Dr Tatiana Kameneva

PhD candidate and honours supervision

Higher degrees by research

Accredited to supervise Masters & Doctoral students as Principal Supervisor.

PhD topics and outlines

Deep reinforcement learning for human-centric guidance in AR/VR: A branch of machine learning called deep reinforcement learning has recently outperformed human experts on tasks such as playing Atari games and the traditional Chinese game Go. In this project, we seek a student to research and develop a deep reinforcement-learning framework for learning task-specific visual enhancement filters, and from this, novel scene visualisations to guide human activity. 

Functional role of beta band frequency oscillations in humans: The field of brain-machine interfaces rapidly grows. New neural decoding algorithms are proposed to control a robotic arm or a wheel chair. Recorded power in beta oscillations (10-45 Hz)  may be used to detect the patient’s attention and readiness to make a movement; therefore, enhancing the existing decoding algorithms.  The project would suit someone with an interest in signal processing. 

Gender and stature identification using image processing and machine learning algorithms: Footprints are unique characteristics of a person and differ in shape, pattern, margin,  and toe marks. In this project, the student will collect data and then develop a machine learning algorithm for automatic identification of gender, age, height and weight of a person from their footprints.

Investigation of the effects of electrical and light stimulation on neural response: Electrical stimulation has been used to restore sensory functions in people who lost their vision or hearing.   A novel way to stimulate neurons is to combine conventional electrical stimulation with targeted optical stimulation.  The aim of this project is to explore the effects of electrical and light stimulation on neural responses in experiments and in computer simulations. 

Nanoparticle enhanced infrared neural stimulation of retinal ganglion cells: In this project, the student will  study  experimentally how retinal ganglion cells respond to combined optical and electrical stimulation, when in vitro bath is enriched with gold nanoparticles. 

Seizure forecasting: Electroencephalography (EEG) is often used to predict a seizure, with varying success between participants. There is an increasing interest to use non-EEG body signals, including electrocardiogram (ECG) to help with seizures detection and prediction.  The aim of this project is to use advanced signal processing to forecast seizures from EEG and ECG data recorded by Seer Medical. 

Honours

Available to supervise honours students.

Honours topics and outlines

Please see PhD topics above: Most PhD topics  above can be adapted to suit a Honours student project.