Dataset Information

The Impact of Artificial Intelligence CNN Based Denoising on FDG PET Radiomics.

ABSTRACT:

Background

With a constantly increasing number of diagnostic images performed each year, Artificial Intelligence (AI) denoising methods offer an opportunity to respond to the growing demand. However, it may affect information in the image in an unknown manner. This study quantifies the effect of AI-based denoising on FDG PET textural information in comparison to a convolution with a standard gaussian postfilter (EARL1).

Methods

The study was carried out on 113 patients who underwent a digital FDG PET/CT (VEREOS, Philips Healthcare). 101 FDG avid lesions were segmented semi-automatically by a nuclear medicine physician. VOIs in the liver and lung as reference organs were contoured. PET textural features were extracted with pyradiomics. Texture features from AI denoised and EARL1 versus original PET images were compared with a Concordance Correlation Coefficient (CCC). Features with CCC values ≥ 0.85 threshold were considered concordant. Scatter plots of variable pairs with R2 coefficients of the more relevant features were computed. A Wilcoxon signed rank test to compare the absolute values between AI denoised and original images was performed.

Results

The ratio of concordant features was 90/104 (86.5%) in AI denoised versus 46/104 (44.2%) with EARL1 denoising. In the reference organs, the concordant ratio for AI and EARL1 denoised images was low, respectively 12/104 (11.5%) and 7/104 (6.7%) in the liver, 26/104 (25%) and 24/104 (23.1%) in the lung. SUVpeak was stable after the application of both algorithms in comparison to SUVmax. Scatter plots of variable pairs showed that AI filtering affected more lower versus high intensity regions unlike EARL1 gaussian post filters, affecting both in a similar way. In lesions, the majority of texture features 79/100 (79%) were significantly (p<0.05) different between AI denoised and original PET images.

Conclusions

Applying an AI-based denoising on FDG PET images maintains most of the lesion's texture information in contrast to EARL1-compatible Gaussian filter. Predictive features of a trained model could be thus the same, however with an adapted threshold. Artificial intelligence based denoising in PET is a very promising approach as it adapts the denoising in function of the tissue type, preserving information where it should.

SUBMITTER: Jaudet C

PROVIDER: S-EPMC8421788 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:IntroductionThe yield per elution of a 68Ge/68Ga generator decreases during its lifespan. This affects the number of patients injected per elution or the injected dose per patient, thereby negatively affecting the cost of examinations and the quality of PET images due to increased image noise. We aimed to investigate whether AI-based PET denoising can offset this decrease in image quality parameters.MethodsAll patients addressed to our PET unit for a 68Ga-DOTATOC PET/CT from April 2020 to February 2021 were enrolled. Forty-four patients underwent their PET scans according to Protocol_FixedDose (150 MBq) and 32 according to Protocol_WeightDose (1.5 MBq/kg). Protocol_WeightDose examinations were processed using the Subtle PET software (Protocol_WeightDoseAI). Liver and vascular SUV mean were recorded as well as SUVmax, SUVmean and metabolic tumour volume (MTV) of the most intense tumoural lesion and its background SUVmean. Liver and vascular coefficients of variation (CV), tumour-to-background and tumour-to-liver ratios were calculated.ResultsThe mean injected dose of 2.1 (0.4) MBq/kg per patient was significantly higher in the Protocol_FixedDose group as compared to 1.5 (0.1) MBq/kg for the Protocol_WeightDose group. Protocol_WeightDose led to noisier images than Protocol_FixedDose with higher CVs for liver (15.57% ± 4.32 vs. 13.04% ± 3.51, p = 0.018) and blood-pool (28.67% ± 8.65 vs. 22.25% ± 10.37, p = 0.0003). Protocol_WeightDoseAI led to less noisy images than Protocol_WeightDose with lower liver CVs (11.42% ± 3.05 vs. 15.57% ± 4.32, p < 0.0001) and vascular CVs (16.62% ± 6.40 vs. 28.67% ± 8.65, p < 0.0001). Tumour-to-background and tumour-to-liver ratios were lower for protocol_WeightDoseAI: 6.78 ± 3.49 vs. 7.57 ± 4.73 (p = 0.01) and 5.96 ± 5.43 vs. 6.77 ± 6.19 (p < 0.0001), respectively. MTVs were higher after denoising whereas tumour SUVmax were lower: the mean% differences in MTV and SUVmax were + 11.14% (95% CI = 4.84-17.43) and -3.92% (95% CI = -6.25 to -1.59).ConclusionThe degradation of PET image quality due to a reduction in injected dose at the end of the 68Ge/68Ga generator lifespan can be effectively counterbalanced by using AI-based PET denoising.

Project description:Background: Diagnostic reports contribute not only to the particular patient, but also to constructing massive training dataset in the era of artificial intelligence (AI). The maximum standardized uptake value (SUVmax) is often described in daily diagnostic reports of [18F] fluorodeoxyglucose (FDG) positron emission tomography (PET) - computed tomography (CT). If SUVmax can be used as an identifier of lesion, that would greatly help AI interpret diagnostic reports. We aimed to clarify whether the lesion can be localized using SUVmax strings. Methods: The institutional review board approved this retrospective study. We investigated a total of 112 lesions from 30 FDG PET-CT images acquired with 3 different scanners. SUVmax was calculated from DICOM files based on the latest Quantitative Imaging Biomarkers Alliance (QIBA) publication. The voxels showing the given SUVmax were exhaustively searched in the whole-body images and counted. SUVmax was provided with 5 different degrees of precision: integer (e.g., 3), 1st decimal places (DP) (3.1), 2nd DP (3.14), 3rd DP (3.142), and 4th DP (3.1416). For instance, when SUVmax = 3.14 was given, the voxels with 3.135 ≤ SUVmax < 3.145 were extracted. We also evaluated whether local maximum restriction could improve the identifying performance, where only the voxels showing the highest intensity within some neighborhood were considered. We defined that "identical detection" was achieved when only single voxel satisfied the criterion. Results: A total of 112 lesions from 30 FDG PET-CT images were investigated. SUVmax ranged from 1.3 to 49.1 (median = 5.6). Generally, when larger and more precise SUVmax values were given, fewer voxels satisfied the criterion. The local maximum restriction was very effective. When SUVmax was determined to 4 decimal places (e.g., 3.1416) and the local maximum restriction was applied, identical detection was achieved in 33.3% (lesions with SUVmax < 2), 79.5% (2 ≤ SUVmax < 5), and 97.8% (5 ≤ SUVmax) of lesions. Conclusion: In this preliminary study, SUVmax of FDG PET-CT could be used as an identifier to localize the lesion if precise SUVmax is provided and local maximum restriction was applied, although the lesions showing SUVmax < 2 were difficult to identify. The proposed method may have potential to make use of diagnostic reports retrospectively for constructing training datasets for AI.