Dr Jun Hua – Innovations in Functional Brain Imaging to Improve Neurosurgery

Dr Jun Hua, Associate Professor at the F. M. Kirby Research Center for Functional Brain Imaging at the Kennedy Krieger Institute and Johns Hopkins University, USA, leads a team focused on developing novel magnetic resonance imaging (MRI) technologies for imaging the structure and function of the brain. Recently, they have been pioneering the development of new MRI techniques that can be used to improve pre-surgical planning for neurological patients and optimise patient outcomes.

MRI scanning: Looking Inside the Body

Magnetic resonance imaging (MRI) is a powerful imaging technique that has revolutionised the diagnosis and treatment of neurological disease. MRI uses strong magnetic fields and radio waves to precisely image the structures inside the body. MRI scans can distinguish between different types of tissue in the body and are generally used to screen for or monitor soft tissue abnormalities, including tumours, soft tissue injuries, joint injuries, spinal injuries, or damage to organs.

Recent years have seen the introduction of functional magnetic resonance imaging, or functional MRI (fMRI), a technique that allows imaging of the activity of the brain through detecting changes in blood oxygenation and flow. fMRI scans can identify which parts of the brain are associated with different aspects of neurological function such as language, sensory information, and motor function.

Dr Jun Hua, Associate Professor at the F. M. Kirby Research Center for Functional Brain Imaging at the Kennedy Krieger Institute and the Department of Radiology at Johns Hopkins University, USA, leads a team focused on the development of novel MRI technologies for imaging the structure and function of the brain. His team works on a number of different MRI-based projects, including the development of novel MRI techniques for imaging the microvasculature and metabolism in the brain.

Current work by Dr Hua and the team is focused on understanding the physiological causes of functional MRI signals in healthy individuals and in the brains of patients with neurological conditions such as Huntington’s disease, Alzheimer’s disease, Parkinson’s disease, schizophrenia, and brain tumours. One of the problems that Dr Hua’s group has begun to address is how to provide a better imaging tool to aid surgeons during their pre-operative planning.

Blood Oxygen Level-dependent Imaging and the Brain

When surgeons operate to remove a brain tumour, they have to make careful decisions about what tissue to remove. Remove too little, and there is a chance that cancer cells could remain in the brain and regrow. Remove too much, and there is a risk of removing healthy brain tissue and causing loss of brain function in the patient.

Some types of brain tumour, such as glioma, are extremely invasive, and it is sometimes impossible to completely remove it entirely. A balance must be made in excising the maximum amount of tumour while limiting damage to the healthy tissue in order to best prolong and improve the quality of the patient’s life.

In order to make these decisions, surgeons are increasingly taking advantage of the information that can be obtained from functional MRI scans. Blood oxygen level-dependent (BOLD) functional magnetic resonance (fMRI) can be used to map the brain pre-operatively to identify which parts of the brain correspond to areas of critical function.

To determine which areas of the brain correspond to which critical functions, patients can be scanned while performing simple tasks that use different parts of the brain. For example, a task such as hand-squeezing will use the part of the patient’s brain associated with motor control, and activity in this area will be detectable on the fMRI scans.

However, this approach is not without challenges. Signal dropouts and distortions are well-known issues with current fMRI scanning methods and are a hurdle to using these imaging methods for assisting with surgery. The most often-used method, gradient-echo (GRE) echo planar imaging (EPI), is associated with well-described image distortions. These often occur at areas close to air cavities, including parts of the brain known as the orbitofrontal and temporal lobes. As these areas are associated with language and critical cognitive functions, they are critical areas that need clear imaging in pre-surgical mapping.

Distortions are particularly bad around cavities caused by previous surgical operations. This causes a problem for surgeons attempting to assess patients who have already had an operation to remove a brain tumour. The area around the initial surgical site is most likely to include tissue that needs excising, and most likely to need precise mapping to ensure that healthy tissue is not removed. Other malformations and surgical features, such as cranial implants, haemorrhage, arteriovenous malformations, and MR-compatible metal head implants can also cause disturbances in the fMRI image.

New and Non-Invasive

In a 2014 study, Dr Hua and his team demonstrated the utility of a new fMRI technique capable of solving the problem of artefacts in the MRI image. Their study described a newly developed approach, known as T2-prepared (T2prep) BOLD, and showed negligible distortion and signal dropout when used to image the whole brain.

