Deep Learning Image Enhancement Algorithms in PET/CT Imaging: a Phantom and Sarcoma Patient Radiomic Evaluation

Overview

Journal Eur J Nucl Med Mol Imaging

Specialties Biophysics
Nuclear Medicine
Radiology

Date 2025 Feb 27

PMID 40014074

Authors

L M Bonney

G M Kalisvaart

F H P van Velden

K M Bradley

A B Hassan

W Grootjans

D R McGowan

Affiliations

Soon will be listed here.

Abstract

Purpose: PET/CT imaging data contains a wealth of quantitative information that can provide valuable contributions to characterising tumours. A growing body of work focuses on the use of deep-learning (DL) techniques for denoising PET data. These models are clinically evaluated prior to use, however, quantitative image assessment provides potential for further evaluation. This work uses radiomic features to compare two manufacturer deep-learning (DL) image enhancement algorithms, one of which has been commercialised, against 'gold-standard' image reconstruction techniques in phantom data and a sarcoma patient data set (N=20).

Methods: All studies in the retrospective sarcoma clinical [ F]FDG dataset were acquired on either a GE Discovery 690 or 710 PET/CT scanner with volumes segmented by an experienced nuclear medicine radiologist. The modular heterogeneous imaging phantom used in this work was filled with [ F]FDG, and five repeat acquisitions of the phantom were acquired on a GE Discovery 710 PET/CT scanner. The DL-enhanced images were compared to 'gold-standard' images the algorithms were trained to emulate and input images. The difference between image sets was tested for significance in 93 international biomarker standardisation initiative (IBSI) standardised radiomic features.

Results: Comparing DL-enhanced images to the 'gold-standard', 4.0% and 9.7% radiomic features measured significantly different (p < 0.0005) in the phantom and patient data respectively (averaged over the two DL algorithms). Larger differences were observed comparing DL-enhanced images to algorithm input images with 29.8% and 43.0% of radiomic features measuring significantly different in the phantom and patient data respectively (averaged over the two DL algorithms).

Conclusion: DL-enhanced images were found to be similar to images generated using the 'gold-standard' target image reconstruction method with more than 80% of radiomic features not significantly different in all comparisons across unseen phantom and sarcoma patient data. This result offers insight into the performance of the DL algorithms, and demonstrate potential applications for DL algorithms in harmonisation for radiomics and for radiomic features in quantitative evaluation of DL algorithms.

References

Lambin P, Leijenaar R, Deist T, Peerlings J, de Jong E, van Timmeren J . Radiomics: the bridge between medical imaging and personalized medicine. Nat Rev Clin Oncol. 2017; 14(12):749-762. DOI: 10.1038/nrclinonc.2017.141. View

Dagogo-Jack I, Shaw A . Tumour heterogeneity and resistance to cancer therapies. Nat Rev Clin Oncol. 2017; 15(2):81-94. DOI: 10.1038/nrclinonc.2017.166. View

Hashimoto F, Onishi Y, Ote K, Tashima H, Reader A, Yamaya T . Deep learning-based PET image denoising and reconstruction: a review. Radiol Phys Technol. 2024; 17(1):24-46. PMC: 10902118. DOI: 10.1007/s12194-024-00780-3. View

Weyts K, Lasnon C, Ciappuccini R, Lequesne J, Corroyer-Dulmont A, Quak E . Artificial intelligence-based PET denoising could allow a two-fold reduction in [F]FDG PET acquisition time in digital PET/CT. Eur J Nucl Med Mol Imaging. 2022; 49(11):3750-3760. PMC: 9399218. DOI: 10.1007/s00259-022-05800-1. View

Mehranian A, Wollenweber S, Walker M, Bradley K, Fielding P, Su K . Image enhancement of whole-body oncology [F]-FDG PET scans using deep neural networks to reduce noise. Eur J Nucl Med Mol Imaging. 2021; 49(2):539-549. PMC: 8803788. DOI: 10.1007/s00259-021-05478-x. View

Mehranian A, Wollenweber S, Walker M, Bradley K, Fielding P, Huellner M . Deep learning-based time-of-flight (ToF) image enhancement of non-ToF PET scans. Eur J Nucl Med Mol Imaging. 2022; 49(11):3740-3749. PMC: 9399038. DOI: 10.1007/s00259-022-05824-7. View

