Our projects

Researchers at the AUT Knowledge Engineering and Discovery Research Institute are involved in a number of high-projects. Learn more about our work and research expertise below.

Our projects

Computational neuro-genetic modelling

Developing new data science technology for mental health diagnosis as part of the MBIE Catalyst Programme

AI in tinnitus

Developing an AI decision tool for the prediction of patients' responses to tinnitus therapy

INTELLECTE

Developing novel information methods and applying them to innovative IIT for complex data modelling

PANTHER project

AUT was a partner of the PANTHER project which was awarded under Erasmus Mundus Action II

China tripartite research

Research collaboration projects between New Zealand and China

Colorectal cancer

Prediction of risk of anastomotic leak, which is a known complication after bowel resection surgery using the preoperative data

Tinnitus project

Visualising and quantifying differences in brain activity of tinnitus patients

Our areas of expertise

Current research projects

  • NeuCube - a novel Spatio-Temporal Machine (STM)
  • Neuromorphic Systems (with University of Manchester and ETH-Zurich)
  • Brain Computer Interface

A holistic framework based on NeuCube for multimodal brain data modelling

For the first time the NeuCube framework will be used to combine all types of spatiotemporal brain data (STBD) related to a given problem (EEG, fMRI, DTI, structural, genetic) to model and understand complex spatio-temporal relationships across the data sets. This is the main method on which the EU H2020 proposal submitted in 2014 is based, with the participation of 8 EU partners and KEDRI, KEDRI being the world leader on the topic. The results will be used in other INTELLECTE/2 projects.

Project team

  • Professor N. Kasabov
  • Dr M. Fiasche' (Polytechnic University of Milan, Italy)
  • Professor G. Indiveri (University of Zurich and ETH, Switzerland)
  • N. Sengupta
  • N. Scott
  • C. McNabb (University of Auckland)
  • E. Capecci

Algorithms for machine learning in spiking neural networks (SNN) and efficient implementation of the NeuCube framework on highly parallel neuromorphic platforms

The following new algorithms for machine learning with SNN and more specifically – for the NeuCube SNN computational framework will be developed: deep learning in a 3D SNN reservoir; time series prediction as a regression problem in NeuCube; rule extraction from a trained NeuCube 3D structure; NeuCube optimisation based on a quantum–inspired methods. The NeuCube framework will be implemented on two types of neuromorphic parallel platforms: SpiNNaker, with 200,000 processing units/neurons (with U Manchester); the INI/ ETH Zurich multicore system (with INI/ETH). The efficiency of the implementation in terms of time and accuracy will be evaluated and published. The results will be used in other INTELLECTE/2 projects.

Project team

  • Professor N. Kasabov
  • Professor S. Furber (University of Manchester, UK)
  • Dr. S. Davidson (University of Manchester, UK)
  • Professor G. Indiveri (University of Zurich and ETH, Switzerland)
  • N. Scott

A neurogenetic model for the analysis and prognosis of AD data

This project will use the methods from some other projects to integrate into a NeuCube model multimodal data such as EEG, fMRI and genetic data, all related to Alzheimer Disease (AD), with the hypothesis that the model can predict early onset of AD and also that new patterns of brain development in relation to AD can be discovered. Data for the project has been obtained from Italy.

Project team

  • Professor N. Kasabov
  • Professor F. C. Morabito (Mediterranea University of Reggio Calabria, Italy)
  • E. Capecci

Current research projects

  • Novel SNN Methods for Neurorehabilitation
  • Integrated Brain Data Modelling (with Humboldt University, Germany)
  • Human Motion Modelling
  • Predicting Response to Treatment (for example addiction, schizophrenia)

A new design of a brain controlled neurorehabilitaion exoskeleton

The main objective will be to develop a working prototype of a BCI device that will link the BCI with a functional electrical stimulation system. A secondary objective will be to look at the feasibility of using NeuCube to map neural activity during task performance. This will produce a powerful tool that will advance the understanding of cortical activation during task performance. Successful models will be used to design and implement a new exoskeleton robotic system with Professor Z.Hou as part of the China Strategic Alliance Agreement.

Project team

  • Professor N. Kasabov
  • Professor Z.-G. Hou (Chinese Academy of Sciences, China)
  • Associat5e Professor D. Taylor
  • Dr D. Shepherd
  • Dr H. Gaeta
  • Professor R. Jones
  • Dr S. Weddell
  • Dr I. Khan
  • J. Chamberlain
  • N. Scott
  • N. Sengupta
  • M. Gholami
  • E. Capecci

Brain Data Networks: methods and systems for computational modelling and analysis of multimodal, multiscale, spatio-temporal brain data

The human brain can be viewed as a dynamic, evolving information-processing system, probably the most complex one. Processing and analysis of information recorded from brain and nervous system activity, and modeling of perception, brain functions, and cognitive processes, aim at understanding the brain and creating brain-like intelligent systems. The main goal of this project is to bundle the diverse knowledge, know-how and technologies of scientists in Germany (Humboldt University) and New Zealand (KEDRI AUT) to create a new methodology and systems for efficient analysis and understanding of complex brain data, through synergistic research, using different data modalities. We plan to combine methods including neuroimaging techniques, novel signal processing methods and complex networks theories within the unifying computational framework called ‘NeuCube’.

