Mechanism-aware and multimodal AI: beyond model-agnostic interpretation.

Journal: Trends in cell biology
Published Date:

Abstract

Artificial intelligence (AI) is widely used for exploiting multimodal biomedical data, with increasingly accurate predictions and model-agnostic interpretations, which are however also agnostic to biological mechanisms. Combining metabolic modelling, 'omics, and imaging data via multimodal AI can generate predictions that can be interpreted mechanistically and transparently, therefore with significantly higher therapeutic potential.

Authors

  • Annalisa Occhipinti
    Computational Systems Biology and Data Analytics Research Group, Middlebrough, UK.
  • Suraj Verma
    School of Computing, Engineering and Digital Technologies, Teesside University, Middlesborough, UK.
  • Le Minh Thao Doan
    School of Computing, Engineering and Digital Technologies, Teesside University, Middlesborough, UK.
  • Claudio Angione
    Department of Computer Science and Information Systems, Teesside University, Middlesbrough, United Kingdom.