Deep-learning-derived input function in dynamic [18F]FDG PET imaging of mice

被引:0
|
作者
Kuttner, Samuel [1 ,2 ,3 ]
Luppino, Luigi T. [2 ]
Convert, Laurence [4 ,5 ]
Sarrhini, Otman [4 ,5 ]
Lecomte, Roger [4 ,5 ,6 ]
Kampffmeyer, Michael C. [2 ]
Sundset, Rune [1 ,3 ]
Jenssen, Robert [2 ]
机构
[1] Univ Hosp North Norway, PET Imaging Ctr, Tromso, Norway
[2] UiT Arctic Univ Norway, Dept Phys & Technol, UiT Machine Learning Grp, Tromso, Norway
[3] UiT Arctic Univ Norway, Dept Clin Med, Nucl Med & Radiat Biol Res Grp, Tromso, Norway
[4] Univ Sherbrooke, Sherbrooke Mol Imaging Ctr CRCHUS, Sherbrooke, PQ, Canada
[5] Univ Sherbrooke, Dept Nucl Med & Radiobiol, Sherbrooke, PQ, Canada
[6] Imaging Res & Technol Inc, Sherbrooke, PQ, Canada
来源
关键词
dynamic positron emission tomography (PET); small-animal PET 18F-FDG PET/CT; Patlak analysis; arterial input function estimation; glucose metabolism; deep learning; prediction model; SMALL-ANIMAL PET; PARTIAL-VOLUME CORRECTION; BRAIN TRANSFER CONSTANTS; GLUCOSE-METABOLISM; F-18-FDG PET; GRAPHICAL EVALUATION; BLOOD; ARTERIAL; QUANTIFICATION;
D O I
10.3389/fnume.2024.1372379
中图分类号
R8 [特种医学]; R445 [影像诊断学];
学科分类号
1002 ; 100207 ; 1009 ;
摘要
Dynamic positron emission tomography and kinetic modeling play a critical role in tracer development research using small animals. Kinetic modeling from dynamic PET imaging requires accurate knowledge of an input function, ideally determined through arterial blood sampling. Arterial cannulation in mice, however, requires complex, time-consuming and terminal surgery, meaning that longitudinal studies are impossible. The aim of the current work was to develop and evaluate a non-invasive, deep-learning-based prediction model (DLIF) that directly takes the PET data as input to predict a usable input function. We first trained and evaluated the DLIF model on 68 [18F]Fluorodeoxyglucose mouse scans with image-derived targets using cross validation. Subsequently, we evaluated the performance of a trained DLIF model on an external dataset consisting of 8 mouse scans where the input function was measured by continuous arterial blood sampling. The results showed that the predicted DLIF and image-derived targets were similar, and the net influx rate constants following from Patlak modeling using DLIF as input function were strongly correlated to the corresponding values obtained using the image-derived input function. There were somewhat larger discrepancies when evaluating the model on the external dataset, which could be attributed to systematic differences in the experimental setup between the two datasets. In conclusion, our non-invasive DLIF prediction method may be a viable alternative to arterial blood sampling in small animal [18F]FDG imaging. With further validation, DLIF could overcome the need for arterial cannulation and allow fully quantitative and longitudinal experiments in PET imaging studies of mice.
引用
收藏
页数:11
相关论文
共 50 条
  • [1] Deep learning derived input-function in dynamic 18F-FDG PET imaging of mice
    Kuttner, S.
    Luppino, L. T.
    Wickstrom, K. K.
    Midtbo, N. T. D.
    Dorraji, E.
    Oteiza, A.
    Martin-Armas, M.
    Fenton, K.
    Convert, L.
    Sarrhini, O.
    Lecomte, R.
    Kampffmeyer, M. C.
    Jenssen, R.
    Axelsson, J.
    Sundset, R.