Project team

  • Professor N. Kasabov
  • Associate Professor D. Taylor
  • Dr G. Wang
  • Dr S. Marks
  • Professor G. Ivanova (Humboldt University of Berlin, Germany)
  • Professor J. Kurhs (Humboldt University of Berlin, Germany)
  • Professor H.C. Hege (Humboldt University of Berlin, Germany)

Motion data analysis technology based on SNN

This project will continue testing if a NeuCube-based approach will be suitable for modelling and understanding of human motion data. Data has been collected in the GATE lab of the FHES. If successful, results can be expected to be used after 2015 in clinical practice and for sport performance evaluation at AUT.

Project team

  • Professor N. Kasabov
  • Associate Professor D. Taylor
  • Professor P. McNair
  • Y. Naude
  • N. Signal
  • N. Scott
  • E. Capecci

Predicting response to treatment of patients with schizophrenia

This project will apply the multimodal brain data integration method used in the project for "A holistic Framework Based on NeuCube for Multimodal Brain Data Modelling" to develop for the first time a NeuCube-based model for predicting the response to clozapine of patients with schizophrenia where full brain data for each patient will be included: EEG, fMRI, structural, DTI, genetic. The project will also continue previous work on developing a NeuCube-based method for more accurate prediction of response to methadone treatment of patients who are under opiates. Now the objective is to investigate for the first time ex-opiate users’ functional recovery on emotion processing and working memory associated with methadone treatment based on EEG data. Test data has already been collected.

Project team

  • Professor N. Kasabov
  • Associate Professor D. Taylor
  • Dr G. Wang
  • Professor R. Kydd (The University of Auckland)
  • Dr B. Russel (The University of Auckland)
  • C. McNabb (The University of Auckland)
  • N. Sengupta
  • M. Gholami
  • E. Capecci

Current research projects

  • Data Analytics for Bioinformatics
  • Personalised Stroke Occurrence Prediction (with NISAN and CAS-China)
  • Personalised Risk of CVD Prediction

Predictive personalised modelling for modelling anaesthetic procedures

The project will continue to utilise a hybrid knowledge-engineering and model-based simulation of patient physiology for early identification of issues in patients under anaesthesia/intensive monitoring. The project aims to develop a decision support system based on NeuCube particularly for haemodynamic management (such as Massive Transfusion Protocol).

Project team

  • Professor N. Kasabov
  • Dr A. Lowe
  • G. Hamano

A web-based system for personalised prediction of stroke occurrence based on climate and personal data

This project will implement as an application web-based system the method for personalised modelling and prediction of stroke occurrence from personal and spatio-temporal climate data. Both an NZ version and a version for China in the Chinese language will be developed.

Project team

  • Professor N. Kasabov
  • Professor V. Feigin
  • Professor Z.-G. Hou (Chinese Academy of Sciences, China)
  • R. Krishnamurthi
  • M. Othman

A holistic framework for personalised predictive modelling of cardiac disorders

This project will continue to explore and utilise the use of internationally available multimodal data for patients with heritable cardiac disorders. The data comprises hundreds of patients with clinical as well as extensive single nucleotide polymorphisms (SNP) data. The results achieved in INTELLECTE will form the foundation of a new model that will take a more holistic approach in that there will be more clinical data to explore.

Project team

  • Professor N. Kasabov
  • Associate Professor C. Higgins
  • Dr D. Love (LabPlus, Auckland City Hospital)
  • Associate Professor D. Parry
  • V. Breen

Current research projects

  • Ecological Data Modelling and Event Prediction e.g.  establishment of harmful species (with SJTU, XU-China and Lincoln Uni.)
  • A holistic approach to Integrating multiple data sets through imputation

Predictive modelling of harmful species establishment from ecological data

This project will continue using SNN NeuCube-based approaches for a more accurate prediction of ecological events, but it will also develop new rule extraction algorithms for a better understanding of the modelled data and the ecological processes. NZ data has already been initially experimented and data from China will be trialled next as part of a tripartite project West China-East China–New Zealand.

Project Team

  • Prof. N. Kasabov
  • A/Prof. R. Pears
  • A/Prof. S. Worner (Lincoln University)
  • Prof. J. Yang (Shanghai Jiao Tong University, China)
  • Prof. S. Ozawa (Kobe University, Japan)
  • R. Hartono

A holistic approach to Integrating multiple data sets through imputation

The project develops a novel method for the integration of multiple data sets with case studies in cancer and more specifically – bladder cancer.