Buvat I, Orlhac F . The Dark Side of Radiomics: On the Paramount Importance of Publishing Negative Results. J Nucl Med. 2019; 60(11):1543-1544. DOI: 10.2967/jnumed.119.235325. View

Huang E, OConnor J, McShane L, Giger M, Lambin P, Kinahan P . Criteria for the translation of radiomics into clinically useful tests. Nat Rev Clin Oncol. 2022; 20(2):69-82. PMC: 9707172. DOI: 10.1038/s41571-022-00707-0. View

Spadarella G, Stanzione A, Akinci DAntonoli T, Andreychenko A, Fanni S, Ugga L . Systematic review of the radiomics quality score applications: an EuSoMII Radiomics Auditing Group Initiative. Eur Radiol. 2022; 33(3):1884-1894. PMC: 9935718. DOI: 10.1007/s00330-022-09187-3. View

10.

Nardone V, Reginelli A, Grassi R, Boldrini L, Vacca G, DIppolito E . Delta radiomics: a systematic review. Radiol Med. 2021; 126(12):1571-1583. DOI: 10.1007/s11547-021-01436-7. View

11.

Hatt M, Krizsan A, Rahmim A, Bradshaw T, Costa P, Forgacs A . Joint EANM/SNMMI guideline on radiomics in nuclear medicine : Jointly supported by the EANM Physics Committee and the SNMMI Physics, Instrumentation and Data Sciences Council. Eur J Nucl Med Mol Imaging. 2022; 50(2):352-375. PMC: 9816255. DOI: 10.1007/s00259-022-06001-6. View

12.

Gallivanone F, DAmbrosio D, Carne I, DArcangelo M, Montagna P, Giroletti E . A tri-modal tissue-equivalent anthropomorphic phantom for PET, CT and multi-parametric MRI radiomics. Phys Med. 2022; 98:28-39. DOI: 10.1016/j.ejmp.2022.04.007. View

13.

Pfaehler E, Beukinga R, de Jong J, Slart R, Slump C, Dierckx R . Repeatability of F-FDG PET radiomic features: A phantom study to explore sensitivity to image reconstruction settings, noise, and delineation method. Med Phys. 2018; 46(2):665-678. PMC: 7380016. DOI: 10.1002/mp.13322. View

14.

Kalisvaart G, van Velden F, Hernandez-Giron I, Meijer K, Ghesquiere-Dierickx L, Brink W . Design and evaluation of a modular multimodality imaging phantom to simulate heterogeneous uptake and enhancement patterns for radiomic quantification in hybrid imaging: A feasibility study. Med Phys. 2022; 49(5):3093-3106. PMC: 9314050. DOI: 10.1002/mp.15537. View

15.

Pfaehler E, van Sluis J, Merema B, van Ooijen P, Berendsen R, van Velden F . Experimental Multicenter and Multivendor Evaluation of the Performance of PET Radiomic Features Using 3-Dimensionally Printed Phantom Inserts. J Nucl Med. 2019; 61(3):469-476. PMC: 7067530. DOI: 10.2967/jnumed.119.229724. View

16.

Shiri I, Rahmim A, Ghaffarian P, Geramifar P, Abdollahi H, Bitarafan-Rajabi A . The impact of image reconstruction settings on 18F-FDG PET radiomic features: multi-scanner phantom and patient studies. Eur Radiol. 2017; 27(11):4498-4509. DOI: 10.1007/s00330-017-4859-z. View

17.

van Griethuysen J, Fedorov A, Parmar C, Hosny A, Aucoin N, Narayan V . Computational Radiomics System to Decode the Radiographic Phenotype. Cancer Res. 2017; 77(21):e104-e107. PMC: 5672828. DOI: 10.1158/0008-5472.CAN-17-0339. View

18.

Marturano F, Guglielmo P, Bettinelli A, Zattoni F, Novara G, Zorz A . Role of radiomic analysis of [F]fluoromethylcholine PET/CT in predicting biochemical recurrence in a cohort of intermediate and high risk prostate cancer patients at initial staging. Eur Radiol. 2023; 33(10):7199-7208. PMC: 10511374. DOI: 10.1007/s00330-023-09642-9. View