Skip to content
DraconiticResearch
Search Sources API
179,152 studies31 sourcesUpdated 56 min ago
PreprintCC BYChemRxiv

Automating Nanoparticle Analysis in Transmission Electron Microscopy with Synthetically Trained Segmentation Models

  • Natalia Sá1
  • Andrew Stewart1
  1. 1Department of Chemistry, University College London, 20 Gordon St, London, United Kingdom, WC1H 0AJ
Open at sourceDOICite

Abstract

Nanoparticle morphology governs the functional behaviour: plasmonic, catalytic, magnetic, electronic, of colloidal and supported nanomaterials, yet distributions of single-particle morphology are seldom reported. Published distributions typically aggregate ensemble averages or a few dozen manually segmented transmission electron microscopy (TEM) images, leaving population-level descriptors statistically underpowered. The rate-limiting step is instance segmentation: convolutional neural networks require extensive pixel-level annotations that are prohibitive to curate for every new particle system. Here we report TEMPOS, a deep-learning instance-segmentation package trained on procedurally generated synthetic TEM micrographs which simulate arbitrary particle distributions, substrate, noise, and imaging aberrations. TEMPOS outputs per-particle shape descriptors to a FAIR-compliant relational database and renders image archives searchable by morphological query. Distributed with a graphical interface for nondomain experts, the package reduces large-dataset analysis from weeks of manual segmentation to minutes of active work, enabling routine statistical morphometry across nanomaterials research.

Related

  1. DeLTA 3.0: A modern segmentation and tracking tool for microscopy images
    Virgile Andreani, Jean-Baptiste Lugagne, Owen M O’connor +1
    Journal of Open Source SoftwarePeer-reviewedCC BY 4.0Full textBiophysicsHAL
  2. A morphology-driven workflow to decipher 3D electron microscopy segmentation in diatoms
    Clarisse Uwizeye, Serena Flori, Joanell Angulo +4
    Cell Reports MethodsPeer-reviewedCC BY 4.0Full textBiophysicsHAL
  3. Crossed laser phase plates for transmission electron microscopy
    Petar N. Petrov, Jessie T. Zhang, Jeremy J. Axelrod +2
    Nature CommunicationsPeer-reviewedCC BYFull text6 citedStructural biologyOpenAlex
  4. Advancing Targeted Nanoparticle mRNA Delivery Using Machine Learning
    Marius Kausas, Valentina Schastlivaia
    ChemRxivPreprintCC BYFull textMolecular biology & gene regulationChemRxiv
  5. Monocyte-Mimetic Nanoparticle Delivery of Verteporfin Promotes Plaque-Stabilizing Remodeling in Atherosclerosis
    T.-Y. Wang, A. Zehnder, S.-P. Wu +3
    bioRxivPreprintAll rights reservedFull textMolecular biology & gene regulationbioRxiv

Cite

BibTeXRIS
BibTeX
@misc{sa2026automating,
  title = {{Automating Nanoparticle Analysis in Transmission Electron Microscopy with Synthetically Trained Segmentation Models}},
  author = {Natalia Sá and Andrew Stewart},
  howpublished = {ChemRxiv},
  year = {2026},
  month = oct,
  doi = {10.26434/chemrxiv.15010426/v1},
  url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15010426/v1},
  note = {CC BY},
}
RIS
TY  - UNPB
TI  - Automating Nanoparticle Analysis in Transmission Electron Microscopy with Synthetically Trained Segmentation Models
AU  - Natalia Sá
AU  - Andrew Stewart
PY  - 2026
DA  - 2026/10/13
PB  - ChemRxiv
DO  - 10.26434/chemrxiv.15010426/v1
UR  - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15010426/v1
AB  - Nanoparticle morphology governs the functional behaviour: plasmonic, catalytic, magnetic, electronic, of colloidal and supported nanomaterials, yet distributions of single-particle morphology are seldom reported. Published distributions typically aggregate ensemble averages or a few dozen manually segmented transmission electron microscopy (TEM) images, leaving population-level descriptors statistically underpowered. The rate-limiting step is instance segmentation: convolutional neural networks require extensive pixel-level annotations that are prohibitive to curate for every new particle system. Here we report TEMPOS, a deep-learning instance-segmentation package trained on procedurally generated synthetic TEM micrographs which simulate arbitrary particle distributions, substrate, noise, and imaging aberrations. TEMPOS outputs per-particle shape descriptors to a FAIR-compliant relational database and renders image archives searchable by morphological query. Distributed with a graphical interface for nondomain experts, the package reduces large-dataset analysis from weeks of manual segmentation to minutes of active work, enabling routine statistical morphometry across nanomaterials research.
ER  -