A deep learning approach to generate synthetic CT in low field MR-guided radiotherapy for lung cases. (November 2022)

• Record Type: Journal Article
• Authors: Lenkowicz, Jacopo; Votta, Claudio; Nardini, Matteo; Quaranta, Flaviovincenzo; Catucci, Francesco; Boldrini, Luca; Vagni, Marica; Menna, Sebastiano; Placidi, Lorenzo; Romano, Angela; Chiloiro, Giuditta; Gambacorta, Maria Antonietta; Mattiucci, Gian Carlo; Indovina, Luca; Valentini, Vincenzo; Cusumano, Davide
• Is Part Of: Radiotherapy and oncology. Volume 176(2022)
• Page Start: 31
• Page End: 38
• Publication Date: 2022-11
• Subjects: Synthetic CT -- Artificial Intelligence -- MR-only Radiotherapy -- Deep Learning -- MR-guided Radiotherapy; Oncology -- Periodicals; Radiotherapy -- Periodicals; Tumors -- Periodicals; Medical Oncology -- Periodicals; Neoplasms -- radiotherapy -- Periodicals; Radiotherapy -- Periodicals; Radiothérapie -- Périodiques; Cancérologie -- Périodiques; Tumeurs -- Périodiques; Electronic journals; 616.9940642
• DOI: 10.1016/j.radonc.2022.08.028
• Languages: English
• ISSNs: 0167-8140

Abstract

Highlights: We propose a deep learning approach to generate synthetic CT from low tesla MR images. Synthetic CT was created for lung lesions enrolling 60 cases. Hybrid approach was proposed and compared with pure sCT. The images generated show an accuracy sufficient to calculate IMRT plans. This is the first experience of sCT generation in lung independently by B intensity. Abstract: Introduction: This study aims to apply a conditional Generative Adversarial Network (cGAN) to generate synthetic Computed Tomography (sCT) from 0.35 Tesla Magnetic Resonance (MR) images of the thorax. Methods: Sixty patients treated for lung lesions were enrolled and divided into training (32), validation (8), internal (10, TA ) and external (10, TB ) test set. Image accuracy of generated sCT was evaluated computing the mean absolute (MAE) and mean error (ME) with respect the original CT. Three treatment plans were calculated for each patient considering MRI as reference image: original CT, sCT (pure sCT) and sCT with GTV density override (hybrid sCT) were used as Electron Density (ED) map. Dose accuracy was evaluated comparing treatment plans in terms of gamma analysis and Dose Volume Histogram (DVH) parameters. Results: No significant difference was observed between the test sets for image and dose accuracy parameters. Considering the whole test cohort, a MAE of 54.9 ± 10.5 HU and a ME of 4.4 ± 7.4 HU was obtained. Mean gamma passing rates for 2%/2mm, and 3%/3mm tolerance criteria were 95.5 ± 5.9% and 98.2 ± 4.1% for pure sCT, 96.1 ± 5.1% and 98.5 ± 3.9% for hybrid sCT: the difference between the two approaches was significant (p = 0.01). As regards DVH analysis, differences in target parameters estimation were found to be within 5% using hybrid approach and 20% using pure sCT. Conclusion: The DL algorithm here presented can generate sCT images in the thorax with good image and dose accuracy, especially when the hybrid approach is used. The algorithm does not suffer from inter-scanner variability, making feasible the implementation of MR-only workflows for palliative treatments.