# Machine Learning to Predict Multicellular Dynamics Driven by Concentrated Polymer Brush-Modified Cellulose Nanofibers

https://mdr.nims.go.jp/datasets/9569aba0-f338-4f3e-8dfd-62c1b6069a60

## File

- [abstract_YOSHIKAWA.pdf](https://mdr.nims.go.jp/filesets/5ee98f19-9b22-40ea-898a-240c8a2af646/download) ([Detail](https://mdr.nims.go.jp/filesets/5ee98f19-9b22-40ea-898a-240c8a2af646.md))

## Id

9569aba0-f338-4f3e-8dfd-62c1b6069a60

## Local identifier



## Visibility

open_to_public

## State

published

## Created at

2025-10-14T11:16:09.033820Z

## Updated at

2025-11-06T03:30:42.513426Z

## Published at

2025-11-06T03:24:51.615411Z

## Doi

https://doi.org/10.48505/nims.5860

## First published url



## Date published



## Recorded date published



## Resource type

conference_presentation

## Manuscript type

na

## Collection



## Title

- title: Machine Learning to Predict Multicellular Dynamics Driven by Concentrated
    Polymer Brush-Modified Cellulose Nanofibers
  title_type: original
  lang: en

## Description

- description: Manipulating the three-dimensional (3D) structures of cells is important
    for facilitating tissue repair and regeneration. We developed a self-assembly
    system for human mesenchymal stem cells (hMSCs) using cellulose nanofibers (CNFs)
    and concentrated polymer brushes (CPBs) to fabricate various cell 3D structures.
    To further generate tissues at an implantable level, it is necessary to carry
    out a large number of experiments using different cell culture conditions and
    material properties; however, this is too complex on a practical level. To address
    this issue, we present a graph neural network-based simulator (GNS) that can be
    trained using assembly process images to predict the assembly status of future
    time steps. A total of 24 (25 steps) time-series images were recorded (four repeats
    for each of the six conditions), and each image was transformed into a graph by
    regarding the cells as nodes and the connecting neighboring cells as edges. Using
    the obtained data, the performances of the GNS were examined under three scenarios
    (i.e., changing a pair of the training or test data) to verify the possibility
    of using the GNS as a predictor for further time steps. It was confirmed that
    the GNS could reasonably reproduce the assembly process, even under the toughest
    scenario wherein the experimental conditions differed between the training and
    test datasets. This result could reduce the number of experiments required to
    find the optimal conditions required for generating cells with desired 3D structures.
    Ultimately, such systems could accelerate progress in regenerative medicine.
  description_type: abstract
  lang: eng

## Creator

- name: Chiaki Yoshikawa
  role: author
  orcid: https://orcid.org/0000-0002-6589-387X
  organization: National Institute for Materials Science
  department: Research Center for Macromolecules and Biomaterials/Macromolecules Field/Polymer
    Surfaces and Devices Team
- name: Hiroshi Mamitsuka
  role: author

## Contact agent



## Publisher



## Managing organization



## Keyword

- subject: Machine learning
  schema: not_defined
- subject: Multicellular dynamics
  schema: not_defined
- subject: Concentrated polymer brush
  schema: not_defined
- subject: Cellulose nanofiber
  schema: not_defined

## Rights

- identifier: http://rightsstatements.org/vocab/InC/1.0/

## Other identifier(s)



## Data origin

- data_origin_type: other

## Embargo



## Journal



## Conference

name: The International Symposium on Exponential Biomedical DX 2024 (eMEDX-24)
start_date: 2024-12-19
end_date: 2024-12-20
identifier: https://www.jaist.ac.jp/ricenter/emedx/emedx-24/

## Related item



## Funding



## Instrument



## Instrument operator



## Instrument managing organization



## Measurement method



## Specimen



## Chemical composition



## Structure for specimen



## Structural feature for specimen



## Specific property for specimen



## Process for specimen treatment



## Computational method



## Energy level/transition state



## Software



## Custom property



## Fileset

- id: 5ee98f19-9b22-40ea-898a-240c8a2af646
  filename: abstract_YOSHIKAWA.pdf
  content_type: application/pdf
  size: 285841
  md5: 8c373f7e3d6215341225ef38f672e51d

## Thumbnail

fileset_id: 5ee98f19-9b22-40ea-898a-240c8a2af646
filename: abstract_YOSHIKAWA.pdf