MARIA: a Multimodal Transformer Model for Incomplete Healthcare Data

Abstract

In healthcare, the integration of multimodal data is pivotal for developing comprehensive diagnostic and predictive models. However, managing missing data remains a significant challenge in real-world applications. We introduce MARIA (Multimodal Attention Resilient to Incomplete datA), a novel transformer-based deep learning model designed to address these challenges through an intermediate fusion strategy. Unlike conventional approaches that depend on imputation, MARIA utilizes a masked self-attention mechanism, which processes only the available data without generating synthetic values. This approach enables it to effectively handle incomplete datasets, enhancing robustness and minimizing biases introduced by imputation methods. We evaluated MARIA against 10 state-of-the-art machine learning and deep learning models across 8 diagnostic and prognostic tasks. The results demonstrate that MARIA outperforms existing methods in terms of performance and resilience to varying levels of data incompleteness, underscoring its potential for critical healthcare applications.

References in corpus (11)

• XGBoost: A Scalable Tree Boosting System
• TabTransformer: Tabular Data Modeling Using Contextual Embeddings
• Handling missing values in healthcare data: A systematic review of deep learning-based imputation techniques
• M${}^3$Care: Learning with Missing Modalities in Multimodal Healthcare Data
• A Systematic Review of Intermediate Fusion in Multimodal Deep Learning for Biomedical Applications
• Robust Multimodal Learning with Missing Modalities via Parameter-Efficient Adaptation
• Multimodal Explainability via Latent Shift applied to COVID-19 stratification
• Multi-objective optimization determines when, which and how to fuse deep networks: an application to predict COVID-19 outcomes
• Class Balancing Diversity Multimodal Ensemble for Alzheimer's Disease Diagnosis and Early Detection
• A graph neural network-based model with Out-of-Distribution Robustness for enhancing Antiretroviral Therapy Outcome Prediction for HIV-1
• Doctor-in-the-Loop: An Explainable, Multi-View Deep Learning Framework for Predicting Pathological Response in Non-Small Cell Lung Cancer

Cited by in corpus (2)

• Not Another Imputation Method: A Transformer-based Model for Missing Values in Tabular Datasets
• Beyond a Single Mode: GAN Ensembles for Diverse Medical Data Generation