Deep learning based synthetic‐CT generation in radiotherapy and PET: A review

Abstract

Recently, deep learning (DL)-based methods for the generation of synthetic computed tomography (sCT) have received significant research attention as an alternative to classical ones. We present here a systematic review of these methods by grouping them into three categories, according to their clinical applications: I) to replace CT in magnetic resonance (MR)-based treatment planning, II) facilitate cone-beam computed tomography (CBCT)-based image-guided adaptive radiotherapy, and III) derive attenuation maps for the correction of positron emission tomography (PET).Appropriate database searching was performed on journal articles published between January 2014 and December 2020.The DL methods' key characteristics were extracted from each eligible study, and a comprehensive comparison among network architectures and metrics was reported. A detailed review of each category was given, highlighting essential contributions, identifying specific challenges, and summarising the achievements. Lastly, the statistics of all the cited works from various aspects were analysed, revealing the popularity and future trends and the potential of DL-based sCT generation. The current status of DLbased sCT generation was evaluated, assessing the clinical readiness of the presented methods.

Adaptive Proton Therapy of Pediatric Head and Neck Cases Using MRI-Based Synthetic CTs: Initial Experience of the Prospective KiAPT Study

“…Other studies report about direct sCTs from MRI (see Refs. [ 58 , 59 , 60 , 61 , 62 ] and references therein). Especially, the machine learning based implementation is gaining popularity.…”

Training of deep cross‐modality conversion models with a small data set, and their application in megavoltage CT to kilovoltage CT conversion

“…In recent years, cross-modality conversion associated with different imaging modalities has gained considerable attention in the radiotherapy (RT) domain. [7][8][9] In particular, the application of magnetic resonance imaging (MRI) to computed tomography (CT) conversion has been widely investigated in the context of MRI-based and MRI-guided RT. 10,11 Cone-beam computed tomography (CBCT) to CT conversion has been adopted to realize a higher accuracy of IGRT 12 and facilitate its application to ART.…”

“…In particular, deep learning has enabled image translation more than image quality enhancement. In recent years, cross‐modality conversion associated with different imaging modalities has gained considerable attention in the radiotherapy (RT) domain 7–9 . In particular, the application of magnetic resonance imaging (MRI) to computed tomography (CT) conversion has been widely investigated in the context of MRI‐based and MRI‐guided RT 10,11 .…”

Deep learning for Dixon MRI-based attenuation correction in PET/MRI of head and neck cancer patients

“…Recently, deep learning has emerged as an alternative MR-AC strategy [14] for the brain [15][16][17] and the pelvic region [18,19] and has shown accurate and robust performances. Although several studies using deep convolutional neural networks for converting MRI to CT also exist for the head/neck region and have been applied for use in radiotherapy [20][21][22][23][24], only limited effort has been put into detailed evaluation of the effect on PET AC where the lower photon energy (511 keV compared to MeV typically used radiotherapy) increases the sensitivity to wrong tissue attenuation coefficients.…”