Back to Projects

AI-enabled Imaging, Emerging Imaging

Engineering whole-body 7 Tesla MRI coil for signal enhancement in early cancer detection

Project ID: 2020_036

1st Supervisor: Özlem Ipek, King’s College London
2nd Supervisor: Shaihan Malik, King’s College London
Clinical Champion: Vicky Goh, King’s College London

Aim of the PhD Porject:

  • Enabling high spatial resolution and big field of view imaging of the human body at 7T MRI.
  • Developing a novel decoupling circuitry to combine a novel parallel transmit RF coil array with multi-channel phased array receivers to accelerate MR image acquisition with maximum MRI signal efficiency.
  • implementing patient-specific tissue heating monitoring during MRI scan to realize a full-implementation of a clinic parallel-transmit body MRI.

Project Description / Background:

Cancer is a major disease burden that affects humanity and causes considerable worldwide mortality and morbidity. MRI scanning could offer a non-invasive early diagnosis of prostate, renal, liver and pancreatic cancer in the future. Ultra-high field MRI (7 Tesla) may drive as an early-diagnosis tool by realizing its advantages for high resolution anatomic and functional imaging of a human body.

Currently whole-body MR imaging can only be performed on 1.5 and 3 Tesla (T) clinical MRI scanners, and their capabilities of early detection of lesions and tumours are limited. 7T MRI is the state-of-the-art clinical MRI scanner for brain and extremities imaging but it is not clinical yet for whole-body imaging. Although 7T MRI offers increased signal-to-noise ratio with increased image contrast, imaging a big field-of-view as whole-body is quite challenging because of the higher resonance frequency (300MHz) at this field strength. The effect of requiring higher frequency radio-frequency (RF) fields is to both increase tissue heating, leading to safety issues, and to reduce the achievable uniformity of the magnetic field used for excitation (B1+). The latter issue occurs because the RF wavelength becomes smaller than the subject, such that wave-propagation effects lead to spatially non-uniform field patterns with limited penetration. In addition, due to increased chemical shift and susceptibility effects at higher field strengths, a higher RF bandwidth is required, which leads to a need for a higher RF peak power. Thus, the main challenges for 7T body imaging are related to the RF fields and how they are generated.

The most likely solution to these challenges is design of novel more efficient RF transmitter coils, coupled with parallel transmission (pTx) in which arrays of coils are used to improve performance. This project will focus on the development of an RF coil array yielding high signal efficiency and decreased tissue heating levels.  The hardware design will be a transmit-only array coil, with a multi-channel phased receive-only array to maximize the image field-of-view and to increased accelerated image acquisition. Although there are some long-standing types of coil design used in MRI – for example the loop, or arrays of loops – the move to higher field has also seen an expansion in innovative design, including the introduction of dipole arrays [1]. We have pioneered new designs of this type, and see their value in reducing SAR per B1+ produced by the coil. In this project we will experiment further with the basic element design, and will also design novel decoupling circuitry to produce new efficient coil designs that can drive the next generation of UHF body MRI. 

The design of this hardware will exploit advances in computational simulation, deep learning and RF hardware prototyping methods. If achieved, this development will enable contrast agent-free high-resolution anatomical and functional body imaging at 7T MR. It will create a platform for further development of investigation of non-invasive markers of cancer pathology.

References:

[1] Ipek Ö et al. Characterization of transceiver surface element designs for 7 tesla magnetic resonance imaging of the prostate: radiative antenna and microstrip. Physics in Medicine & Biology 57 (2),  343 (2011).

Figure 1: Multi-channel parallel-transmit crossed-dipole antenna designs (dark blue) for abdomen, cardiac, pelvis and shoulder imaging combined with local multi-channel loop receive coil array (upper row).

Back to Projects