PET/MR imaging of bone lesions—Implications for PET quantification from imperfect attenuation correction
Open accessAbstract
Purpose Accurate attenuation correction (AC) is essential for quantitative analysis of PET tracer distribution. In MR, the lack of cortical bone signal makes bone segmentation difficult and may require implementation of special sequences. The purpose of this study was to evaluate the need for accurate bone segmentation in MR-based AC for whole-body PET/MR imaging.Methods In 22 patients undergoing sequential PET/CT and 3-T MR imaging, modified CT AC maps were produced by replacing pixels with values of >100 HU, representing mostly bone structures, by pixels with a constant value of 36 HU corresponding to soft tissue, thereby simulating current MR-derived AC maps. A total of 141 FDG-positive osseous lesions and 50 soft-tissue lesions adjacent to bones were evaluated. The mean standardized uptake value (SUVmean) was measured in each lesion in PET images reconstructed once using the standard AC maps and once using the modified AC maps. Subsequently, the errors in lesion tracer uptake for the modified PET images were calculated using the standard PET image as a reference.Results Substitution of bone by soft tissue values in AC maps resulted in an underestimation of tracer uptake in osseous and soft tissue lesions adjacent to bones of 11.2 ± 5.4% (range 1.5-30.8%) and 3.2 ± 1.7% (range 0.2-4%), respectively. Analysis of the spine and pelvic osseous lesions revealed a substantial dependence of the error on lesion composition. For predominantly sclerotic spine lesions, the mean underestimation was 15.9 ± 3.4% (range 9.9-23.5%) and for osteolytic spine lesions, 7.2 ± 1.7% (range 4.9-9.3%), respectively.Conclusion CT data simulating treating bone as soft tissue as is currently done in MR maps for PET AC leads to a substantial underestimation of tracer uptake in bone lesions and depends on lesion composition, the largest error being seen in sclerotic lesions. Therefore, depiction of cortical bone and other calcified areas in MR AC maps is necessary for accurate quantification of tracer uptake values in PET/MR imaging.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Systematic underestimation of SUV in bone tissue or lesions close to bone of >10% compared to PET/CT has been described, whereas SUV differences were <5% in other normal tissues.
“Systematic underestimation of SUV in bone tissue or lesions close to the bone of >10% compared to PET/CT has been described [60,62], whereas SUV differences were <5% in other normal tissues [60].”
From Influences on PET Quantification and Interpretation (2022) · cites this paper as [62]
The Results report mean SUV underestimation of 11.2% across bone lesions and errors generally below 5% in other body areas. But Table 2 reports a mean underestimation of only 3.2% for soft-tissue lesions near bone, not >10%. The comparison was between PET images reconstructed with modified and standard attenuation-correction maps, rather than a direct PET/MR-versus-PET/CT measurement. (LLM, conf=0.88)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
The citation lists haven’t been fetched yet
References 14 · Cited by 251
- Authors
- Samarin, A., Burger, C., Wollenweber, S. D., Crook, D. W., Burger, I. A., Schmid, D. T., von Schulthess, G. K., & Kuhn, F. P.
- Affiliations
- University Hospital Zurich (CH)
- Venue
- European Journal of Nuclear Medicine and Molecular Imaging
- Published
- 2012-04-14
- Volume
- 39
- Issue
- 7
- Pages
- 1154-1160
- ISSN
- 1619-7070 (ISSN-L)
- 1619-7089
- OpenAlex
- W2037425560
- Cited by
- 251
- References
- 14
- Citations by year
- 2026: 2 · 2025: 2 · 2024: 4 · 2023: 9 · 2022: 9 · 2021: 11 · 2020: 19 · 2019: 22 · 2018: 21 · 2017: 28 · 2016: 20 · 2015: 31 · 2014: 30 · 2013: 31 · 2012: 12
- FWCI
- 25.03
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Licence
- www.springer.com/tdm (tdm)
- Locations
- European Journal of Nuclear Medicine and Molecular Imaging (publishedVersion)
- PubMed (publishedVersion)
- Zurich Open Repository and Archive (University of Zurich) (acceptedVersion, other-oa)
- reroDoc Digital Library (submittedVersion)
- Zurich Open Repository and Archive (University of Zurich)
- Topics
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Medical Imaging Techniques and Applications
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Advanced MRI Techniques and Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Medical Image Segmentation Techniques
- Keywords
- PET/MR, attenuation correction, bone segmentation, cortical bone, FDG uptake, bone lesions, standardized uptake value, PET quantification, sclerotic bone lesions, osteolytic lesions, soft tissue lesions
- MeSH terms
- Adult, Aged, Bone Neoplasms, Humans, Image Processing, Computer-Assisted, Magnetic Resonance Imaging, Male, Middle Aged, Retrospective Studies, Tomography, X-Ray Computed, Radiopharmaceuticals, Fluorodeoxyglucose F18, Positron-Emission Tomography
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
