Smart Medical Imaging

EPSRC Centre for Doctoral Training


2019_011 - Non-invasive early detection of cancer metastasis in lymph nodes using 3D ultrafast and super-resolution ultrasound

1st supervisor: Mengxing Tang
2nd supervisor: Robert Eckersley

Approximately 90% of all cancer deaths occur because of the secondary tumours that result from metastases that have travelled via lymphatics. Early detection of signs of metastasis in lymph nodes (LNs) is of crucial importance for patient management and prognosis. Currently there is no effective non-invasive imaging technologies for this purpose, due to limitations of image resolution and/or contrast and accessibility of the existing modalities including MRI and CT. Clinically the lack of such non-invasive staging capability often causes sub-optimal treatment (over or under treatment) and cause patient suffering and significant waste of resources.

To give an example, at present patients with breast cancer are advised to have some or all of their lymph nodes (LN) removed from the armpit together with the cancer in the breast. This is because cancer cells can spread through tissue fluid channels (lymphatics) and these channels then pass through LN where the cancer cells can cluster and grow. However in countries like the UK, around 70% of patients with early breast cancer will not have cancer deposits in their LN. Due to the lack of an effective imaging tool to non-invasively inspect LNs, many patients are subjected to an expensive surgical procedure that does not benefit them and may lead to long-term problems such as lymphoedema (arm swelling).

The challenges for imaging such LNs with existing imaging modalities are 1) LNs are small – typically less than 1cm in diameter; 2) LNs can be located at depths of up to several cm. 3) LNs have dual circulation systems, blood flow and lymph flow, where lymph flow is very slow and hard to detect/image. 4) metastasis as small as a few mms is still clinically significant and requires 3D high resolution and sensitivity imaging currently not achievable.

This project aims to develop imaging techniques to address this unmet clinical need, by taking advantage of four recent advances. Firstly ultrafast ultrasound, with up to tens of thousands of imaging frames per second, enables significant improvement in the information content of the acquisition and up to an order of magnitude improvement in signal to noise ratio. Secondly microbubble contrast agents increase the sensitivity of ultrasound in detecting blood flow by up to two orders of magnitude, and provide highly visible individual scatters. Thirdly, super-resolution ultrasound, detecting and localising moving individual microbubbles in e.g. blood vessels, can achieve a spatial resolution of down to 20 microns at centimetre depths in vivo. Finally, new ultrasound transducer technologies have made it possible to achieve fast 3D image acquisition.

We hypothesise that by taking advantage of these technological advances and being able to generate super-resolution images at multi-centimetre depth with both blood and lymph flow information in 3D, it is feasible to non-invasively detect lymph node metastasis in vivo.