Elisabeth Pfaehler1, Liesbet Mesotten2,3, Gem Kramer4, Michiel Thomeer2,5, Karolien Vanhove2,6, Johan de Jong7, Peter Adriaensens8, Otto S Hoekstra4, Ronald Boellaard7,4. 1. Department of Nuclear Medicine and Molecular Imaging, Medical Imaging Center, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands. e.a.g.pfaehler@umcg.nl. 2. Faculty of Medicine and Life Sciences, Hasselt University, Agoralaan Building D, 3590, Diepenbeek, Belgium. 3. Department of Nuclear Medicine, Ziekenhuis Oost Limburg, Schiepse Bos 6, 3600, Genk, Belgium. 4. Department of Radiology and Nuclear Medicine, VU University Medical Center, Amsterdam, The Netherlands. 5. Department of Respiratory Medicine, Ziekenhuis Oost Limburg, Schiepse Bos 6, 3600, Genk, Belgium. 6. Department of Respiratory Medicine, AZ Vesalius Hospital, Hazelereik 51, 3700, Tongeren, Belgium. 7. Department of Nuclear Medicine and Molecular Imaging, Medical Imaging Center, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands. 8. Institute for Materials Research (IMO) - Division Chemistry, Hasselt University, Agoralaan Building D, 3590, Diepenbeek, Belgium.
Abstract
BACKGROUND: Positron emission tomography (PET) is routinely used for cancer staging and treatment follow-up. Metabolic active tumor volume (MATV) as well as total MATV (TMATV-including primary tumor, lymph nodes and metastasis) and/or total lesion glycolysis derived from PET images have been identified as prognostic factor or for the evaluation of treatment efficacy in cancer patients. To this end, a segmentation approach with high precision and repeatability is important. However, the implementation of a repeatable and accurate segmentation algorithm remains an ongoing challenge. METHODS: In this study, we compare two semi-automatic artificial intelligence (AI)-based segmentation methods with conventional semi-automatic segmentation approaches in terms of repeatability. One segmentation approach is based on a textural feature (TF) segmentation approach designed for accurate and repeatable segmentation of primary tumors and metastasis. Moreover, a convolutional neural network (CNN) is trained. The algorithms are trained, validated and tested using a lung cancer PET dataset. The segmentation accuracy of both segmentation approaches is compared using the Jaccard coefficient (JC). Additionally, the approaches are externally tested on a fully independent test-retest dataset. The repeatability of the methods is compared with those of two majority vote (MV2, MV3) approaches, 41%SUVMAX, and a SUV > 4 segmentation (SUV4). Repeatability is assessed with test-retest coefficients (TRT%) and intraclass correlation coefficient (ICC). An ICC > 0.9 was regarded as representing excellent repeatability. RESULTS: The accuracy of the segmentations with the reference segmentation was good (JC median TF: 0.7, CNN: 0.73). Both segmentation approaches outperformed most other conventional segmentation methods in terms of test-retest coefficient (TRT% mean: TF: 13.0%, CNN: 13.9%, MV2: 14.1%, MV3: 28.1%, 41%SUVMAX: 28.1%, SUV4: 18.1%) and ICC (TF: 0.98, MV2: 0.97, CNN: 0.99, MV3: 0.73, SUV4: 0.81, and 41%SUVMAX: 0.68). CONCLUSION: The semi-automatic AI-based segmentation approaches used in this study provided better repeatability than conventional segmentation approaches. Moreover, both algorithms lead to accurate segmentations for both primary tumors as well as metastasis and are therefore good candidates for PET tumor segmentation.
