Utility of Deep Learning to Address Missing Modalities from Multi-Modal Medical Imaging Studies: A Systematic Review

Jinzhao Qian; Ankita Joshi; Hailong Li; Nehal A. Parikh; Jonanthan R. Dillman; Lili He

doi:10.47852/bonviewAIA52026392

Authors

Jinzhao Qian Imaging Research Center, Cincinnati Children’s Hospital Medical Center and Department of Computer Science, University of Cincinnati, USA https://orcid.org/0009-0007-3742-0032
Ankita Joshi Imaging Research Center, Cincinnati Children’s Hospital Medical Center, USA https://orcid.org/0009-0002-6712-2937
Hailong Li Imaging Research Center, Cincinnati Children’s Hospital Medical Center, Department of Radiology, Cincinnati Children’s Hospital Medical Center and Department of Radiology, University of Cincinnati College of Medicine, USA
Nehal A. Parikh Perinatal Institute, Cincinnati Children’s Hospital Medical Center and Department of Pediatrics, University of Cincinnati College of Medicine, USA
Jonanthan R. Dillman Imaging Research Center, Cincinnati Children’s Hospital Medical Center, Department of Radiology, Cincinnati Children’s Hospital Medical Center and Department of Radiology, University of Cincinnati College of Medicine, USA
Lili He Imaging Research Center, Cincinnati Children’s Hospital Medical Center, Department of Computer Science, University of Cincinnati, Department of Radiology, Cincinnati Children’s Hospital Medical Center, Department of Radiology, University of Cincinnati College of Medicine, Department of Biomedical Engineering, University of Cincinnati College of Medicine and Department of Biomedical Informatics, University of Cincinnati College of Medicine, USA

DOI:

https://doi.org/10.47852/bonviewAIA52026392

Keywords:

deep learning, missing modalities, image synthesis, knowledge transfer, latent space, medical image analysis

Abstract

Missing modalities pose a significant challenge on multi-modal studies by disrupting the comprehensive analysis of diverse data sources. Deep learning addresses this issue by employing algorithms that can effectively infer and integrate the absent information, thereby ensuring robustness and accuracy of the models while increasing the study’s statistical power. This study aims to provide a systematic literature review on deep learning solutions for missing imaging modalities in multi-modal medical data analysis. Articles on PubMed, IEEE explore digital library, and Scopus were searched in the range from January 2013 to May 2025. This systematic search and review identified 234 articles. Adhering to the specified search criteria, 61 published studies were eligible. Among these, 47% employed image synthesis methods, 20% applied knowledge transfer methods, and 33% used latent feature space-based methods. The paper explores the research gaps and challenges associated within each of these categories. Additionally, this review paper illuminates the popular public datasets for multi-modal studies with missing modalities. Furthermore, it presents evaluation metrics and their key attributes. The review concludes with its limitations and a detailed discussion of current challenges and future directions in this domain.

Received: 6 June 2025 | Revised: 17 September 2025 | Accepted: 24 September 2025

Conflicts of Interest

The authors declare that they have no conflicts of interest to this work.

Data Availability Statement

Data sharing is not applicable to this article as no new data were created or analyzed in this study.

Author Contribution Statement

Jinzhao Qian: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Data curation, Writing – original draft, Writing – review & editing, Visualization. Ankita Joshi: Conceptualization, Methodology, Validation, Investigation, Writing – original draft, Writing – review & editing, Visualization, Supervision. Hailong Li: Writing – review & editing, Project administration. Nehal A. Parikh: Writing – review & editing. Jonanthan R. Dillman: Writing – review & editing. Lili He: Conceptualization, Resources, Writing – review & editing, Supervision, Project administration, Funding acquisition.