Dataset Information

Cerebral blood flow measurements with ¹⁵O-water PET using a non-invasive machine-learning-derived arterial input function.

ABSTRACT: Cerebral blood flow (CBF) can be measured with dynamic positron emission tomography (PET) of ¹⁵O-labeled water by using tracer kinetic modelling. However, for quantification of regional CBF, an arterial input function (AIF), obtained from arterial blood sampling, is required. In this work we evaluated a novel, non-invasive approach for input function prediction based on machine learning (MLIF), against AIF for CBF PET measurements in human subjects.Twenty-five subjects underwent two 10 min dynamic ¹⁵O-water brain PET scans with continuous arterial blood sampling, before (baseline) and following acetazolamide medication. Three different image-derived time-activity curves were automatically segmented from the carotid arteries and used as input into a Gaussian process-based AIF prediction model, considering both baseline and acetazolamide scans as training data. The MLIF approach was evaluated by comparing AIF and MLIF curves, as well as whole-brain grey matter CBF values estimated by kinetic modelling derived with either AIF or MLIF.The results showed that AIF and MLIF curves were similar and that corresponding CBF values were highly correlated and successfully differentiated before and after acetazolamide medication. In conclusion, our non-invasive MLIF method shows potential to replace the AIF obtained from blood sampling for CBF measurements using ¹⁵O-water PET and kinetic modelling.

SUBMITTER: Kuttner S

PROVIDER: S-EPMC8392760 | biostudies-literature | 2021 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Cerebral blood flow measurements with <sup>15</sup>O-water PET using a non-invasive machine-learning-derived arterial input function.

Kuttner Samuel S Wickstrøm Kristoffer Knutsen KK Lubberink Mark M Tolf Andreas A Burman Joachim J Sundset Rune R Jenssen Robert R Appel Lieuwe L Axelsson Jan J

Journal of cerebral blood flow and metabolism : official journal of the International Society of Cerebral Blood Flow and Metabolism 20210208 9

Cerebral blood flow (CBF) can be measured with dynamic positron emission tomography (PET) of <sup>15</sup>O-labeled water by using tracer kinetic modelling. However, for quantification of regional CBF, an arterial input function (AIF), obtained from arterial blood sampling, is required. In this work we evaluated a novel, non-invasive approach for input function prediction based on machine learning (MLIF), against AIF for CBF PET measurements in human subjects.Twenty-five subjects underwent two 1 ...[more]

PMID: 33557691

Similar Datasets

Project description:BackgroundDopamine transporter (DAT) imaging is used in the diagnostic work-up in suspected parkinsonian syndromes and dementia with Lewy bodies but cannot differentiate between these syndromes, and an extra brain imaging examination of the regional cerebral blood flow (rCBF) or glucose metabolism is often needed for differential diagnosis. The requirement of two different imaging examinations is resource-consuming and inconvenient for the patients. Therefore, imaging of both cortical blood flow and DAT imaging with the same radiotracer would be more convenient and cost-effective. The aim of this study was to test whether relative regional cerebral blood flow (rCBFR) can be measured with the DAT-specific positron emission tomography (PET) tracer [18F]FE-PE2I (FE-PE2I), by validation with cerebral perfusion measured with [15O]H2O PET (H2O).MethodsThe rCBFR was quantified by kinetic modeling for FE-PE2I (R1) and H2O (F). The R1 was calculated using the simplified reference tissue model, and F was calculated with a modified Koopman double-integration method. The linear relationship and intraclass correlation (ICC) between R1 and F were tested in image data derived from 29 patients with recent onset parkinsonism and 30 healthy controls.ResultsThere was a strong linear correlation across all subjects between R1 and F in the frontal, parietal, temporal, cingulate and occipital cortex as well as in the striatum (r ≥ 0.731-0.905, p < 0.001) with a good-to-excellent ICC, ranging from 0.727 to 0.943 (p < 0.001).ConclusionsOur results suggest that FE-PE2I may be used as a proxy for cerebral perfusion, thus potentially serving as a radiotracer for assessment of both DAT availability and rCBFR in one single dynamic scan. This could be valuable in the differential diagnosis of parkinsonian syndromes.Trial registrationEUDRA-CT 2015-003045-26. Registered 23 October 2015 https://www.clinicaltrialsregister.eu/ctr-search/search?query=2015-003045-26.

Dataset Information

Cerebral blood flow measurements with ¹⁵O-water PET using a non-invasive machine-learning-derived arterial input function.

Publications

Cerebral blood flow measurements with <sup>15</sup>O-water PET using a non-invasive machine-learning-derived arterial input function.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets

Dataset Information

Cerebral blood flow measurements with 15O-water PET using a non-invasive machine-learning-derived arterial input function.

Publications

Cerebral blood flow measurements with <sup>15</sup>O-water PET using a non-invasive machine-learning-derived arterial input function.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets

Cerebral blood flow measurements with ¹⁵O-water PET using a non-invasive machine-learning-derived arterial input function.