Physics-Informed Deep Learning for Improved Input Function Estimation in Motion-Blurred Dynamic [${}^{18}$F]FDG PET Images

2025-10-24

AI Summary

This study addresses the challenge of accurately estimating the arterial input function (AIF) in dynamic [¹⁸F]FDG PET imaging under motion-induced blurring. We propose a physics-guided deep learning method built upon the physics-informed neural network (PINN) framework, which embeds a two-tissue compartment kinetic model into the network training process and incorporates physiological constraint losses to jointly optimize data-driven learning and domain-specific prior knowledge. Evaluated on 70 murine dynamic PET datasets with ground-truth measured AIFs, our model achieves high-accuracy AIF estimation even under severe motion blur—significantly outperforming purely data-driven approaches. The estimated AIFs exhibit improved physiological plausibility and superior generalizability across subjects and motion conditions. This work establishes a novel paradigm for noninvasive, rapid, and robust AIF acquisition in preclinical PET quantification.

Technology Category

• Computer Vision: Low Level & Physics-based Vision
• Machine Learning: Bio-inspired Learning
• Planning, Routing, and Scheduling: Activity and Plan Recognition

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Kinetic modeling enables extit{in vivo} quantification of tracer uptake and glucose metabolism in [${}^{18}$F]Fluorodeoxyglucose ([${}^{18}$F]FDG) dynamic positron emission tomography (dPET) imaging of mice. However, kinetic modeling requires the accurate determination of the arterial input function (AIF) during imaging, which is time-consuming and invasive. Recent studies have shown the efficacy of using deep learning to directly predict the input function, surpassing established methods such as the image-derived input function (IDIF). In this work, we trained a physics-informed deep learning-based input function prediction model (PIDLIF) to estimate the AIF directly from the PET images, incorporating a kinetic modeling loss during training. The proposed method uses a two-tissue compartment model over two regions, the myocardium and brain of the mice, and is trained on a dataset of 70 [${}^{18}$F]FDG dPET images of mice accompanied by the measured AIF during imaging. The proposed method had comparable performance to the network without a physics-informed loss, and when sudden movement causing blurring in the images was simulated, the PIDLIF model maintained high performance in severe cases of image degradation. The proposed physics-informed method exhibits an improved robustness that is promoted by physically constraining the problem, enforcing consistency for out-of-distribution samples. In conclusion, the PIDLIF model offers insight into the effects of leveraging physiological distribution mechanics in mice to guide a deep learning-based AIF prediction network in images with severe degradation as a result of blurring due to movement during imaging.

Problem

• Estimating arterial input function from motion-blurred PET images
• Improving robustness of AIF prediction using physics-informed deep learning
• Addressing image degradation in dynamic FDG PET imaging of mice

Innovation

• Physics-informed deep learning predicts input function
• Incorporates kinetic modeling loss for training
• Uses two-tissue compartment model for robustness