Deep learning based low-activity PET reconstruction of [11C]PiB and [18F]FE-PE2I in neurodegenerative disorders

Raphaël Sura Daveau; Ian Law; Otto Mølby Henriksen; Steen Gregers Hasselbalch; Ulrik Bjørn Andersen; Lasse Anderberg; Liselotte Højgaard; Flemming Littrup Andersen; Claes Nøhr Ladefoged

doi:10.1016/j.neuroimage.2022.119412

Deep learning based low-activity PET reconstruction of [¹¹C]PiB and [¹⁸F]FE-PE2I in neurodegenerative disorders

Neuroimage. 2022 Oct 1:259:119412. doi: 10.1016/j.neuroimage.2022.119412. Epub 2022 Jun 24.

Authors

Raphaël Sura Daveau¹, Ian Law¹, Otto Mølby Henriksen¹, Steen Gregers Hasselbalch², Ulrik Bjørn Andersen¹, Lasse Anderberg¹, Liselotte Højgaard¹, Flemming Littrup Andersen¹, Claes Nøhr Ladefoged³

Affiliations

¹ Department of Clinical Physiology and Nuclear Medicine, Rigshospitalet, University of Copenhagen, Denmark.
² Danish Dementia Research Centre, Rigshospitalet, University of Copenhagen, Denmark.
³ Department of Clinical Physiology and Nuclear Medicine, Rigshospitalet, University of Copenhagen, Denmark. Electronic address: claes.noehr.ladefoged@regionh.dk.

PMID: 35753592
DOI: 10.1016/j.neuroimage.2022.119412

Abstract

Purpose: Positron Emission Tomography (PET) can support a diagnosis of neurodegenerative disorder by identifying disease-specific pathologies. Our aim was to investigate the feasibility of using activity reduction in clinical [¹⁸F]FE-PE2I and [¹¹C]PiB PET/CT scans, simulating low injected activity or scanning time reduction, in combination with AI-assisted denoising.

Methods: A total of 162 patients with clinically uncertain Alzheimer's disease underwent amyloid [¹¹C]PiB PET/CT and 509 patients referred for clinically uncertain Parkinson's disease underwent dopamine transporter (DAT) [¹⁸F]FE-PE2I PET/CT. Simulated low-activity data were obtained by random sampling of 5% of the events from the list-mode file and a 5% time window extraction in the middle of the scan. A three-dimensional convolutional neural network (CNN) was trained to denoise the resulting PET images for each disease cohort.

Results: Noise reduction of low-activity PET images was successful for both cohorts using 5% of the original activity with improvement in visual quality and all similarity metrics with respect to the ground-truth images. Clinically relevant metrics extracted from the low-activity images deviated < 2% compared to ground-truth values, which were not significantly changed when extracting the metrics from the denoised images.

Conclusion: The presented models were based on the same network architecture and proved to be a robust tool for denoising brain PET images with two widely different tracer distributions (delocalized, ([¹¹C]PiB, and highly localized, [18F]FE-PE2I). This broad and robust application makes the presented network a good choice for improving the quality of brain images to the level of the standard-activity images without degrading clinical metric extraction. This will allow for reduced dose or scan time in PET/CT to be implemented clinically.

Keywords: Alzheimer's disease; Deep learning; PET denoising; Parkinson's disease; [(11)C]PiB; [(18)F]FE-PE2I.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Deep Learning*
Humans
Nortropanes*
Parkinson Disease*
Positron Emission Tomography Computed Tomography
Positron-Emission Tomography / methods

Substances

Nortropanes