Innovations in Stroke Rehabilitation: Reconnecting and Rewiring the Brain and Limbs 

Prof. Dr. Raymond Kai-Yu Tong 



Neuroplasticity refers to the brain’s ability to form new connections throughout life, which is particularly evident in response to stroke. Brain lesions are a long-term consequence of stroke, resulting in deficits in language, motor, or cognition, depending on the lesion location. Through neuroplastic processes, the lesioned brain can reorganize its structure, function, and connections to adapt to such damage. Innovations in stroke rehabilitation may facilitate reconnect and rewire the brain and limbs. We investigate the characteristics of brain waves (EEG), brain activities (fMRI) and muscle activities (EMG) related to the paretic upper limb movements after stroke. Understanding how brain structure and function reorganize after stroke is crucial in managing stroke recovery. Then we developed interactive control strategies to control different rehabilitation training systems for hand training in clinical trials, such as functional electrical stimulation (FES) and rehabilitation robot (Hand of Hope). The system incorporated the EMG and EEG as the bio-parameters to indicate the voluntary effort from a subject.  We applied these engineering-based technologies in the field of Neurorehabilitation, robotic system to provide external assistive force during the rehabilitation training. The clinical studies showed functional improvement in the clinical outcome measures on the upper limb and lower limb after the exoskeleton robotic training for 20-session on stroke survivors, and changes have been observed in the resting-state functional MRI (rs-fMRI).



Biomedical Engineering Contributions in Covid-19 Times

Scientia Prof. Dr. Nigel Lovell


As a response to the increasing burden of chronic disease and the ageing population on health care expenditure, considerable focus has been placed on appropriate technologies for promoting self-care and for supporting ageing-in-place. Such technologies are even more critical in the face of emerging health threats such as the COVID-19 pandemic. A number of medical device technologies aimed at relieving the burden of disease and improving quality of life will be explored. These devices, developed at the Graduate School of Biomedical Engineering (GSBmE), UNSW over the past two decades include telehealth monitoring and decision support systems for chronic disease management; wearable ambulatory technologies based around triaxial accelerometry for estimating risks of falling and for automatically detecting falls; and a range of neural interface technologies for restoring and potentially augmenting sensory loss. The talk will also highlight the future of implantable, wearable and telehealth technologies in future models of patient care and health service delivery especially in the current global pandemic.



Silk-based Biomaterial for Tissue Regeneration

Prof. Dr. James C.H. Goh


Innovative approaches to different fields within biomedical engineering and life sciences have largely been biologically inspired. This is especially so in the field of tissue engineering and regenerative medicine, whereby researchers have looked upon nature for inspiration in strategies and design parameters for scaffold materials and architectures for specific tissues. Biopolymers have been largely studied and silk fibroin has shown to be an excellent example due to its unique molecular and supra-molecular structure, its customizable ligands-based bioactivity, its ability to self assembles and its ability to be manipulated into various forms and structures. There exists an array of techniques to process silk fibroin into various forms with tailored mechanical and biological properties, to provide the necessary cellular, architectural, and chemical cues for the specific tissue types. The material can be processed into powders, films, gels, sponges, foams, yarns, knitted, woven mats as well as 3D printable bio-inks for various interesting tissue engineering applications. Numerous research have investigated applying the material in regeneration of tissues such as bone, cartilage, tendon & ligament, intervertebral discs, skin and cardiovascular tissues. However, limitations persist in its widespread use due to source-based variations and lack in standardization of processing protocols.



Application of machine learning methods for automated detection of ASD

Prof. Dr. Rajendra U Acharya


Autism Spectrum Disorder (ASD) is a neurobiological disorder that affects children’s behavior, social interaction, and communication. This is due to the abnormal brain wiring in autistic individuals which causes a reduction in learning rate and language impairment. ASD is influenced by a variety of factors from genetic to environmental such as exposure to radiation during pregnancy. Globally, 1 in 160 children is expected to have ASD, and boys are generally five times more likely to suffer from ASD as the mutation causing autism is found in the X chromosome. Hence, the mutation became dominant in males who possess XY chromosomes, and recessive in females if the mutation exists in only one of the XX chromosomes.

Fortunately, learning therapies are available to enhance neural connectivity and improve social communication. Learning therapies are extremely important for ASD children who are entering the adolescence stage as hormonal changes during puberty have a direct effect on brain development, thereby enhancing their speech integration ability. Thus, early diagnosis and intervention are crucial for ASD children to improve their quality of life and facilitate faster integration into society.

Currently, diagnosis of ASD relies on clinical assessment to determine if children possess ASD symptoms such as abnormal behavior and lack of interactive skills. However, the heterogeneous nature of ASD increases the difficulty for early diagnosis of ASD children, resulting in delayed treatment. Therefore, this warrants the need for a much more sensitive and reliable diagnostic tool that relies on biomarkers instead of clinical features. Electroencephalograms (EEG) which reflect the activity of the brain serves as a potential biomarker for ASD diagnosis. As such, researchers have explored a variety of machine learning models to detect ASD patients using EEG signals. These machine learning models extracted salient features from the EEG recordings to train their classifier to recognize abnormal EEG patterns pertaining to ASD. Many machine learning studies have yielded good performance for automated ASD detection based on EEG, hence reflecting its potential as a powerful diagnostic tool that can be used as an adjunct tool by the clinicians to confirm their manual screening.