AUTHOR=Lamontagne-Caron Rémi , Duchesne Simon 

TITLE=A scoping review of magnetic resonance angiography and perfusion image synthesis

JOURNAL=Frontiers in Dementia

VOLUME=Volume 3 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/dementia/articles/10.3389/frdem.2024.1408782

DOI=10.3389/frdem.2024.1408782

ISSN=2813-3919

ABSTRACT=Deregulation of the cerebrovascular system has been linked to neurodegeneration, part of a putative causal pathway into etiologies such as Alzheimer's disease (AD). In medical imaging, time-of-flight magnetic resonance angiography (TOF-MRA) and perfusion MRI are the most common modalities used to study this system. However, due to acquisition time and lack of resources, many large scale studies of AD did not or are not acquiring these images as part of their study protocols; this creates a conundrum, as the lack of evidence limits our knowledge of the extent between the cerebrovascular system and neurodegenerative diseases. Meanwhile, deep learning approaches have been used in recent developments to generate synthetic medical images from existing contrasts. In this review, we study the use of artificial intelligence in the generation of synthetic TOF-MRA and perfusion-related images from existing neuroanatomical and neurovascular acquisitions (e.g., T1-, T2-, and FLAIR-weighted imaged) for the study of the cerebrovascular system. Although no studies used these specific contrasts, thirteen papers managed to generate either TOF-MRA or perfusion MRI from a variety of imaging modalities. From those, the studies from Huang et al., Li et al. and Fujita et al. showed that structural MRI on its own can be used to synthesize perfusion map. Other studies demonstrated that synthetic images could have a greater signal-to-noise ratio compared to real images and that some models trained on healthy subjects could generalize their outputs to an unseen population, such as stroke patients. These findings suggest that generating synthetic TOF-MRA and perfusion MRI images holds significant potential for enhancing neurovascular studies, particularly in cases where direct acquisition is not feasible. This approach could also provide valuable insights for retrospective studies of several cerebrovascular related diseases such as stroke and dementia. While promising, further research is needed to refine these models using larger datasets, assess their sensitivity and specificity, and ensure their applicability across diverse populations. The use of similar models to generate TOF-MRA and perfusion MRI using commonly acquired data appears to be the next feet necessary for the retrospective study of the cerebrovascular system and elucidate its role in the development of dementia.