Scan-wise generalized PET denoising with contrastive adversarial learning

Clinical Snapshot

Key Findings

• P Value: Not reported

• Effect Size: Consistent improvements in PSNR and SSIM compared to baseline methods

• Primary Outcome: Superior denoising performance with improved peak signal-to-noise ratio and structural similarity index

• Nnt Or Sensitivity: Reduced bias and standard deviation in Alzheimer-related regions

• Confidence Interval: Not reported

Clinical Application

Requires specialized deep learning infrastructure and expertise; implementation complexity unclear Relevant for Australian dementia research centres and nuclear medicine departments using tau PET; potential application in PBS-funded PET studies for Alzheimer's disease Patients undergoing tau PET imaging, particularly in longitudinal Alzheimer's disease studies

Abstract

Objective.Deep learning has significantly advanced low-count positron emission tomography (PET) denoising. However, models trained on specific distributions often yield biased outputs when applied to scans with different activity distributions caused by anatomical and physiological variations (distribution shifts). Existing methods fail to generalize well across these scan-wise variations. Our goal is to formulate PET denoising as a scan-wise domain generalization (DG) problem to mitigate these variations and achieve robust, unbiased denoising for unseen scans.Approach.We propose a contrastive adversarial DG (CADG) framework to learn scan-invariant features efficiently. We leverage the property that multiple low-count noise realizations can be generated from a single raw list-mode PET scan to form a scan-wise domain distribution. Unlike conventional adversarial training with cross-entropy (CE) loss, we propose a contrastive adversarial framework that minimizes the mutual information between feature and scan-wise domains. Considering that each subject can have multiple scans in longitudinal studies, we further propose an ordered and memory-queued contrastive adversarial framework. This method efficiently includes realizations from the same scan as positive pairs, different scans of the same subject as pseudo-positive pairs, and different subjects as negative pairs in a memory batch. We explicitly exploit their ordered relationship as prior knowledge using a novel noisy-robust multi-positive ordinal contrastive loss.Main results.We systematically validated the effectiveness of our methods using 1920 noise realizations derived from 80 subjects with 192 longitudinal scans of MK-6240 tau PET data. The proposed CADG approach demonstrated superior denoising performance compared to CE-based adversarial methods and standard baselines. The ordered contrastive loss successfully improved the peak signal-to-noise ratio and structural similarity index consistently while reducing bias and standard deviation in Alzheimer-related regions and the whole brain.Significance.To our knowledge, this is the first attempt to alleviate the performance degradation in cross-scan denoising from the perspective of DG. Our study is also a pioneering work in utilizing longitudinal scans as pseudo-positives within an ordered contrastive learning scheme to exploit fine-grained relationships for robust clinical PET imaging applications.

References

1. 1.Liu, X., Marin, T., Vafay Eslahi, S., Tiss, A., Chemli, Y., Najmaoui, Y., Jang, S., El Fakhri, G., & Ouyang, J. (2026). Scan-wise generalized PET denoising with contrastive adversarial learning. Physics in Medicine and Biology. https://doi.org/10.1088/1361-6560/ae7231