Deep learning based unpaired image-to-image translation applications for medical physics: a systematic review

Abstract

Purpose: There is a growing number of publications on the application of unpaired image-to-image (I2I) translation in medical imaging. However, a systematic review covering the current state of this topic for medical physicists is lacking. The aim of this article is to provide a comprehensive review of current challenges and opportunities for medical physicists to apply I2I translation in practice. Methods and Materials: The PubMed electronic database was searched using terms referring to unpaired (unsupervised), I2I translation, and medical imaging. This review has been reported in compliance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. From each full-text article, we extracted information extracted regarding technical and clinical applications of methods, Transparent Reporting for Individual Prognosis Or Diagnosis (TRIPOD) study type, performance of algorithm and accessibility of source code and pre-trained models. Results: Among 461 unique records, 55 full-text articles were included in the review. The major technical applications described in the selected literature are segmentation (26 studies), unpaired domain adaptation (18 studies), and denoising (8 studies). In terms of clinical applications, unpaired I2I translation has been used for automatic contouring of regions of interest in MRI, CT, X-ray and ultrasound images, fast MRI or low dose CT imaging, CT or MRI only based radiotherapy planning, etc. Only 5 studies validated their models using an independent test set and none were externally validated by independent researchers. Finally, 12 articles published their source code and only one study published their pre-trained models. Conclusion: I2I translation of medical images offers a range of valuable applications for medical physicists. However, the scarcity of external validation studies of I2I models and the shortage of publicly available pre-trained models limits the immediate applicability of the proposed methods in practice.

Analysis of disentangled representation learning for high-resolution dynamic MRI synthesis

“…Instead, disentangled representations are formulated on the assumption that it is possible to factorize an input image in a set of meaningful, independent features, each representing a mode of variation within the image, and encode them into distinct dimensions. 15,16 Despite similar application field as GAN, encouraging a representation to be disentangled benefits from a more interpretable generation process, a more controllable synthesis, increase the transferability between tasks and is particularly suitable for privacy aware application such as federated learning applications. 17,18 In this work, we focus on pediatric dynamic MRI data 19 acquired to study equinus which is the most common musculo-skeletal deformity in children with cerebral palsy.…”

Segmentation and quantitative analysis of optical coherence tomography (OCT) images of laser burned skin based on deep learning

“…Deep learning methods can be used to help achieve automated, high-speed, and highly accurate assisted diagnosis for various diseases (Esteva et al 2019, Lee et al 2019, Chen et al 2023, De Biase et al 2023. U-Net is a deep learning model that uses a universal convolutional neural network (CNN) to extract imaging features from complex data sets (Ronneberger et al 2015).…”