Multimodal Federated Learning With Missing Modalities through Feature Imputation Network

TL;DR

This paper introduces a lightweight feature imputation network for multimodal federated learning, effectively handling missing modalities in healthcare data without relying on synthetic data or expensive generative models.

Contribution

It proposes a novel low-dimensional feature translator to reconstruct missing modality features, improving federated learning performance in healthcare applications.

Findings

Consistent performance improvements across three datasets

Effective handling of missing modalities without synthetic data

Applicable to both homogeneous and heterogeneous settings

Abstract

Multimodal federated learning holds immense potential for collaboratively training models from multiple sources without sharing raw data, addressing both data scarcity and privacy concerns, two key challenges in healthcare. A major challenge in training multimodal federated models in healthcare is the presence of missing modalities due to multiple reasons, including variations in clinical practice, cost and accessibility constraints, retrospective data collection, privacy concerns, and occasional technical or human errors. Previous methods typically rely on publicly available real datasets or synthetic data to compensate for missing modalities. However, obtaining real datasets for every disease is impractical, and training generative models to synthesize missing modalities is computationally expensive and prone to errors due to the high dimensionality of medical data. In this paper, we propose a novel, lightweight, low-dimensional feature translator to reconstruct bottleneck features of the missing modalities. Our experiments on three different datasets (MIMIC-CXR, NIH Open-I, and CheXpert), in both homogeneous and heterogeneous settings consistently improve the performance of competitive baselines. The code and implementation details are available at: https://github.com/bhattarailab/FedFeatGen