Neuro-Symbolic Artificial Intelligence for Patient Monitoring

TL;DR

The paper addresses the need for interpretable and data-efficient patient monitoring using Neuro-Symbolic AI. It proposes a hierarchical process architecture that fuses low-level sensor observations with symbolic domain knowledge to infer patient states and high-level activities. It surveys NeSy paradigms such as Logic Tensor Networks, DeepProbLog, ProbLog, and CSSMs to address sample efficiency, uncertainty, and explainability within HAR-based medical monitoring. The authors provide a roadmap for deploying NeSy in hospital environments and broader care contexts, highlighting practical deployment challenges and potential benefits.

Abstract

In this paper we argue that Neuro-Symbolic AI (NeSy-AI) should be applied for patient monitoring. In this context, we introduce patient monitoring as a special case of Human Activity Recognition and derive concrete requirements for this application area. We then present a process architecture and discuss why NeSy-AI should be applied for patient monitoring. To further support our argumentation, we show how NeSy-AI can help to overcome certain technical challenges that arise from this application area.

Figures (1)

• Figure 1: The different levels of the patient monitoring process.