Project Team

  • Prof. N. Kasabov
  • V. Breen
  • PEBL team

Current research projects

  • Moving Object Recognition Using DVS and NeuCube
  • Novel Methods for Radio Astronomy
  • Environmental Event Prediction (e.g. risk of earthquakes)

Methods for computer vision and image processing of dynamic objects

This project will develop a new method and an engineering system for fast video and image scene data analysis, including neuromorphic computation with dynamic vision sensors (DVS, also called artificial retina) and neuromorphic chips, both produced at INI/ETH Zurich.

Project Team

  • Prof. N. Kasabov
  • A/Prof. W. Yan
  • Prof. R. Klette (The University of Auckland)
  • A/Prof. D. Bailey (Massey University)
  • Prof. G. Indiveri (University of Zurich and ETH Zurich, Switzerland)
  • W. Cui
  • F. Alvi

Big data in radio-astronomy and pulsar recognition

This project will continue to look into the problem of complex pattern recognition from large and fast radio- astronomy data, particularly for the identification of dispersed transient and pulsar signatures, using evolving neuromorphic systems. We propose to apply techniques of neuromorphic systems, most particularly the NeuCube (Radioastronomy variant) architecture for spectro-temporal data modelling based on SNN. First experiments show promising results with very low computational cost. This is especially true when paired with a dedicated neuromorphic hardware platform, such as the SpiNNaker. With further development, the NeuCube has the potential to significantly improve the state-of-the-art in dispersed transient and pulsar search, and become a core technology in projects like the SKA.

Project Team

  • Prof. N. Kasabov
  • Prof. S.Gulyaev
  • A/Prof. B. Pfahringer (The University of Waikato)
  • Dr. G. Dobbie (The University of Auckland)
  • Prof. B. Stappers (The University of Manchester)
  • N. Scott
  • M. Mahmood

Predicting risk of earthquakes in New Zealand: dream or reality?

Predicting risk of earthquakes is a very challenging task, but our hypothesis is that with the SNN NeuCube framework it is possible to achieve it with a high accuracy subject to sufficient spatiotemporal data is available on-line. Preliminary experiments on small data from GeoNet showed promising results, but this project will develop a new model that treats for the first time the problem in its entirety, using all available fast and large stream data in an on-line mode (from the GeoNet site).

Project Team

  • Prof. N. Kasabov
  • A/Prof. R. Pears
  • R. Hartono

Current research projects

  • Data Visualisation
  • Multisensory Data Processing in Real Time
  • Modelling audio-visual stimulus perception related to language aspects

Intelligent data visualisation and human-computer interaction methods

In the context of visualisation of big-data and complex spatial and temporal datasets such as NeuCube, this project explores multi-user scenarios, virtual reality technologies for immersive data visualisation, advanced human-computer interaction technologies, e.g., haptic feedback, and advanced GPU based real-time rendering technologies (ray-tracing, radiosity).

Project Team

  • Prof. N. Kasabov
  • Dr. S. Marks
  • Prof. M. Billinghurst (University of Canterbury)
  • Prof. R. Jones (University of Canterbury)
  • Dr. S. Weddell (University of Canterbury)
  • Dr. M. Sagar
  • Dr. N. Scott
  • M. Gholami
  • J. Estevez

Integrative and collaborative methods for smart system design

Our team will continue to explore a new collaborative approach to distributed software/hardware system design Aura-based collaboration in a computer-mediated collaboration environment. Reducing cognition stress and avoiding interruptions are two primary research objectives of awareness systems for collaboration studies, which have not been achieved yet. The Aura-based collaboration comprises communication approaches for facilitating collaboration among distributed people, enabling people to dynamically manage their availability status without compromising their interpersonal privacy. Based on this concept, a delegation-like application will be developed to capture, understand and predict user’s availability status. The novelty of this study is to employ data mining and neuromorphic techniques.

Project Team

  • Prof. N. Kasabov
  • Prof. S. MacDonell
  • A/Prof. T. Clear
  • Dr. A. Connor
  • Prof. Z. Salcic (The University of Auckland)
  • Dr. A. Malik (The University of Auckland)
  • Prof. M. Winikoff (University of Otago)
  • A/Prof. D. Damian (University of Victoria, Canada)
  • D. Zhang

Modelling audio-visual stimulus perception related to language aspects

The hypothesis is that the NeuCube architecture can model not only brain data measured in certain areas of the brain, but also the perception of presented stimulus data such as speech and image and to relate this information to language concepts in multilingual subjects. A proof of concept method will be explored for the first time.

Project Team

  • Prof. N. Kasabov
  • Prof. S. Harvey
  • A. Wendt

NeuCube

NeuCube is the world-first development environment and a  computational architecture for the creation of Brain-Like Artificial Intelligence (BLAI), that includes applications across domain areas. It is based on the latest neural network models, called spiking neural networks (SNN).

Find out more

Publications

View the latest research outputs by researchers at the AUT Knowledge Engineering and Discovery Research Institute.

View all

Our research groups

Find out more about the different research groups at the AUT Knowledge Engineering and Discovery Research Institute.

Find out more