YOLOX-based detection of C60 molecules in time-resolved TEM
Abstract
Low contrast and heterogeneous backgrounds hinder automated molecular detection in time-resolved transmission electron microscopy (TEM). We prepared C60-loaded carbon nanohorns by solvent-free mechanochemistry and compared four preprocessing pipelines using YOLOX detectors trained on expert-annotated frames. Wiener filtering combined with contrastlimited adaptive histogram equalization (CLAHE) yielded the highest F1 score of 0.713 at an intersection-over-union (IoU) threshold of 0.30 on temporally held-out frames from the same sequence. Applied to later frames, the model detected partially overlapping C60 candidates and yielded declining detection counts. Some projected intermolecular separations decreased to 0.90 nm, with a cumulative electron dose on the order of 108 e– nm–2 over approximately 15 s. These observations were consistent with, but not diagnostic of, [2+2] cycloaddition. The model also detected some fused structures as C60, indicating that counts reflect C60-like image features rather than true monomer populations. The workflow provides frame-wise localization and counting of molecular candidates. Multiclass detection of monomers, and fused structures, combined with tracking that accounts for detection errors and independent validation, could support quantitative kinetic analysis observed by atomic-resolution TEM.
Related
- Bimodal spectroscopy integrating multi-wavelength time-resolved photoacoustic spectroscopy and near-infrared spectroscopy with deep learning for quantitative detection of serum biochemical indicators
- Time-Resolved Transcriptomic and Widely Targeted Metabolomic Profiling of Tomato Seedlings During Short-Term Chilling
- Inferring Cell Fate Trajectories in Time-Resolved Metabolic RNA Labeling data
- Sample-efficient time-resolved X-ray crystallography with serial small wedges
- Parameter-Based Prioritization of Time-Resolved LC–HRMS Ion Tracks for Untargeted Reaction Monitoring
BibTeX
@misc{saito2026yoloxbased,
title = {{YOLOX-based detection of C60 molecules in time-resolved TEM}},
author = {Yoshito Saito and Masaya Sakakibara and Kenta Itakura and Takayuki Nakamuro},
howpublished = {ChemRxiv},
year = {2026},
month = sep,
doi = {10.26434/chemrxiv.15009699/v1},
url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15009699/v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - YOLOX-based detection of C60 molecules in time-resolved TEM
AU - Yoshito Saito
AU - Masaya Sakakibara
AU - Kenta Itakura
AU - Takayuki Nakamuro
PY - 2026
DA - 2026/09/30
PB - ChemRxiv
DO - 10.26434/chemrxiv.15009699/v1
UR - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15009699/v1
AB - Low contrast and heterogeneous backgrounds hinder automated molecular detection in time-resolved transmission electron microscopy (TEM). We prepared C60-loaded carbon nanohorns by solvent-free mechanochemistry and compared four preprocessing pipelines using YOLOX detectors trained on expert-annotated frames. Wiener filtering combined with contrastlimited adaptive histogram equalization (CLAHE) yielded the highest F1 score of 0.713 at an intersection-over-union (IoU) threshold of 0.30 on temporally held-out frames from the same sequence. Applied to later frames, the model detected partially overlapping C60 candidates and yielded declining detection counts. Some projected intermolecular separations decreased to 0.90 nm, with a cumulative electron dose on the order of 108 e– nm–2 over approximately 15 s. These observations were consistent with, but not diagnostic of, [2+2] cycloaddition. The model also detected some fused structures as C60, indicating that counts reflect C60-like image features rather than true monomer populations. The workflow provides frame-wise localization and counting of molecular candidates. Multiclass detection of monomers, and fused structures, combined with tracking that accounts for detection errors and independent validation, could support quantitative kinetic analysis observed by atomic-resolution TEM.
ER -