    EUROPEAN JOURNAL OF NUCLEAR MEDICINE AND MOLECULAR IMAGING, 2022, 49 (SUPPL 1) : S245 - S245
  • [2] Machine learning derived input-function in a dynamic 18F-FDG PET study of mice
    Kuttner, Samuel
    Wickstrom, Kristoffer Knutsen
    Kalda, Gustav
    Dorraji, S. Esmaeil
    Martin-Armas, Montserrat
    Oteiza, Ana
    Jenssen, Robert
    Fenton, Kristin
    Sundset, Rune
    Axelsson, Jan
    BIOMEDICAL PHYSICS & ENGINEERING EXPRESS, 2020, 6 (01)
  • [3] Extraction of Input Function from Rat [18F]FDG PET Images
    Kudomi, Nobuyuki
    Bucci, Marco
    Oikonen, Vesa
    Silvennoinen, Mika
    Kainulainen, Heikki
    Nuutila, Pirjo
    Iozzo, Patricia
    Roivainen, Anne
    MOLECULAR IMAGING AND BIOLOGY, 2011, 13 (06) : 1241 - 1249
  • [4] Extraction of Input Function from Rat [18F]FDG PET Images
    Nobuyuki Kudomi
    Marco Bucci
    Vesa Oikonen
    Mika Silvennoinen
    Heikki Kainulainen
    Pirjo Nuutila
    Patricia Iozzo
    Anne Roivainen
    Molecular Imaging and Biology, 2011, 13 : 1241 - 1249
  • [5] Comparison of eight methods for the estimation of the image-derived input function in dynamic [18F]-FDG PET human brain studies
    Zanotti-Fregonara, Paolo
    Fadaili, El Mostafa
    Maroy, Renaud
    Comtat, Claude
    Souloumiac, Antoine
    Jan, Sebastien
    Ribeiro, Maria-Joao
    Gaura, Veronique
    Bar-Hen, Avner
    Trebossen, Regine
    JOURNAL OF CEREBRAL BLOOD FLOW AND METABOLISM, 2009, 29 (11): : 1825 - 1835
  • [6] Improved Derivation of Input Function in Dynamic Mouse [18F]FDG PET Using Bladder Radioactivity Kinetics
    Koon-Pong Wong
    Xiaoli Zhang
    Sung-Cheng Huang
    Molecular Imaging and Biology, 2013, 15 : 486 - 496
  • [7] Improved Derivation of Input Function in Dynamic Mouse [18F]FDG PET Using Bladder Radioactivity Kinetics
    Wong, Koon-Pong
    Zhang, Xiaoli
    Huang, Sung-Cheng
    MOLECULAR IMAGING AND BIOLOGY, 2013, 15 (04) : 486 - 496
  • [8] Is the Physical Decay Correction of the F-18-FDG Input Function in Dynamic PET Imaging Justified?
    Laffon, Eric
    Barret, Olivier
    Marthan, Roger
    Ducassou, Dominique
    JOURNAL OF NUCLEAR MEDICINE TECHNOLOGY, 2009, 37 (02) : 111 - 113
  • [9] Hibernoma:: 18F FDG PET/CT imaging
    Subramaniam, Rathan M.
    Clayton, Amy C.
    Karantanis, Dimitrios
    Collins, Douglas A.
    JOURNAL OF THORACIC ONCOLOGY, 2007, 2 (06) : 569 - 570
  • [10] A deep learning model for generating [18F]FDG PET Images from early-phase [18F]Florbetapir and [18F]Flutemetamol PET images
    Sanaat, Amirhossein
    Boccalini, Cecilia
    Mathoux, Gregory
    Perani, Daniela
    Frisoni, Giovanni B.
    Haller, Sven
    Montandon, Marie-Louise
    Rodriguez, Cristelle
    Giannakopoulos, Panteleimon
    Garibotto, Valentina
    Zaidi, Habib
    EUROPEAN JOURNAL OF NUCLEAR MEDICINE AND MOLECULAR IMAGING, 2024, 51 (12) : 3518 - 3531