Med-Real2Sim: Non-Invasive Medical Digital Twins using Physics-Informed Self-Supervised Learning

TL;DR

This paper introduces Med-Real2Sim, a physics-informed self-supervised learning approach to create non-invasive digital twins of physiological processes, enabling in-silico disease modeling and clinical trials without invasive data collection.

Contribution

It presents a novel physics-informed SSL method for identifying digital twin parameters solely from noninvasive patient data, bridging inverse problems and self-supervised learning.

Findings

Successfully modeled cardiac hemodynamics from echocardiogram videos.

Enabled unsupervised disease detection using digital twins.

Demonstrated potential for in-silico clinical trials.

Abstract

A digital twin is a virtual replica of a real-world physical phenomena that uses mathematical modeling to characterize and simulate its defining features. By constructing digital twins for disease processes, we can perform in-silico simulations that mimic patients' health conditions and counterfactual outcomes under hypothetical interventions in a virtual setting. This eliminates the need for invasive procedures or uncertain treatment decisions. In this paper, we propose a method to identify digital twin model parameters using only noninvasive patient health data. We approach the digital twin modeling as a composite inverse problem, and observe that its structure resembles pretraining and finetuning in self-supervised learning (SSL). Leveraging this, we introduce a physics-informed SSL algorithm that initially pretrains a neural network on the pretext task of learning a differentiable simulator of a physiological process. Subsequently, the model is trained to reconstruct physiological measurements from noninvasive modalities while being constrained by the physical equations learned in pretraining. We apply our method to identify digital twins of cardiac hemodynamics using noninvasive echocardiogram videos, and demonstrate its utility in unsupervised disease detection and in-silico clinical trials.