Ongoing Research Programs

1. A Longitudinal Study of Stroke Patients

The aim of the study is to examine memory and other cognitive functions. We hope to determine what factors predispose an individual to the development of such deficits and what impact these deficits have on the functioning of the individual. This study involves a number of assessments and laboratory investigations which will be carried out in 3 stages. (Contact Prof. P. Sachdev for details)

2. Transcranial Magnetic Stimulation (TMS):

In collaboration with the Mood Disorders Unit (Professor P Mitchell), the NPI is conducting research into TMS. Transcranial magnetic stimulation involves the treatment of certain mental illnesses using machines that generate magnetic fields in rapid pulses. Traditionally, neurologists have used these machines to investigate whether or not nerve pathways are intact. For example, magnetic stimulation of motor areas of the brain can be used to produce a muscular twitch - in the thumb for instance – and this provides useful diagnostic information at the flick of a switch. TMS has been shown to be a non-invasive technique, apparently free of side effects (such as memory impairment), and while it does not induce a seizure, it is capable of modifying the activity of specific brain areas.

The use of TMS as a treatment for psychiatric disturbance first came to attention when it was noticed that patients with Parkinson’s disease who were also depressed felt better after having TMS tests. Since then, a number of studies have been conducted examining the use of TMS to treat depression in a number of countries including the USA, Germany Israel, Austria and Spain. Studies have largely considered TMS as a treatment for Major Depression, which is not effectively treated by antidepressant medication.

At Prince of Wales Hospital (POWH), research has been conducted to ascertain the effectiveness of TMS in treating depression in patients resistant to pharmacological treatments. Like other experimental centres, research at POWH initially focussed on stimulating left frontal areas of the brain, which have been shown to be underactive in some depressed patients. However, there are also grounds to suppose that stimulation of frontal areas of the brain on both sides may be superior to single-sided stimulation. POWH is currently conducting the first such research investigating whether or not this is the case. Another arm of the TMS research is being used to examine whether psychomotor slowing seen in some patients who experience severe depression occurs because motor areas of the brain and their connected pathways are not working properly, or because the patients cannot ‘drive’ these motor areas properly for other reasons (e.g., voluntary factors like motivation may play a role).

So far, most of the excitement surrounding TMS is based upon potential rather than proven effectiveness. Positive results have however, been reported in Germany, Israel and the United States. The research at POWH is still at an experimental stage - and at present, it is too early to comment on the effectiveness of TMS in our studies. However, this research represents a new frontier of non-invasive brain modulation, perhaps with the potential to replace current procedures such as ECT, and will certainly contribute to our understanding of brain-behaviour relationships. (Contact Prof. P. Sachdev for details)

3. Brain Imaging In Alzheimer’s Disease

This study is using specialised brain scans to evaluate brain blood flow abnormalities in Alzheimer’s disease, and to examine how this vaires during a "memory stress test". It is hoped that this research will help our understanding of brain abnormalities in early Alzheimer’s disease.(Contact Dr. J. Trollor for details)

4. Vagus Nerve Stimulation for Resistant Depression

VNS, or the long-term, intermittent stimulation of the left Vagus Nerve by a commercially available device (the NeuroCybernetic Prosthesis) is an experimental treatment that has shown initial promise in studies in the USA. The NPI has received ethics approval to conduct a limited open study on this treatment, and we are seeking appropriate referrals from psychiatrists.

1. Current DSM-IV Major Depressive Episode of ≥ two years) duration and/or has had a history of recurrent MDEs (at least four lifetime MDEs including the current MDE).
   
2. Failure to respond to drug treatment, with at least two treatments from different treatment categories during the current episode of MDE, and failure to respond to psychotherapy or cognitive-behaviour therapy (as appropriate).
   
3. A score ≥ 20 on the 24-item Hamilton Rating Scale of Depression.
   
4. Ability to consent to a surgical procedure involving the implantation of a stimulator (resembling a pacemaker).

Please click HERE to download more information.

NB: Subject will have to pay for the device (about $10,000), but will not incur any other out-of-pocket expenses.

For referral, please contact;

1. 1. Prof P Sachdev (02-93823763) p.sachdev@unsw.edu.au or
   
2. 2. Dr Julian Trollor (02-93823755) j.trollor@unsw.edu.au or
   
3. 3. Dr Gin Malhi (02-93823719) g.malhi@unsw.edu.au
   

5. Neuroimage Processing and Analysis

Computational Neuroanatomy: Computational neuroanatomy is emerging as an exciting methodology to characterize shape and neuroanatomical configuration of different brains. It encompasses a triad of techniques. These techniques provide information about both global and local differences in brain shapes and sizes.

