You are here:

Ben Sinclair

Dr Ben Sinclair is a neuroscientist with a background in physics. He has worked in astrophysics, characterizing the magnetic field of Saturn and the matter/energy makeup of the universe, and in particle physics at CERN, researching the theory of supersymmetry. His current research is split into two areas. The first is using physical exercise as a treatment for brain diseases. After finding that an unsupervised moderate intensity exercise regime did little to help patients at risk of Alzheimer’s Disease, Dr Sinclair is now looking at more vigorous and assisted exercise treatments, teaming up with the Isodynamics Company to test their Reviver device. The second avenue of research is using artificial intelligence in brain imaging. After finding that success in epilepsy surgery is associated with PET image characteristics, he is now using these images in AI algorithms to see whether we can use these characteristics to decide who we should send to surgery. Dr Sinclair is a Post-Doctoral Fellow in the following research groups in the Department of Neuroscience:

Find out more about Dr Sinclair's research and services.

Projects

Artificial Intelligence to Improve Surgical Outcomes In Epilepsy

PET, MRI

Around 39 million people worldwide suffer from epilepsy. This disease has profound physical, psychological and social consequences. For around 30% of epilepsy patients, medication is unable to control the seizures and for these patients, the most common treatment is brain surgery to remove the damaged tissue. This procedure is effective and leads to an improved quality of life, and a meaningful reduction in seizures for approximately 70% of patients. However, there can be serious side effects, such as reductions in memory, language deficits and motor complications. Although patients are thoroughly examined prior to surgery to assess their suitability for the procedure, for around 30% of patients do not obtain reductions in seizure frequency.

The decision to undergo brain surgery is a difficult decision for patients suffering from epilepsy. We want to help patients and their doctors have more certainty around this decision, and improve outcomes for people suffering with epilepsy.

We are using PET and MRI brain imaging from patients to try to learn what patterns in the images predict good surgical outcome, and use this as a tool for helping neurologists and patients decide whether to proceed to surgery.

We recently discovered that the presence of abnormalities observed in PET images on the opposite side of the brain from the tissue to be resected was associated with poor surgical outcomes, but only for patients whose epliepsy orginates on the right side of the brain!

We suspect that there are other patterns in the images presently unknown which could predict whether the surgery will be successful. We are attempting to find these patterns using machine learning.

At present we have collected and screened data of 145 epilepsy patients who had surgery and underwent PET scans prior to surgery, with about 250 more yet to be screened. This is the largest known data set of this kind.

So far data from a subset of 82 patients has been analysed. Summary measures from the PET and MRI images have been extracted and used to predict surgical outcome with an efficiency of 0.75(±0.09) using support vector machines. This analysis was limited to patients who had MRI scans before and after surgery, which substantially reduced our sample size. An additional limitation was that the full images were not input into the machine learning algorithm, only summary measures of the images.

We intend to use AI to discover features on MRI and PET scans predictive of good surgical outcomes (i.e. reduction in seizure frequency). We will use unsupervised generative deep learning methods to encode PET and MRI images to and from a latent space. We will predict surgical outcomes from the latent encodings, and use the generative model to answer – “How would this patient’s MRI or PET have to change in order for them to have a good surgical outcome?” By comparing synthetic images to the original images, we can identify features predictive of good surgical outcomes. We refer to this method as the creation of Generative Visual Rationales.

Without AI, we can only make tests with a priori hypothesis and prior domain knowledge. With AI we automatically extract image features without having to define these manually, potentially identifying features that we have not thought of or identified before.

REVIVER

Watch video (0.38min): Australian swimming legend Ian Thorpe visits the REVIVER Centre

Physical activity is critical for the maintenance of good physical health, and the evidence is strong that physical activity promotes neurological health [1]. However, 55% of Australians do not meet government guidelines for amount of physical activity [2]. The elderly, those with disabilities, and those with neurological disorders face additional barriers to exercise [3], for example those with Parkinson’s disease (PD) achieve ~30% less physical exercise than their healthy peers [4, 5]. So while physical activity helps prevent cardiovascular events, diabetes mellitus, and osteoporosis [6], suppresses depression symptoms and cognitive decline [7, 8], those with neurological conditions rarely reap these benefits.

