Back

DiatomNet: An automatic Diatom genus identification system through microscopic images and Deep Learning

Tabik, S.; Villar, P.; Casado, J.; Fernandez, D.; Sanchez, P.

2025-02-11 bioinformatics
10.1101/2025.02.10.635050 bioRxiv
Show abstract

Diatoms are microscopic organisms belonging to the algae kingdom. They adapt to the ecosystem and modify their shape and texture depending on hundreds of ecosystem variables. Hence, these micro-organism are considered as the most accurate indicator to measure water quality. Commonly, the recognition of the class of diatoms in a microscopic image has always been done by expert biologists knowledgeable about the morphometric characteristics of these organisms. This work proposes a new automatic diatom genus recognition system from microscopic images using state-of-the-art deep CNNs. In particular, 1) we developed a public high quality database organized into 44 genus-level diatom classes, 2) designed a robust diatom classification model, and 3) provided a user-friendly interface to utilize our automatic diatom recognition tool.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

1
PLOS ONE
● 5266 papers in training set
Top 6%
26.7%
Similar papers in this journal
2
BMC Bioinformatics
○ 457 papers in training set
Top 1%
6.8%
Similar papers in this journal
3
Scientific Reports
○ 3612 papers in training set
Top 15%
5.5%
Similar papers in this journal
4
PLOS Computational Biology
● 1863 papers in training set
Top 7%
4.9%
Similar papers in this journal
5
Bioinformatics
◐ 1204 papers in training set
Top 5%
4.1%
6
Computers in Biology and Medicine
○ 128 papers in training set
Top 1.0%
4.1%
Similar papers in this journal
50% of probability mass above
7
Biology Methods and Protocols
◐ 61 papers in training set
Top 0.2%
3.5%
8
IEEE Access
● 35 papers in training set
Top 0.3%
3.3%
9
Neuroinformatics
○ 46 papers in training set
Top 0.2%
3.2%
10
Computational and Structural Biotechnology Journal
● 242 papers in training set
Top 2%
2.7%
11
Frontiers in Neuroinformatics
○ 41 papers in training set
Top 0.2%
2.5%
12
Computer Methods and Programs in Biomedicine
○ 28 papers in training set
Top 0.5%
1.7%
13
GigaScience
◐ 212 papers in training set
Top 3%
1.4%
14
Frontiers in Bioinformatics
○ 49 papers in training set
Top 0.7%
1.3%
15
Briefings in Bioinformatics
◐ 354 papers in training set
Top 6%
1.1%
16
Informatics in Medicine Unlocked
○ 22 papers in training set
Top 0.8%
1.1%
17
Animals
○ 23 papers in training set
Top 0.6%
1.1%
18
Sensors
○ 43 papers in training set
Top 0.9%
1.1%
19
Gigabyte
● 62 papers in training set
Top 0.9%
1.1%
20
Vision Research
○ 29 papers in training set
Top 0.3%
1.0%
21
Bioinformatics Advances
◐ 203 papers in training set
Top 4%
0.8%
22
Ecological Informatics
○ 33 papers in training set
Top 0.7%
0.8%
23
Bioengineering
○ 29 papers in training set
Top 1%
0.8%
24
Mathematics
○ 11 papers in training set
Top 0.4%
0.6%
25
Scientific Data
○ 209 papers in training set
Top 3%
0.6%
26
Plant Methods
○ 42 papers in training set
Top 0.9%
0.6%

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.