Pablo Borrelli1, John Ly2,3, Reza Kaboteh1, Johannes Ulén4, Olof Enqvist4,5, Elin Trägårdh6,7, Lars Edenbrandt1,8. 1. Department of Clinical Physiology, Sahlgrenska University Hospital, Gothenburg, Sweden. 2. Department of Radiology, Kristianstad Hospital, Kristianstad, Sweden. john.ly@med.lu.se. 3. Department of Translational Medicine and Wallenberg Center for Molecular Medicine, Lund University, Malmö, Sweden. john.ly@med.lu.se. 4. Eigenvision AB, Malmö, Sweden. 5. Department of Electrical Engineering, Chalmers University of Technology, Gothenburg, Sweden. 6. Department of Translational Medicine and Wallenberg Center for Molecular Medicine, Lund University, Malmö, Sweden. 7. Department of Clinical Physiology and Nuclear Medicine, Skåne University Hospital, Malmö, Sweden. 8. Department of Molecular and Clinical Medicine, Institute of Medicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden.
Abstract
BACKGROUND: [18F]-fluorodeoxyglucose (FDG) positron emission tomography with computed tomography (PET-CT) is a well-established modality in the work-up of patients with suspected or confirmed diagnosis of lung cancer. Recent research efforts have focused on extracting theragnostic and textural information from manually indicated lung lesions. Both semi-automatic and fully automatic use of artificial intelligence (AI) to localise and classify FDG-avid foci has been demonstrated. To fully harness AI's usefulness, we have developed a method which both automatically detects abnormal lung lesions and calculates the total lesion glycolysis (TLG) on FDG PET-CT. METHODS: One hundred twelve patients (59 females and 53 males) who underwent FDG PET-CT due to suspected or for the management of known lung cancer were studied retrospectively. These patients were divided into a training group (59%; n = 66), a validation group (20.5%; n = 23) and a test group (20.5%; n = 23). A nuclear medicine physician manually segmented abnormal lung lesions with increased FDG-uptake in all PET-CT studies. The AI-based method was trained to segment the lesions based on the manual segmentations. TLG was then calculated from manual and AI-based measurements, respectively and analysed with Bland-Altman plots. RESULTS: The AI-tool's performance in detecting lesions had a sensitivity of 90%. One small lesion was missed in two patients, respectively, where both had a larger lesion which was correctly detected. The positive and negative predictive values were 88% and 100%, respectively. The correlation between manual and AI TLG measurements was strong (R2 = 0.74). Bias was 42 g and 95% limits of agreement ranged from - 736 to 819 g. Agreement was particularly high in smaller lesions. CONCLUSIONS: The AI-based method is suitable for the detection of lung lesions and automatic calculation of TLG in small- to medium-sized tumours. In a clinical setting, it will have an added value due to its capability to sort out negative examinations resulting in prioritised and focused care on patients with potentially malignant lesions.
BACKGROUND:[18F]-fluorodeoxyglucose (FDG) positron emission tomography with computed tomography (PET-CT) is a well-established modality in the work-up of patients with suspected or confirmed diagnosis of lung cancer. Recent research efforts have focused on extracting theragnostic and textural information from manually indicated lung lesions. Both semi-automatic and fully automatic use of artificial intelligence (AI) to localise and classify FDG-avid foci has been demonstrated. To fully harness AI's usefulness, we have developed a method which both automatically detects abnormal lung lesions and calculates the total lesion glycolysis (TLG) on FDG PET-CT. METHODS: One hundred twelve patients (59 females and 53 males) who underwent FDG PET-CT due to suspected or for the management of known lung cancer were studied retrospectively. These patients were divided into a training group (59%; n = 66), a validation group (20.5%; n = 23) and a test group (20.5%; n = 23). A nuclear medicine physician manually segmented abnormal lung lesions with increased FDG-uptake in all PET-CT studies. The AI-based method was trained to segment the lesions based on the manual segmentations. TLG was then calculated from manual and AI-based measurements, respectively and analysed with Bland-Altman plots. RESULTS: The AI-tool's performance in detecting lesions had a sensitivity of 90%. One small lesion was missed in two patients, respectively, where both had a larger lesion which was correctly detected. The positive and negative predictive values were 88% and 100%, respectively. The correlation between manual and AI TLG measurements was strong (R2 = 0.74). Bias was 42 g and 95% limits of agreement ranged from - 736 to 819 g. Agreement was particularly high in smaller lesions. CONCLUSIONS: The AI-based method is suitable for the detection of lung lesions and automatic calculation of TLG in small- to medium-sized tumours. In a clinical setting, it will have an added value due to its capability to sort out negative examinations resulting in prioritised and focused care on patients with potentially malignant lesions.
Authors: Judi N A van Diessen; Matthew La Fontaine; Michel M van den Heuvel; Erik van Werkhoven; Iris Walraven; Wouter V Vogel; José S A Belderbos; Jan-Jakob Sonke Journal: Radiother Oncol Date: 2019-11-22 Impact factor: 6.280
Authors: Joan Fledelius; Anne Winther-Larsen; Azza A Khalil; Karin Hjorthaug; Jørgen Frøkiær; Peter Meldgaard Journal: Am J Nucl Med Mol Imaging Date: 2018-02-05
Authors: Nichole T Tanner; Alexander Porter; Michael K Gould; Xiao-Jun Li; Anil Vachani; Gerard A Silvestri Journal: Chest Date: 2017-01-20 Impact factor: 9.410
Authors: D Groheux; G Quere; E Blanc; C Lemarignier; L Vercellino; C de Margerie-Mellon; P Merlet; S Querellou Journal: Diagn Interv Imaging Date: 2016-08-24 Impact factor: 4.026
Authors: Edwin A Usmanij; Lioe-Fee de Geus-Oei; Esther G C Troost; Liesbeth Peters-Bax; Erik H F M van der Heijden; Johannes H A M Kaanders; Wim J G Oyen; Olga C J Schuurbiers; Johan Bussink Journal: J Nucl Med Date: 2013-07-17 Impact factor: 10.057
Authors: Paul Flechsig; Amit Mehndiratta; Uwe Haberkorn; Clemens Kratochwil; Frederik L Giesel Journal: Semin Nucl Med Date: 2015-07 Impact factor: 4.446
Authors: Michael K Gould; Jessica Donington; William R Lynch; Peter J Mazzone; David E Midthun; David P Naidich; Renda Soylemez Wiener Journal: Chest Date: 2013-05 Impact factor: 9.410
Authors: Yu Guo; Yuanming Feng; Jian Sun; Ning Zhang; Wang Lin; Yu Sa; Ping Wang Journal: Comput Math Methods Med Date: 2014-05-29 Impact factor: 2.238
Authors: Pablo Borrelli; José Luis Loaiza Góngora; Reza Kaboteh; Johannes Ulén; Olof Enqvist; Elin Trägårdh; Lars Edenbrandt Journal: EJNMMI Phys Date: 2022-02-03