Deep learning based low-activity PET reconstruction of [11C]PiB and [18F]FE-PE2I in neurodegenerative disorders
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 o...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-10-01
|
| Series: | NeuroImage |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1053811922005298 |
| _version_ | 1818198155411324928 |
|---|---|
| author | Raphaël Sura Daveau Ian Law Otto Mølby Henriksen Steen Gregers Hasselbalch Ulrik Bjørn Andersen Lasse Anderberg Liselotte Højgaard Flemming Littrup Andersen Claes Nøhr Ladefoged |
| author_facet | Raphaël Sura Daveau Ian Law Otto Mølby Henriksen Steen Gregers Hasselbalch Ulrik Bjørn Andersen Lasse Anderberg Liselotte Højgaard Flemming Littrup Andersen Claes Nøhr Ladefoged |
| author_sort | Raphaël Sura Daveau |
| collection | DOAJ |
| description | 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. |
| first_indexed | 2024-12-12T02:01:22Z |
| format | Article |
| id | doaj.art-e0aff247b6e74ab8900410d957141011 |
| institution | Directory Open Access Journal |
| issn | 1095-9572 |
| language | English |
| last_indexed | 2024-12-12T02:01:22Z |
| publishDate | 2022-10-01 |
| publisher | Elsevier |
| record_format | Article |
| series | NeuroImage |
| spelling | doaj.art-e0aff247b6e74ab8900410d9571410112022-12-22T00:42:10ZengElsevierNeuroImage1095-95722022-10-01259119412Deep learning based low-activity PET reconstruction of [11C]PiB and [18F]FE-PE2I in neurodegenerative disordersRaphaël Sura Daveau0Ian Law1Otto Mølby Henriksen2Steen Gregers Hasselbalch3Ulrik Bjørn Andersen4Lasse Anderberg5Liselotte Højgaard6Flemming Littrup Andersen7Claes Nøhr Ladefoged8Department of Clinical Physiology and Nuclear Medicine, Rigshospitalet, University of Copenhagen, DenmarkDepartment of Clinical Physiology and Nuclear Medicine, Rigshospitalet, University of Copenhagen, DenmarkDepartment of Clinical Physiology and Nuclear Medicine, Rigshospitalet, University of Copenhagen, DenmarkDanish Dementia Research Centre, Rigshospitalet, University of Copenhagen, DenmarkDepartment of Clinical Physiology and Nuclear Medicine, Rigshospitalet, University of Copenhagen, DenmarkDepartment of Clinical Physiology and Nuclear Medicine, Rigshospitalet, University of Copenhagen, DenmarkDepartment of Clinical Physiology and Nuclear Medicine, Rigshospitalet, University of Copenhagen, DenmarkDepartment of Clinical Physiology and Nuclear Medicine, Rigshospitalet, University of Copenhagen, DenmarkDepartment of Clinical Physiology and Nuclear Medicine, Rigshospitalet, University of Copenhagen, Denmark; Corresponding author.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.http://www.sciencedirect.com/science/article/pii/S1053811922005298Parkinson's disease[18F]FE-PE2IAlzheimer's disease[11C]PiBDeep learningPET denoising |
| spellingShingle | Raphaël Sura Daveau Ian Law Otto Mølby Henriksen Steen Gregers Hasselbalch Ulrik Bjørn Andersen Lasse Anderberg Liselotte Højgaard Flemming Littrup Andersen Claes Nøhr Ladefoged Deep learning based low-activity PET reconstruction of [11C]PiB and [18F]FE-PE2I in neurodegenerative disorders NeuroImage Parkinson's disease [18F]FE-PE2I Alzheimer's disease [11C]PiB Deep learning PET denoising |
| title | Deep learning based low-activity PET reconstruction of [11C]PiB and [18F]FE-PE2I in neurodegenerative disorders |
| title_full | Deep learning based low-activity PET reconstruction of [11C]PiB and [18F]FE-PE2I in neurodegenerative disorders |
| title_fullStr | Deep learning based low-activity PET reconstruction of [11C]PiB and [18F]FE-PE2I in neurodegenerative disorders |
| title_full_unstemmed | Deep learning based low-activity PET reconstruction of [11C]PiB and [18F]FE-PE2I in neurodegenerative disorders |
| title_short | Deep learning based low-activity PET reconstruction of [11C]PiB and [18F]FE-PE2I in neurodegenerative disorders |
| title_sort | deep learning based low activity pet reconstruction of 11c pib and 18f fe pe2i in neurodegenerative disorders |
| topic | Parkinson's disease [18F]FE-PE2I Alzheimer's disease [11C]PiB Deep learning PET denoising |
| url | http://www.sciencedirect.com/science/article/pii/S1053811922005298 |
| work_keys_str_mv | AT raphaelsuradaveau deeplearningbasedlowactivitypetreconstructionof11cpiband18ffepe2iinneurodegenerativedisorders AT ianlaw deeplearningbasedlowactivitypetreconstructionof11cpiband18ffepe2iinneurodegenerativedisorders AT ottomølbyhenriksen deeplearningbasedlowactivitypetreconstructionof11cpiband18ffepe2iinneurodegenerativedisorders AT steengregershasselbalch deeplearningbasedlowactivitypetreconstructionof11cpiband18ffepe2iinneurodegenerativedisorders AT ulrikbjørnandersen deeplearningbasedlowactivitypetreconstructionof11cpiband18ffepe2iinneurodegenerativedisorders AT lasseanderberg deeplearningbasedlowactivitypetreconstructionof11cpiband18ffepe2iinneurodegenerativedisorders AT liselottehøjgaard deeplearningbasedlowactivitypetreconstructionof11cpiband18ffepe2iinneurodegenerativedisorders AT flemminglittrupandersen deeplearningbasedlowactivitypetreconstructionof11cpiband18ffepe2iinneurodegenerativedisorders AT claesnøhrladefoged deeplearningbasedlowactivitypetreconstructionof11cpiband18ffepe2iinneurodegenerativedisorders |