For those with neurodegenerative diseases such as Parkinson’s Disease, exercise is an innovative approach to managing disease with research already showing it may directly alter the neurodegenerative process [9], have a positive effect on cognitive performance both acute and accumulatively [10], and modulate cortico-spinal pathways commonly associated with motor dysfunction [11-13]. Given these protective and preventative aspects of physical activity, any strategy or technology which increases access to exercise, or provides exercise in a more effective and efficient manner, will have profound preventative and public health implications. In this proposal we focus on Parkinsonism as an exemplar of a neurodegenerative condition, ensuring that findings are easily interpretable and not confounded by the use of an overly heterogenous patient population. However, the findings will be relevant to any condition where exercise ameliorates physical decline, but there are barriers to participation, e.g. Alzheimer’s Disease (342,000 in Australia), Multiple Sclerosis (25,600) and ageing (3.8 million over 65).

REVIVER

Parkinson’s Disease effects over 80,000 people in Australia, presenting with impairments to gait and balance and an increased risk of falls and debility. Atypical Parkinsonism shares the motor deficits of Parkinson’s Disease, but is unresponsive to Parkinson’s Disease treatments, including Levodopa, in fact no clinical trial has yielded a positive treatment result for Atypical Parkinsonism. Physical activity is routinely prescribed as an intervention to treat the motor symptoms of Parkinson’s Disease, and clinical trials support its efficacy in improving these symptoms [14-17]. Physical activity could therefore offer similar benefits to patients with Atypical Parkinsonism. However, exercise interventions that increase physical activity are often time consuming, difficult and unmotivating. This makes the development of high volume, intensive rehabilitation strategies that are both accessible and engaging an important challenge [5, 18].

To address this challenge, we are running a trial of an exercise assistance device which provides a new method of strengthening and balance training, allowing people to exercise far beyond their unassisted capacity, including those who are so debilitated that they are no longer able to engage in any form of exercise. The Reviver™ (Fig 1), has been developed in Australia by the Isodynamics Company. The device works by placing the user at a tilted angle and rotating them around (video demonstration: [19]). This form of motion requires minimal dynamic movement, and with no risk of falling it is widely accessible, regardless of state of disability This exercise assistance machine has been proven tolerable by paraplegics, those with motor-neuron disease, late stage PD patients and wheelchair-bound multiple-sclerosis patients. While the Reviver builds on concepts from machine-aided physical therapies, the use of gravity and reflex movements to facilitate exercise in impaired populations is a novel innovation, providing the primary point of difference between this and previous research on traditional exercises provided in rehabilitation.

1. Hillman, C.H., et al., Nat Rev Neurosci, 2008. 9(1): p. 58-65.
2. Health, A.I.o., et al., Insufficient physical activity. 2019, AIHW: Canberra.
3. Jaarsma, E.A., et al., Scandinavian journal of medicine & science in sports, 2014. 24(6): p. 871-881.
4. Toth, M.J., et al., Neurology, 1997. 48(1): p. 88-91.
5. van Nimwegen, M., et al., J Neurol, 2011. 258(12): p. 2214-21.
6. Warburton, D.E., et al., CMAJ, 2006. 174(6): p. 801-9.
7. Blumenthal, J.A., et al., Arch Intern Med, 1999. 159(19): p. 2349-56.
8. Lautenschlager, N.T., et al., JAMA, 2008. 300(9): p. 1027-37.
9. Petzinger, G.M., et al., J Neurosci, 2007. 27(20): p. 5291-300.
10. Chang, Y.-K., et al., Brain research, 2012. 1453: p. 87-101.
11. Cantone, M., et al., Clinical Neurophysiology, 2014. 125(8): p. 1509-1532.
12. Teo, W.-P., et al., Clinical Neurophysiology, 2014. 125(3): p. 562-568.
13. Ternes, A.-M., et al., Journal of clinical and experimental neuropsychology, 2019. 41(3): p. 320-329.
14. Tomlinson, C.L., et al., BMJ, 2012. 345: p. e5004.
15. Shu, H.F., et al., PLoS One, 2014. 9(7): p. e100503.
16. Chung, C.L., et al., Clin Rehabil, 2016. 30(1): p. 11-23.
17. Allen, N.E., et al., Mov Disord, 2011. 26(9): p. 1605-15.
18. Abbruzzese, G., et al., Parkinsonism Relat Disord, 2016. 22 Suppl 1: p. S60-4.
19. Reviver-Demonstration:, https://www.youtube.com/watch?v=JvpVNvvDYd0.

Video

Watch video (0.38min): Australian swimming legend Ian Thorpe visits the REVIVER Centre