AI based automatic measurement of split renal function in [18F]PSMA-1007 PET/CT
(2025) In EJNMMI reports 9.- Abstract
Background: Prostate-specific membrane antigen (PSMA) is an important target for positron emission tomography (PET) with computed tomography (CT) in prostate cancer. In addition to overexpression in prostate cancer cells, PSMA is expressed in healthy cells in the proximal tubules of the kidneys. Consequently, PSMA PET is being explored for renal functional imaging. Left and right renal uptake of PSMA targeted radiopharmaceuticals have shown strong correlations to split renal function (SRF) as determined by other methods. Manual segmentation of kidneys in PET images is, however, time consuming, making this method of measuring SRF impractical. In this study, we designed, trained and validated an artificial intelligence (AI) model for... (More)
Background: Prostate-specific membrane antigen (PSMA) is an important target for positron emission tomography (PET) with computed tomography (CT) in prostate cancer. In addition to overexpression in prostate cancer cells, PSMA is expressed in healthy cells in the proximal tubules of the kidneys. Consequently, PSMA PET is being explored for renal functional imaging. Left and right renal uptake of PSMA targeted radiopharmaceuticals have shown strong correlations to split renal function (SRF) as determined by other methods. Manual segmentation of kidneys in PET images is, however, time consuming, making this method of measuring SRF impractical. In this study, we designed, trained and validated an artificial intelligence (AI) model for automatic renal segmentation and measurement of SRF in [18F]PSMA-1007 PET images. Results: Kidneys were segmented in 135 [18F]PSMA-1007 PET/CT studies used to train the AI model. The model was evaluated in 40 test studies. Left renal function percentage (LRF%) measurements ranged from 40 to 67%. Spearman correlation coefficients for LRF% measurements ranged between 0.98 and 0.99 when comparing segmentations made by 3 human readers and the AI model. The largest LRF% difference between any measurements in a single case was 3 percentage points. The AI model produced measurements similar to those of human readers. Conclusions: Automatic measurement of SRF in PSMA PET is feasible. A potential use could be to provide additional data in investigation of renal functional impairment in patients treated for prostate cancer.
(Less)
- author
- Valind, Kristian
LU
; Ulén, Johannes LU ; Gålne, Anni LU
; Jögi, Jonas LU
; Minarik, David LU and Trägårdh, Elin LU
- organization
- publishing date
- 2025-12
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- AI, CNN, PET, PSMA, Renal function, Segmentation, SRF
- in
- EJNMMI reports
- volume
- 9
- article number
- 20
- publisher
- Springer Nature
- external identifiers
-
- pmid:40518464
- scopus:105007992322
- ISSN
- 3005-074X
- DOI
- 10.1186/s41824-025-00254-8
- language
- English
- LU publication?
- yes
- additional info
- Publisher Copyright: © The Author(s) 2025.
- id
- 6669292c-657a-47cf-89dc-f04afa46e1c0
- date added to LUP
- 2025-06-23 09:01:57
- date last changed
- 2025-07-07 09:12:12
@article{6669292c-657a-47cf-89dc-f04afa46e1c0, abstract = {{<p>Background: Prostate-specific membrane antigen (PSMA) is an important target for positron emission tomography (PET) with computed tomography (CT) in prostate cancer. In addition to overexpression in prostate cancer cells, PSMA is expressed in healthy cells in the proximal tubules of the kidneys. Consequently, PSMA PET is being explored for renal functional imaging. Left and right renal uptake of PSMA targeted radiopharmaceuticals have shown strong correlations to split renal function (SRF) as determined by other methods. Manual segmentation of kidneys in PET images is, however, time consuming, making this method of measuring SRF impractical. In this study, we designed, trained and validated an artificial intelligence (AI) model for automatic renal segmentation and measurement of SRF in [<sup>18</sup>F]PSMA-1007 PET images. Results: Kidneys were segmented in 135 [<sup>18</sup>F]PSMA-1007 PET/CT studies used to train the AI model. The model was evaluated in 40 test studies. Left renal function percentage (LRF%) measurements ranged from 40 to 67%. Spearman correlation coefficients for LRF% measurements ranged between 0.98 and 0.99 when comparing segmentations made by 3 human readers and the AI model. The largest LRF% difference between any measurements in a single case was 3 percentage points. The AI model produced measurements similar to those of human readers. Conclusions: Automatic measurement of SRF in PSMA PET is feasible. A potential use could be to provide additional data in investigation of renal functional impairment in patients treated for prostate cancer.</p>}}, author = {{Valind, Kristian and Ulén, Johannes and Gålne, Anni and Jögi, Jonas and Minarik, David and Trägårdh, Elin}}, issn = {{3005-074X}}, keywords = {{AI; CNN; PET; PSMA; Renal function; Segmentation; SRF}}, language = {{eng}}, publisher = {{Springer Nature}}, series = {{EJNMMI reports}}, title = {{AI based automatic measurement of split renal function in [<sup>18</sup>F]PSMA-1007 PET/CT}}, url = {{http://dx.doi.org/10.1186/s41824-025-00254-8}}, doi = {{10.1186/s41824-025-00254-8}}, volume = {{9}}, year = {{2025}}, }