Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

Scan-wise generalized PET denoising with contrastive adversarial learning.

Xiaofeng Liu¹, Thibault Marin¹, Samira Vafay Eslahi²

¹Biomedical Imaging Institute and Dept. of Radiology and Biomedical Imaging, Yale University, New Haven, CT 06520, United States of America.

Physics in Medicine and Biology

|May 22, 2026

Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Open-Geometry PET: Quantifying the Trade-off Between Time-of-Flight Resolution and Angular Coverage.

Physics in medicine and biology·2026

Same author

An evolutionary neural architecture search for magnetic resonance image reconstructions.

International journal of computer assisted radiology and surgery·2026

Same author

An Open Multi-Center Whole-Body FDG PET/CT Foundation Model for Tumor Segmentation.

ArXiv·2026

Same author

Direct Cardiac T1 Mapping with Subspace Modeling and Free-breathing Data Acquisition.

IEEE transactions on bio-medical engineering·2026

Same author

Individualized Treatment Effect Inference of Head and Neck Cancer with Multimodal Data.

APSIPA Transactions on Signal and Information Processing·2026

Same author

Revolutionizing MDD diagnosis: the integrated role of circRNAs and biosensor technology.

Molecular psychiatry·2026

Same journal

Deep learning-based dose prediction to enhance planning efficiency in cervical brachytherapy with hybrid applicators.

Physics in medicine and biology·2026

Same journal

Corrigendum: Referenceless MR thermometry-a comparison of five methods (2017<i>Phys. Med. Biol</i>.<b>62</b>1-16).

Physics in medicine and biology·2026

Same journal

Corrigendum: Measured and Monte Carlo simulated electron backscatter to the monitor chamber for the varian TrueBeam linac (2016<i>Phys. Med. Biol</i>.<b>61</b>8779).

Physics in medicine and biology·2026

Same journal

Corrigendum: 3D range-modulator for scanned particle therapy: development, Monte Carlo simulations and experimental evaluation (2017<i>Phys. Med. Biol</i>.<b>62</b>7075).

Physics in medicine and biology·2026

Same journal

Recent progress in applications of computing to radiotherapy (ICCR 2016).

Physics in medicine and biology·2026

Same journal

Novel TMS coils designed using an inverse boundary element method.

Physics in medicine and biology·2026

See all related articles

This summary is machine-generated.

Deep learning for PET denoising struggles with data variations. A new contrastive adversarial domain generalization method improves robustness and reduces bias in low-count PET scans across different subjects and scans.

Area of Science:

Medical Imaging
Artificial Intelligence
Nuclear Medicine

Background:

Deep learning (DL) excels at low-count Positron Emission Tomography (PET) denoising.
Distribution shifts from anatomical/physiological variations cause biased outputs and poor generalization in existing DL models.
Current methods fail to address scan-wise variations effectively.

Purpose of the Study:

Formulate PET denoising as a scan-wise Domain Generalization (DG) problem to mitigate variations.
Achieve robust and unbiased denoising for unseen PET scans.
Develop a DL framework that generalizes well across different scan distributions.

Main Methods:

Propose a contrastive adversarial domain generalization framework to learn scan-invariant features.

Keywords:

contrastive learning domain generalization low-count denoising positron emission tomography

Related Experiment Videos

Utilize multiple noise realizations from single raw list-mode PET scans to create scan-wise domains.

Introduce an ordered and memory-queued contrastive adversarial framework for longitudinal data, employing a novel noisy-robust multipositive ordinal contrastive loss.

Main Results:

The contrastive adversarial DG approach outperformed cross-entropy-based adversarial methods and standard baselines.
Ordered contrastive loss enhanced peak signal-to-noise ratio and structural similarity index.
Bias and standard deviation were reduced in Alzheimer's-related regions and the whole brain.

Conclusions:

This study is the first to address cross-scan denoising degradation using domain generalization.
Utilizing longitudinal scans as pseudo-positives within an ordered contrastive learning scheme is a pioneering approach.
The proposed methods offer a pathway to robust clinical PET imaging applications.