Deep Learning Based Cell Segmentation of 3D Confocal Images of Wood for Enzymatic Hydrolysis Applications

Abstract

Plant cell walls are an abundant renewable resource for biobased products, but their recalcitrance limits enzymatic hydrolysis efficiency. Cell and tissue morphological parameters provide reliable biomarkers for predicting enzymatic hydrolysis efficiency. Accurate cell segmentation in 3D microscopy images is essential for quantifying these parameters. Although deep learning has achieved high performance in biological image segmentation, confocal microscopy's limited penetration depth leaves elongated wood cells partially imaged, creating open-ended structures that challenge existing segmentation methods. Based on an artificial longitudinal domain restriction, an adapted framework is presented here for the training of a 3D U-net model for the cell segmentation of wood images. Quantitative evaluation on a dedicated benchmark dataset identified a model configuration that reduced segmentation errors and produced more robust predictions during early enzymatic hydrolysis. These results support the use of the framework to quantify morphological parameters relevant to assessing and predicting biomass conversion efficiency.

2D U-NET Deep Learning Architecture for Atherosclerotic Lesion Segmentation and Quantification
W. Giles, Z. Venetianer, N. Patel +5
bioRxiv31 Oct 2026PreprintAll rights reservedFull textBioinformatics & computational biologybioRxiv
MorphoDeep enables fast 3D cell instance segmentation by learning cell interfaces and junctions
G. David, K. Biasuz, B. Gallean +3
bioRxiv16 Oct 2026PreprintCC BYFull textBioinformatics & computational biologybioRxiv
Deep Learning-Meets-Active Surfaces: A Unified Framework for Multi-Object Segmentation and Shape Analysis in 3D Fluorescence Microscopy
Quentin Rapilly, Anaïs Badoual, You-Hsuan Liu +1
HAL30 Sep 2026PreprintCC BY 4.0Full textBioinformatics & computational biologyHALbioRxiv
Event segmentation aligns emotional learning and episodic memory
Patrick Laing, Josh Cisler, Joseph E. Dunsmoor
PsyArXiv6 Oct 2026PreprintCC BY 4.0Full textNeuroscience PhysiologyOSF
Automatic Generation of Expert-Level Neuron Segmentation Masks from Fluorescence Microscopy Images for Non-Invasive Deep Learning Analysis of Phase-Contrast Images
Gerard Villarroya-Piqué, Víctor M. González, Esther Serrano-Pertierra +3
arXiv28 Sep 2026PreprintCC BY-NC-SA 4.0Full textNeurosciencearXiv

Cite

BibTeX RIS

BibTeX

@misc{faraj2026deep,
  title = {{Deep Learning Based Cell Segmentation of 3D Confocal Images of Wood for Enzymatic Hydrolysis Applications}},
  author = {A. Faraj and S. Hossein Khani and G. Paës and Y. Refahi},
  howpublished = {bioRxiv},
  year = {2026},
  month = oct,
  doi = {10.64898/2026.10.26.761974},
  url = {https://www.biorxiv.org/content/10.64898/2026.10.26.761974v1},
  note = {CC BY-NC-ND},
}

RIS

TY  - UNPB
TI  - Deep Learning Based Cell Segmentation of 3D Confocal Images of Wood for Enzymatic Hydrolysis Applications
AU  - A. Faraj
AU  - S. Hossein Khani
AU  - G. Paës
AU  - Y. Refahi
PY  - 2026
DA  - 2026/10/26
PB  - bioRxiv
DO  - 10.64898/2026.10.26.761974
UR  - https://www.biorxiv.org/content/10.64898/2026.10.26.761974v1
AB  - Plant cell walls are an abundant renewable resource for biobased products, but their recalcitrance limits enzymatic hydrolysis efficiency. Cell and tissue morphological parameters provide reliable biomarkers for predicting enzymatic hydrolysis efficiency. Accurate cell segmentation in 3D microscopy images is essential for quantifying these parameters. Although deep learning has achieved high performance in biological image segmentation, confocal microscopy's limited penetration depth leaves elongated wood cells partially imaged, creating open-ended structures that challenge existing segmentation methods. Based on an artificial longitudinal domain restriction, an adapted framework is presented here for the training of a 3D U-net model for the cell segmentation of wood images. Quantitative evaluation on a dedicated benchmark dataset identified a model configuration that reduced segmentation errors and produced more robust predictions during early enzymatic hydrolysis. These results support the use of the framework to quantify morphological parameters relevant to assessing and predicting biomass conversion efficiency.
ER  -

Abstract

Related

Cite