2026-04-29T02:37:29Zhttps://repisalud.isciii.es/rest/oai/request
oai:repisalud.isciii.es:20.500.12105/233632024-11-29T02:05:32Zcom_20.500.12105_15322com_20.500.12105_2051col_20.500.12105_16967
Repisalud author Vidaurre-Gallart, Isabel author Fernaud-Espinosa, Isabel author Cosmin-Toader, Nicusor author Talavera-Martinez, Lidia author Martin-Abadal, Miguel author Benavides-Piccione, Ruth author Gonzalez-Cid, Yolanda author Pastor, Luis author DeFelipe, Javier author Garcia-Lorenzo, Marcos 2024-10-04T13:16:19Z 2024-10-04T13:16:19Z 2022-03-17 Vidaurre-Gallart I, Fernaud-Espinosa I, Cosmin-Toader N, Talavera-Martinez L, Martin-Abadal M, Benavides-Piccione R, et al. A Deep Learning-Based Workflow for Dendritic Spine Segmentation. Front Neuroanat. 2022 Mar 17;16:817903. 1662-5129 https://hdl.handle.net/20.500.13003/19402 https://hdl.handle.net/20.500.12105/23363 35370569 10.3389/fnana.2022.817903 Frontiers in Neuroanatomy 2-s2.0-85127860017 786633300001 L2015504591 The morphological analysis of dendritic spines is an important challenge for the neuroscientific community. Most state-of-the-art techniques rely on user-supervised algorithms to segment the spine surface, especially those designed for light microscopy images. Therefore, processing large dendritic branches is costly and time-consuming. Although deep learning (DL) models have become one of the most commonly used tools in image segmentation, they have not yet been successfully applied to this problem. In this research article, we study the feasibility of using DL models to automatize spine segmentation from confocal microscopy images. Supervised learning is the most frequently used method for training DL models. This approach requires large data sets of high-quality segmented images (ground truth). As mentioned above, the segmentation of microscopy images is time-consuming and, therefore, in most cases, neuroanatomists only reconstruct relevant branches of the stack. Additionally, some parts of the dendritic shaft and spines are not segmented due to dyeing problems. In the context of this research, we tested the most successful architectures in the DL biomedical segmentation field. To build the ground truth, we used a large and high-quality data set, according to standards in the field. Nevertheless, this data set is not sufficient to train convolutional neural networks for accurate reconstructions. Therefore, we implemented an automatic preprocessing step and several training strategies to deal with the problems mentioned above. As shown by our results, our system produces a high-quality segmentation in most cases. Finally, we integrated several postprocessing user-supervised algorithms in a graphical user interface application to correct any possible artifacts. eng Automatic 3D image segmentation Artificial neural network Confocal microscopy Reconstruction algorithms Pyramidal cells A Deep Learning-Based Workflow for Dendritic Spine Segmentation research article