PET/CT scanners deliver two kinds of radiation in a single examination: the radiotracer injected into the patient, and the X-ray dose from the CT component. When the CT is used only for attenuation correction and anatomical localisation rather than full diagnostic imaging, it can still account for more than half of the patient’s total radiation exposure. That overhead poses a major limitation for dose-sensitive populations, including pregnant women and paediatric patients. A new study published in EJNMMI Physics reports that a carefully chosen ultra-low-dose CT protocol can strip away most of that exposure while leaving the PET images quantitatively intact.
The research, led by Georgios I. Angelis and colleagues, systematically evaluated how progressive reductions in CT dose affect CT image quality, the accuracy of the attenuation map derived from the CT, and the resulting PET quantification. The team began with NEMA phantom experiments, a standardised way of characterising imaging performance under controlled conditions, before moving to prospective validation in clinical studies using two widely used tracers, [18F]FDG and [68Ga]DOTATATE.
The central result is striking: tin-filtered ultra-low-dose CT achieved a reduction of more than 90% in CT effective dose while still preserving accurate attenuation correction and PET quantification. The selected protocol combined tin filtration with 140 kVp and a tube current of 6 mAs, a setting the authors identified as offering the optimal balance between dose reduction and image quality, avoiding artefacts and maintaining robust performance across the quantitative PET metrics they measured.
Validation in patients compared PET images reconstructed using the ultra-low-dose CT against images reconstructed using standard-dose CT. The comparison covered regional SUV metrics, voxel-wise error analysis, structural similarity, lesion detectability, and the concordance of SUV-based clinical classifications. Across organs, the mean bias in SUVmean between the two reconstructions was −0.005, or −0.33%, for [18F]FDG studies and −0.025, or −0.72%, for [68Ga]DOTATATE studies, with narrow Bland–Altman limits of agreement.
The voxel-wise analysis added an important nuance. Where small differences did appear between images, they were driven primarily by respiratory misalignment between the PET and CT acquisitions rather than by noise in the ultra-low-dose CT itself. This distinction matters because it suggests the residual discrepancies are a known limitation of PET/CT acquisition in general, not a penalty introduced by pushing the CT dose down to ultra-low levels.
Clinically relevant measures held up as well. Lesion detectability was preserved, as were SUVmax and SUVpeak, the metrics most often used to characterise and monitor lesions. Across clinically relevant SUV thresholds, the two reconstruction approaches showed concordance greater than 95%, meaning that SUV-based classifications would rarely change depending on which CT protocol was used for attenuation correction.
The authors conclude that tin-filtered ultra-low-dose CT enables genuinely low-dose whole-body PET/CT, achieving total effective doses below 1 mSv while preserving quantitative accuracy, lesion conspicuity and clinical interpretability.
The study is published as open-access original research, with the clinical validation covering both [18F]FDG and [68Ga]DOTATATE PET/CT studies.
The phantom work established the dose-response relationship, and the prospective clinical studies confirmed that the selected tin-filtered protocol transfers from the bench to patient imaging with negligible quantitative cost.
Subject of Research: Ultra-low-dose CT protocols for attenuation correction in PET/CT imaging
Article Title: Quantitative accuracy and clinical utility of ultra-low dose CT protocols for attenuation correction in PET/CT
Article References: Angelis, G. I., Gholami, Y. H., Willowson, K. P., Ayessa, S. L., Bailey, E., Roach, P., Bailey, D. L., & Meikle, S. R. (2026). Quantitative accuracy and clinical utility of ultra-low dose CT protocols for attenuation correction in PET/CT. EJNMMI Physics. https://doi.org/10.1186/s40658-026-00948-3
Image Credits: AI Generated
DOI: 10.1186/s40658-026-00948-3
Keywords: PET/CT, attenuation correction, ultra-low-dose CT, tin filter, radiation dose, FDG, DOTATATE, SUV quantification, lesion detectability, paediatric imaging, NEMA phantom, dose reduction
News Source: Harold Sullivan. (October 29, 2026). Ultra-Low-Dose Tin-Filtered CT Slashes Radiation in PET/CT Without Losing Accuracy. Scienmag.