Dr Hua and his group scanned the brains of volunteers undertaking simple tasks designed to use different parts of the brain. These were a visual stimulation task consisting of a blue/yellow flashing checkerboard projected onto the inside of the MRI scanner in front of the volunteer, and a motor task requiring the participant to tap one finger during the flashing period. The team found that their T2prep BOLD technique was able to accurately show areas of activity in the brain with minimal distortions and dropouts.

Having confirmed that they had a more accurate and reliable method for detecting brain activity, Dr Hua and his team then wanted to know if the technique could successfully be used as a pre-operative tool. They conducted a pilot clinical study in four patients with either a brain tumour or a brain lesion caused by epilepsy.

In this study, the group imaged patients using standard imaging techniques such as EPI, and their new T2prep BOLD imaging method. Signal dropouts and distortions were found in the EPI images because of the presence of blood products and air-filled cavities. However, T2prep BOLD imaging produced images with minimal imaging artefacts. The team’s method also showed equivalent structural detail to that of existing anatomical brain scans.

The patients also underwent tests of brain function. First, patients undertook a sentence completion test commonly used in pre-surgical language mapping. During the sentence completion test, the patients were scanned with the existing EPI method, and a T2prep BOLD scan. The team found that in a patient with a tumour very close to an important language area in the brain, activation of this language centre was not detected by EPI imaging whereas activity was visible using T2prep BOLD fMRI. This was also the case for a patient with an epilepsy lesion located in the same area.

These studies demonstrated that the new T2prep BOLD imaging method is capable of mapping brain function extremely close to existing lesions and tumours. This technique could, therefore, constitute a significant step forward in pre-surgery planning, allowing surgeons to determine healthy areas of tissue around brain tumours and then more accurately excise tumours while minimising healthy tissue loss.

Diffusion Tensor Imaging: Visualising Fibre Bundles in the Brain

Diffusion Tensor Imaging (DTI) is a non-invasive imaging method that can visualise the trajectories of fibre bundles in the brain. It provides critical information on the spatial relationship of the fibre tracts to the margins of resectable lesions, which allows neurosurgeons to plan the safest surgical trajectory for lesion resection without damaging eloquent fibre tracts in the vicinity. DTI is frequently used by neurosurgeons together with BOLD fMRI in presurgical brain mapping.

Recently, Dr Hua and his team extended their efforts to the development of novel DTI methods that can provide artefact free images in the presence of metal implants. In a recent study published in Radiology, one of the most prestigious journals in the field, they showed that T2prep BOLD fMRI and diffusion prepared DTI significantly reduced distortion and signal dropout throughout the brain compared to conventional EPI-based methods in healthy individuals wearing metallic orthodontic braces. Thus, diffusion prepared DTI can be applied to presurgical mapping in a way similar to T2prep fMRI, providing an alternative MRI method in the presence of strong susceptibility artefacts.

Next Steps

The current gold-standard method for mapping brain activity is known as electrical stimulation mapping, or ESM. This is performed during an operation to map brain function in the area around a brain lesion. However, this is a highly demanding and invasive technique. While this technique can accurately tell surgeons where language-related brain tissue is located, it requires a longer duration of surgery, runs the risk of inducing seizures, and must be performed in awake patients.

T2prep BOLD imaging technique can therefore not only improve the quality of non-invasive brain activity imaging, but also provide a viable, non-invasive alternative to existing methods. As ESM can also only be used on patients who are awake during their surgery, there is the risk of creating considerable distress, whereas T2prep BOLD can be used on awake or anaesthetised patients, allowing for a much less distressing scanning procedure.

After the success of their initial studies, Dr Hua and the team hope to test their new method in a larger cohort of patients. The next step to validating this technique for clinical use is to compare the T2prep BOLD scan results with those from gold-standard brain mapping techniques such as ESM.

The group believes that once their technique has been validated in a larger patient cohort, the feasibility and usefulness of its use in non-invasive pre-surgical mapping will be significant. This, in turn, will have a positive impact on patient health and surgical outcomes through aiding accurate surgical resections, and also negating the need for distressing and invasive mapping techniques such as ESM.

Finally, the team hopes that T2prep BOLD imaging could also be used for patients with other brain malformations, and plan to recruit patients with malformations that could cause imaging artefacts, including those with surgical cavities, cranial implants, tumours, arteriovenous malformations, internal calcification, and lesions resulting from other causes. If successful in clinical trials, the T2prep BOLD imaging technique could help transform the surgical care of a large number of patients across a wide range of neurological disorders.