Skip to main navigation Skip to search Skip to main content

The application and progress of AI-based image analysis in tumor organoid research

  • Yicheng Xu
  • , Shuai Fan
  • , Yuxin Chen
  • , Wanting Xu
  • , Haodong Lu
  • , Yutong Zhou
  • , Yuetong Liu
  • , Qing Du
  • , Wenyu Wang
  • , Tonghui Yu*
  • , Lei Dong*
  • *Corresponding author for this work
  • Beijing Institute of Technology
  • School of Mathematics, Harbin Institute of Technology

Research output: Contribution to journal › Review article › peer-review

Abstract

The development of preclinical models that recapitulate the physiological and pathological features of human tumors remains a central challenge in cancer research. Advances in cell biology have enabled the generation of three-dimensional tumor organoids, which closely mirror patient-specific therapeutic responses and facilitate the study of disease mechanisms. However, the trend of these models necessitates a shift from traditional, invasive analytical methods toward non-invasive, high-throughput imaging approaches. Here, we review the current state of tumor organoid culture and the emerging application of artificial intelligence (AI) in their evaluation. We discuss how AI-driven technologies are revolutionizing the analysis of fluorescence imaging, viability assessments, and dynamic cell tracking, thereby overcoming the limitations of manual interpretation. Finally, we provide a perspective on how integrating deep learning with organoid technology will enhance the precision and efficiency of drug discovery and personalized oncology.

Original languageEnglish
Article number116756
JournaliScience
Volume29
Issue number8
DOIs
Publication statusPublished - 21 Aug 2026
Externally publishedYes

Keywords

  • artificial intelligence
  • artificial intelligence applications
  • cancer
  • cancer
  • deep learning
  • image analysis
  • tumor organoid

Fingerprint

Dive into the research topics of 'The application and progress of AI-based image analysis in tumor organoid research'. Together they form a unique fingerprint.

Cite this