Back to Projects

AI-enabled Imaging, Emerging Imaging

Quantitative myocardial tissue characterisation using a new affordable low magnetic field (0.55T) MRI scanner

Project ID: 2022_025

1st Supervisor: Dr. Sebastien Roujol , King’s College London
2nd Supervisor: Prof. Amedeo Chiribiri , King’s College London
Clinical Supervisor: Prof. Amedeo Chiribiri , King’s College London

Aim of the PhD Project:

  • To develop novel magnetic resonance imaging (MRI) techniques for myocardial tissue characterisation for newly introduced low magnetic field MRI scanners for optimal accuracy, precision, and reproducibility. 
  • To automate the workflow of these techniques (scan prescription, image acquisition/reconstruction, and post-processing/analysis) using AI. 
  • To evaluate the proposed methods in phantom/healthy volunteers/patients

Project description/background:

Cardiac magnetic resonance imaging (MRI) is commonly used for the clinical management of patients with cardiovascular diseases. MRI relaxation time parameters including T1, T2, and T2* are important markers exploited for the diagnostic of a variety of cardiac conditions such as myocarditis, amyloidosis, diffuse fibrosis, inflammation/oedema, acute and chronic myocardial infarction. The variations in T1/T2/T2* times can be exploited to generate images with a variety of MRI contrast. However, the relative nature of the MRI signal makes the interpretation of such images very subjective. Quantitative estimation of these parameters enables to remove this subjectivity in the assessment process by generating quantitative maps where each pixel represents a number in ms for each of these parameters. Conventional clinical techniques enable the acquisition of one 2D parametric map in one ~12s breathhold. Therefore, the acquisition of all parametric maps with sufficient spatial coverage of the heart requires a large number of breathhold acquisition resulting in prolonged protocols, increasing patient discomfort, image artifacts, and cost. Despite these limitations, these techniques are now available in most commercially available MRI scanners and used routinely in the clinic. 

Unfortunately, cardiac MRI examinations are commonly performed using high magnetic field MRI scanners (1.5T/3T) which are expensive, which has been an important limiting factor to their worldwide spread. Affordable low magnetic field (0.55T) MRI scanners have been recently developed as an alternative to break such barriers.  

The aim of this project will be to develop the next generation of time efficient quantitative cardiac MRI acquisition techniques for low field MRI scanners and to automate quantitative map analysis using AI to reduce operator dependency, analysis time, and cost. 

The ideal candidate should have a strong interest in biomedical engineering, medical physics, computer science and related fields. Novel MRI acquisition/reconstruction techniques as well as new AI approaches will be developed in a truly interdisciplinary project to build a unique broad set of skills, currently in need in both the academic and industrial job market.  

Please refer to the caption.

Figure 1. Example of myocardial T1 maps acquired using high field MRI scanners using our recently developed accelerated FAST1 approach (left) and conventional approach (right). A five-time acceleration rate was obtained without compromising accuracy and reproducibility of T1 maps.

 

Back to Projects