Supplementary material from "How many specimens make a sufficient training set for automated 3D feature extraction?" ...
Deep learning has emerged as a robust tool for automating feature extraction from 3D images, offering an efficient alternative to labour-intensive and potentially biased manual image segmentation methods. However, there has been limited exploration into the optimal training set sizes, including asse...
Main Authors: | , , , , , |
---|---|
Format: | Article in Journal/Newspaper |
Language: | unknown |
Published: |
The Royal Society
2024
|
Subjects: | |
Online Access: | https://dx.doi.org/10.6084/m9.figshare.c.7288922.v1 https://rs.figshare.com/collections/Supplementary_material_from_How_many_specimens_make_a_sufficient_training_set_for_automated_3D_feature_extraction_/7288922/1 |
id |
ftdatacite:10.6084/m9.figshare.c.7288922.v1 |
---|---|
record_format |
openpolar |
spelling |
ftdatacite:10.6084/m9.figshare.c.7288922.v1 2024-09-15T18:31:00+00:00 Supplementary material from "How many specimens make a sufficient training set for automated 3D feature extraction?" ... Mulqueeney, James Michael Searle-Barnes, Alex Brombacher, Anieke Sweeney, Marisa Goswami, Anjali Ezard, Thomas 2024 https://dx.doi.org/10.6084/m9.figshare.c.7288922.v1 https://rs.figshare.com/collections/Supplementary_material_from_How_many_specimens_make_a_sufficient_training_set_for_automated_3D_feature_extraction_/7288922/1 unknown The Royal Society https://dx.doi.org/10.6084/m9.figshare.c.7288922 Creative Commons Attribution 4.0 International https://creativecommons.org/licenses/by/4.0/legalcode cc-by-4.0 Evolutionary biology not elsewhere classified article Collection 2024 ftdatacite https://doi.org/10.6084/m9.figshare.c.7288922.v110.6084/m9.figshare.c.7288922 2024-07-03T11:06:21Z Deep learning has emerged as a robust tool for automating feature extraction from 3D images, offering an efficient alternative to labour-intensive and potentially biased manual image segmentation methods. However, there has been limited exploration into the optimal training set sizes, including assessing whether artificial expansion by data augmentation can achieve consistent results in less time and how consistent these benefits are across different types of traits. In this study, we manually segmented 50 planktonic foraminifera specimens from the genus Menardella to determine the minimum number of training images required to produce accurate volumetric and shape data from internal and external structures. The results reveal unsurprisingly that deep learning models improve with a larger number of training images with eight specimens being required to achieve 95% accuracy. Furthermore, data augmentation can enhance network accuracy by up to 8.0%. Notably, predicting both volumetric and shape measurements for ... Article in Journal/Newspaper Planktonic foraminifera DataCite |
institution |
Open Polar |
collection |
DataCite |
op_collection_id |
ftdatacite |
language |
unknown |
topic |
Evolutionary biology not elsewhere classified |
spellingShingle |
Evolutionary biology not elsewhere classified Mulqueeney, James Michael Searle-Barnes, Alex Brombacher, Anieke Sweeney, Marisa Goswami, Anjali Ezard, Thomas Supplementary material from "How many specimens make a sufficient training set for automated 3D feature extraction?" ... |
topic_facet |
Evolutionary biology not elsewhere classified |
description |
Deep learning has emerged as a robust tool for automating feature extraction from 3D images, offering an efficient alternative to labour-intensive and potentially biased manual image segmentation methods. However, there has been limited exploration into the optimal training set sizes, including assessing whether artificial expansion by data augmentation can achieve consistent results in less time and how consistent these benefits are across different types of traits. In this study, we manually segmented 50 planktonic foraminifera specimens from the genus Menardella to determine the minimum number of training images required to produce accurate volumetric and shape data from internal and external structures. The results reveal unsurprisingly that deep learning models improve with a larger number of training images with eight specimens being required to achieve 95% accuracy. Furthermore, data augmentation can enhance network accuracy by up to 8.0%. Notably, predicting both volumetric and shape measurements for ... |
format |
Article in Journal/Newspaper |
author |
Mulqueeney, James Michael Searle-Barnes, Alex Brombacher, Anieke Sweeney, Marisa Goswami, Anjali Ezard, Thomas |
author_facet |
Mulqueeney, James Michael Searle-Barnes, Alex Brombacher, Anieke Sweeney, Marisa Goswami, Anjali Ezard, Thomas |
author_sort |
Mulqueeney, James Michael |
title |
Supplementary material from "How many specimens make a sufficient training set for automated 3D feature extraction?" ... |
title_short |
Supplementary material from "How many specimens make a sufficient training set for automated 3D feature extraction?" ... |
title_full |
Supplementary material from "How many specimens make a sufficient training set for automated 3D feature extraction?" ... |
title_fullStr |
Supplementary material from "How many specimens make a sufficient training set for automated 3D feature extraction?" ... |
title_full_unstemmed |
Supplementary material from "How many specimens make a sufficient training set for automated 3D feature extraction?" ... |
title_sort |
supplementary material from "how many specimens make a sufficient training set for automated 3d feature extraction?" ... |
publisher |
The Royal Society |
publishDate |
2024 |
url |
https://dx.doi.org/10.6084/m9.figshare.c.7288922.v1 https://rs.figshare.com/collections/Supplementary_material_from_How_many_specimens_make_a_sufficient_training_set_for_automated_3D_feature_extraction_/7288922/1 |
genre |
Planktonic foraminifera |
genre_facet |
Planktonic foraminifera |
op_relation |
https://dx.doi.org/10.6084/m9.figshare.c.7288922 |
op_rights |
Creative Commons Attribution 4.0 International https://creativecommons.org/licenses/by/4.0/legalcode cc-by-4.0 |
op_doi |
https://doi.org/10.6084/m9.figshare.c.7288922.v110.6084/m9.figshare.c.7288922 |
_version_ |
1810472592738353152 |