BACKGROUND: Positron emission tomography (PET) is routinely used for cancer staging and treatment follow-up. Metabolic active tumor volume (MATV) as well as total MATV (TMATV-including primary tumor, lymph nodes and metastasis) and/or total lesion glycolysis derived from PET images have been identified as prognostic factor or for the evaluation of treatment efficacy in cancerpatients. To this end, a segmentation approach with high precision and repeatability is important. However, the implementation of a repeatable and accurate segmentation algorithm remains an ongoing challenge. METHODS: In this study, we compare two semi-automatic artificial intelligence (AI)-based segmentation methods with conventional semi-automatic segmentation approaches in terms of repeatability. One segmentation approach is based on a textural feature (TF) segmentation approach designed for accurate and repeatable segmentation of primary tumors and metastasis. Moreover, a convolutional neural network (CNN) is trained. The algorithms are trained, validated and tested using a lung cancer PET dataset. The segmentation accuracy of both segmentation approaches is compared using the Jaccard coefficient (JC). Additionally, the approaches are externally tested on a fully independent test-retest dataset. The repeatability of the methods is compared with those of two majority vote (MV2, MV3) approaches, 41%SUVMAX, and a SUV > 4 segmentation (SUV4). Repeatability is assessed with test-retest coefficients (TRT%) and intraclass correlation coefficient (ICC). An ICC > 0.9 was regarded as representing excellent repeatability. RESULTS: The accuracy of the segmentations with the reference segmentation was good (JC median TF: 0.7, CNN: 0.73). Both segmentation approaches outperformed most other conventional segmentation methods in terms of test-retest coefficient (TRT% mean: TF: 13.0%, CNN: 13.9%, MV2: 14.1%, MV3: 28.1%, 41%SUVMAX: 28.1%, SUV4: 18.1%) and ICC (TF: 0.98, MV2: 0.97, CNN: 0.99, MV3: 0.73, SUV4: 0.81, and 41%SUVMAX: 0.68). CONCLUSION: The semi-automatic AI-based segmentation approaches used in this study provided better repeatability than conventional segmentation approaches. Moreover, both algorithms lead to accurate segmentations for both primary tumors as well as metastasis and are therefore good candidates for PET tumor segmentation.
Entities:
Keywords:
Convolutional neural network; Repeatability; Textural segmentation; Tumor segmentation PET
Authors: Walter Jentzen; Lutz Freudenberg; Ernst G Eising; Melanie Heinze; Wolfgang Brandau; Andreas Bockisch Journal: J Nucl Med Date: 2007-01 Impact factor: 10.057
Authors: Mathieu Hatt; John A Lee; Charles R Schmidtlein; Issam El Naqa; Curtis Caldwell; Elisabetta De Bernardi; Wei Lu; Shiva Das; Xavier Geets; Vincent Gregoire; Robert Jeraj; Michael P MacManus; Osama R Mawlawi; Ursula Nestle; Andrei B Pugachev; Heiko Schöder; Tony Shepherd; Emiliano Spezi; Dimitris Visvikis; Habib Zaidi; Assen S Kirov Journal: Med Phys Date: 2017-05-18 Impact factor: 4.071
Authors: Jarkko Johansson; Kati Alakurtti; Juho Joutsa; Jussi Tohka; Ulla Ruotsalainen; Juha O Rinne Journal: Nucl Med Commun Date: 2016-10 Impact factor: 1.690
Authors: Hilke Vorwerk; Gabriele Beckmann; Michael Bremer; Maria Degen; Barbara Dietl; Rainer Fietkau; Tammo Gsänger; Robert Michael Hermann; Markus Karl Alfred Herrmann; Ulrike Höller; Michael van Kampen; Wolfgang Körber; Burkhard Maier; Thomas Martin; Michael Metz; Ronald Richter; Birgit Siekmeyer; Martin Steder; Daniela Wagner; Clemens Friedrich Hess; Elisabeth Weiss; Hans Christiansen Journal: Radiother Oncol Date: 2009-03-30 Impact factor: 6.280
Authors: Guilherme D Kolinger; David Vállez García; Gerbrand M Kramer; Virginie Frings; Egbert F Smit; Adrianus J de Langen; Rudi A J O Dierckx; Otto S Hoekstra; Ronald Boellaard Journal: EJNMMI Res Date: 2019-02-07 Impact factor: 3.138
Authors: Thomas W Georgi; Axel Zieschank; Kevin Kornrumpf; Lars Kurch; Osama Sabri; Dieter Körholz; Christine Mauz-Körholz; Regine Kluge; Stefan Posch Journal: PLoS One Date: 2022-04-18 Impact factor: 3.240
Authors: Elisabeth Pfaehler; Daniela Euba; Andreas Rinscheid; Otto S Hoekstra; Josee Zijlstra; Joyce van Sluis; Adrienne H Brouwers; Constantin Lapa; Ronald Boellaard Journal: EJNMMI Phys Date: 2022-08-09