Topic 1: Automated brain extraction, segmentation and cortical surface generation using 3D MRI data.

Description: 3D segmentation of brain MRIs into white matter, grey matter and CSF (cerebrospinal fluid) is crucial to many aspects of brain research, such as: quantitative volumetric analysis, morphological analysis and visualization. The aim of this topic is to develop methods of automating these processes. We will develop new algorithms to eliminate/minimise labor-intensive work such as manual tracing and measuring etc. One of the obvious application or extension of this topic is the automatic analysis of cerebral atrophy and quantification of changes of cerebral volume in a longitudinal MRI study. We will use T1-weighted MRI data available at NPI, PoWH.

Topic 2: MRI data recovery by removing and correcting acquisition artifacts.

Description: An image artifact is any feature which appears in an image which is not present in the original imaged object. Artifacts could also distort an image, rendering an image difficult or impossible to analyze. Among many T1-weighted MRI data NPI has acquired so far, a considerable portion of the images has worse than expected quality such as the poor grey and white matter differentiation. Some novel algorithms need to be developed to remove/correct such artifacts so that these MRI scans can still be reliably used. The methods should also find a greater audience once successfully developed since poor g/w differentiation caused by field inhomogeneity or other problems is common in MRI acquisition. A knowledge based method using multispectral histogram is tentatively proposed at this stage since the FLAIR sequence image of the same subject is also available to assist the analysis.

Topic 3: Automated quantification of changes in ventricular volume.

Description:. There is growing interest in the quantification of brain volumes using 3D MRI. We have started in developing an automated method for quantification of ventricular volume change from baseline and follow-up MRIs. Although we are studying only the changes in ventricular volumes at this stage, we will further explore the possibility of developing automated quantification of whole brain/white matter/grey matter/CSF changes once we have made progress in segmentation. Besides, a reliable measure in the changes in ventricular volume also provides valuable insight into the longitudinal structural development of both normal and pathological brains.

Topic 4: Computerised measurement of white matter hyperintensities in FLAIR sequence MRI.

Description: Signal hyperintensities on FLAIR sequence MRI are seen commonly in the brains of elderly individuals. We plan to develop an automated computer algorithm to replace the labor intensive manual rating of signal hyperintensities on MRI. The result of computerised method will be compared with that of the manual rating.

Topic 5: Computation of CBV/CBF (Cerebral Blood Volume/Flow) from time-course perfusion MRI.

Description: Maps of cerebral perfusion can be generated by nuclear medicine imaging techniques based on the radioactive tracers such as SPECT and PET. More recently MRI has acquired similar capacities along with the development of new acquisition protocols such as EPI etc. As MR scanners are frequently available and impose no radioactivity on the subject, perfusion MRI can be combined with routine scanning and repeated on several sessions. The objective of this study is to explore the possibilities of generating CBV/CBF maps from perfusion MRI data sets with new algorithms. The methods and algorithms reported in the literature are based on parametric mapping of time-course data, which may suffer from the use of inaccurate mathematical and/or physiological models, in particular when the models are applied to pathological tissues.

Topic 6: Graphical display of perfusion MRI data: visualizing multidimensional space.

Description:. Visualization of multidimensional data is an important part of computational data analysis. Perfusion MRI is a 4D (3D spatially plus time-course) data set. With proper assumptions, computation and analysis, we can generate a great deal of useful information from such 4D data. These include a subject? regional CBV/CBF, MTT (Mean Transit Time) maps and magnetic field homogeneity map which is associated with the head anatomical structures etc. Visualization of such data set will be implemented systematically and efficiently.

Topic 7: Automated arterial input function (AIF) computation for MR perfusion analysis.

Description: During bolus passage brain tissue T2 increases proportional to the concentration of the contrast agent. Relative CBV can be calculated in respect with the signal attenuation. However, to calculate the flow, the AIF of the contrast has to be acquired. The aim of this study is to develop an automatic algorithm for identifying arterial pixels, MCA (Middle Cerebral Artery) in particular from the 4D time-course data.

(Contact Prof. Sachdev 02-93823724 or Dr. Wen 02-93723730)