Navigating the Multiverse: A Hitchhiker’s Guide to Selecting Harmonisation Methods for Multimodal Biomedical Data

Abstract

AbstractIntroductionThe application of machine learning (ML) techniques in classification and prediction tasks has greatly advanced our comprehension of biological systems. There is a notable shift in the trend towards integration methods that specifically target the simultaneous analysis of multiple modes or types of data, showcasing superior results compared to individual analyses. Despite the availability of diverse ML architectures for researchers interested in embracing a multimodal approach, the current literature lacks a comprehensive taxonomy that includes the pros and cons of these methods to guide the entire process. Closing this gap is imperative, necessitating the creation of a robust framework. This framework should not only categorise the diverse ML architectures suitable for multimodal analysis but also offer insights into their respective advantages and limitations. Additionally, such a framework can act as a guide for selecting an appropriate workflow for multimodal analysis. This comprehensive taxonomy would furnish a clear guidance and aid in informed decision-making within the progressively intricate realm of biomedical and clinical data analysis, and is imperative for advancing personalised medicine.ObjectiveThe aims of the work are to comprehensively study and describe the harmonisation processes that are performed and reported in the literature and present a working guide that would enable planning and selecting an appropriate integrative model.MethodsA systematic review of publications that report the multimodal harmonisation of biomedical and clinical data has been performed.ResultsWe present harmonisation as a dual process of representation and integration, each with multiple methods and categories. The taxonomy of the various representation and integration methods are classified into six broad categories and detailed with the advantages, disadvantages and examples. A guide flowchart that describes the step-by-step processes that are needed to adopt a multimodal approach is also presented along with examples and references.ConclusionsThis review provides a thorough taxonomy of methods for harmonising multimodal data and introduces a foundational 10-step guide for newcomers to implement a multimodal workflow.