Multimodal Fusion for Differential Disease Diagnosis A Comprehensive Review

DOI

10.2991/978-94-6239-723-1_13How to use a DOI?

Keywords

Disease Prediction; Multimodal Data Fusion; Symptom Extraction; Laboratory Feature Engineering; Reference Range Normalization; Approximate Nearest Neighbors (ANN); Classification Calibration; Explainable Artificial Intelligence (XAI); Clinical Decision Support Systems (CDSS)

Abstract

The cornerstone of effective patient care is an accurate and, even more importantly, timely diagnosis. There is a growing body of evidence which shows more and more the truth of the statement that the combination of different data sources—clinical narratives, laboratory assays, and imaging—produces better results than single-source diagnostic approaches. This paper presents a review of 29 peer-reviewed studies. They include classics in machine learning (SVM, Random Forest), deep learning models (CNNs, Transformer variations), and multimodal fusion techniques (early, late, attention-based). Furthermore, an integrated perspective is provided by examining contemporary clinical decision-support systems and near-real-time prediction models like MUFASA. Reported performance across studies varies significantly, with accuracy ranging from 61.9% to 99% and AUC values from 0.65 to 0.99 depending on dataset characteristics and modeling approaches. In spite of this, a few obstacles still remain: the issues of interpretability, the need for powerful computing, and the challenge of deployment. The review highlights key challenges and design considerations for explainable multimodal diagnostic systems, including the role of embedding-based representations, retrieval mechanisms, and feature engineering strategies reported across prior studies. It also summarizes recent advances, evaluates performance metrics and limitations, and identifies key challenges including interpretability, data heterogeneity, and clinical deployment constraints.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Joshil Das
AU  - Mayank Sharma
AU  - Sandeep Saxena
AU  - Hardik Bindal
AU  - Sahil Kumar Aggarwal
AU  - Dhruv Sagar Saxena
PY  - 2026
DA  - 2026/07/14
TI  - Multimodal Fusion for Differential Disease Diagnosis A Comprehensive Review
BT  - Proceedings of the International Conference on Responsible, Risk-aware, and Regulated AI (RRRAI 2026)
PB  - Atlantis Press
SP  - 139
EP  - 148
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-723-1_13
DO  - 10.2991/978-94-6239-723-1_13
ID  - Das2026
ER  -