Institute of Artificial Intelligence, Ankara University, Ankara 06100, TürkiyeDepartment of Computer Engineering, Faculty of Engineering, Başkent University, Ankara 06790, TürkiyeDepartment of Computer Engineering, Faculty of Engineering, Başkent University, Ankara 06790, TürkiyeInstitute of Artificial Intelligence, Ankara University, Ankara 06100, TürkiyeFaculty of Medicine and Health Technology, Tampere University, 33720 Tampere, FinlandGraduate School of Natural and Applied Sciences, Ankara University, Ankara 06830, TürkiyeInstitute of Artificial Intelligence, Ankara University, Ankara 06100, TürkiyeDepartment of Biology, Faculty of Science, Ankara University, Ankara 06100, TürkiyeFungi play a critical role in ecosystems, contributing to biodiversity and providing economic and biotechnological value. In this study, we developed a novel deep learning-based framework for the classification of seven macrofungi species from the genera <i>Mycena</i> and <i>Marasmius</i>, leveraging their unique ecological and morphological characteristics. The proposed approach integrates a custom convolutional neural network (CNN) with a self-organizing map (SOM) adapted for supervised learning and a Kolmogorov–Arnold Network (KAN) layer to enhance classification performance. The experimental results demonstrate significant improvements in classification metrics when using the CNN-SOM and CNN-KAN architectures. Additionally, advanced pretrained models such as MaxViT-S and ResNetV2-50 achieved high accuracy rates, with MaxViT-S achieving 98.9% accuracy. Statistical analyses using the chi-square test confirmed the reliability of the results, emphasizing the importance of validating evaluation metrics statistically. This research represents the first application of SOM in fungal classification and highlights the potential of deep learning in advancing fungal taxonomy. Future work will focus on optimizing the KAN architecture and expanding the dataset to include more fungal classes, further enhancing classification accuracy and ecological understanding.https://www.mdpi.com/1424-8220/25/6/1642macrofungi classificationmachine learningdeep learningself-organizing mapsMaxViT-Small |
| spellingShingle | Fatih Ekinci Guney Ugurlu Giray Sercan Ozcan Koray Acici Tunc Asuroglu Eda Kumru Mehmet Serdar Guzel Ilgaz Akata Classification of Mycena and <i>Marasmius</i> Species Using Deep Learning Models: An Ecological and Taxonomic Approach Sensors macrofungi classification machine learning deep learning self-organizing maps MaxViT-Small |
| title | Classification of Mycena and <i>Marasmius</i> Species Using Deep Learning Models: An Ecological and Taxonomic Approach |
| title_full | Classification of Mycena and <i>Marasmius</i> Species Using Deep Learning Models: An Ecological and Taxonomic Approach |
| title_fullStr | Classification of Mycena and <i>Marasmius</i> Species Using Deep Learning Models: An Ecological and Taxonomic Approach |
| title_full_unstemmed | Classification of Mycena and <i>Marasmius</i> Species Using Deep Learning Models: An Ecological and Taxonomic Approach |
| title_short | Classification of Mycena and <i>Marasmius</i> Species Using Deep Learning Models: An Ecological and Taxonomic Approach |
| title_sort | classification of mycena and i marasmius i species using deep learning models an ecological and taxonomic approach |
| topic | macrofungi classification machine learning deep learning self-organizing maps MaxViT-Small |
| url | https://www.mdpi.com/1424-8220/25/6/1642 |
| work_keys_str_mv | AT fatihekinci classificationofmycenaandimarasmiusispeciesusingdeeplearningmodelsanecologicalandtaxonomicapproach AT guneyugurlu classificationofmycenaandimarasmiusispeciesusingdeeplearningmodelsanecologicalandtaxonomicapproach AT giraysercanozcan classificationofmycenaandimarasmiusispeciesusingdeeplearningmodelsanecologicalandtaxonomicapproach AT korayacici classificationofmycenaandimarasmiusispeciesusingdeeplearningmodelsanecologicalandtaxonomicapproach AT tuncasuroglu classificationofmycenaandimarasmiusispeciesusingdeeplearningmodelsanecologicalandtaxonomicapproach AT edakumru classificationofmycenaandimarasmiusispeciesusingdeeplearningmodelsanecologicalandtaxonomicapproach AT mehmetserdarguzel classificationofmycenaandimarasmiusispeciesusingdeeplearningmodelsanecologicalandtaxonomicapproach AT ilgazakata classificationofmycenaandimarasmiusispeciesusingdeeplearningmodelsanecologicalandtaxonomicapproach |