Welcome to AI4PAN, The Artificial Intelligence for Pandemics group centered at
The University of Queensland (UQ).
The group's focus is the application of data science, machine learning, statistical learning,
applied mathematics, computation, and other "artificial intelligence" techniques for managing
pandemics both at the epidemic and clinical level.
The host response as an alternative early diagnostic for viral infection
by Meagan Carney and Kirsty Short
Click to expand
Since the initial outbreak of Coronavirus Disease 2019 (COVID-19) in Wuhan, China, over 215 million subsequent cases
have been recorded. The causative agent, Severe Acute Respiratory Coronavirus 2 (SARS-CoV-2) is highly transmissible
and characterized by a heterogenous respiratory disease that ranges from mild symptoms to acute respiratory distress
syndrome and death. Over 4.5 million deaths have been recorded to date.
Direct detection of viral RNA via quantitative polymerase chain reaction (qPCR) is a highly sensitive and specific
diagnostic tool for SARS-CoV-2 and the current gold standard for testing. However, whilst highly sensitive, a certain
threshold of viral RNA must be present for subsequent amplification and detection by qPCR. Accordingly, it is possible
for a close contact of a SARS-CoV-2 positive individual to initially test negative for the virus but then later in the
incubation period, when there is increased viral replication, to test positive. As a result, current public health
guidelines in Australia require close contacts of a SARS-CoV-2 positive case, as well as passengers from overseas, to
quarantine for the entirety of the viral incubation period (14 days). The time taken for an infected individual to be
identified by qPCR also has implications for anti-viral therapeutics. Current monoclonal antibody therapeutics are most
efficacious if given early in infection. Thus, the importance of identifying SARS-CoV-2 positive individuals as early
as possible early in infection is appreciable. Here, we hypothesise that there is a gene signature in the nasopharynx
that can be detected in SARS-CoV-2 positive individuals prior to the detection of viral RNA using qPCR. To test this
hypothesis we will use a combination of clinical samples and unsupervised machine learning clustering methods to
potentially develop a new, early diagnostic for SARS-CoV-2 infection.
Large scale COVID19 vaccination behaviour monitoring on social media
by Ajay Hemanth Sampath Kumar, Aminath Shausan and Afshin (Ash) Rahimi
Bounding the extinction time of a spatial epidemic
by Ross McVinish and Xiao-Yu Anita Lin
Endothelial cells are not productively infected by SARS-CoV-2
by Lilian Schimmel, Keng Yih Chew, Claudia J Stocks, Teodor E Yordanov, Patricia Essebier,
Arutha Kulasinghe, James Monkman, Anna Flavia Ribeiro dos Santos Miggiolaro, Caroline Cooper,
Lucia de Noronha, Kate Schroder, Anne Karine Lagendijk, Larisa I Labzin, Kirsty R Short and Emma J Gordon.
The host response as an alternative early diagnostic for viral infection by Meagan Carney
and Kirsty Short
Dr Meagan Carney's
research interests include probability and statistics of extremes in chaotic systems,
rare event analysis, dynamical systems, unsupervised learning algorithms, and
applications of importance sampling methods.
A/Prof. Marcus Gallagher's
expertise is in Artificial Intelligence, Optimization and Machine Learning
algorithms, including the theory, development and practical applications of these
techniques. He collaborated with Queensland Health in the area of anomaly detection
in drug prescription data.
A/Prof. Cecilia González Tokman's
expertise is in Dynamical Systems, Ergodic Theory and related areas.
Her recent work focuses on random dynamical systems, transfer operators, Lyapunov exponents
and coherent structures.
Dr. Hamid Khataee's
expertise is in Mathematical and computational modelling, Applied mathematics, Theoretical Biology
including computational modelling of morphological dynamics of cell populations, mechanics of
molecular motors, and quantification and modelling of epidemic data.
Prof. Geoff McLachlan's
expertise is in statistics in the related fields of classification, cluster and discriminant
analyses, image analysis, machine learning, neural networks, and pattern recognition, and in
the field of statistical inference.
A/Prof. Yoni Nazarathy's (coordinator)
expertise is in Machine Learning, Applied Probability,
Statistics, Operations Research, Simulation, Scientific Computing, Control Theory,
Queueing Theory, Scheduling, and Mathematical Education. His involvement with AI4PAN
is through the Safe Blues project.
Dr. Zoltan Neufeld's
expertise is in Mathematical and computational modelling, Applied mathematics, Mathematical Biology
including analysis of epidemic models, computational models of multicellular tissue biology,
collective cell motility, mechano-biology, pattern formation and tissue development.
Dr Hien Duy Nguyen's
main research focus is to explore the relationships between regression data and
mixture models, and to leverage such relationships to better analyse data that arise
from bioinformatics, economics, image analysis, neuroimaging, and proteomics data.
Interests also in the construction of expectation-maximization and
minorization-maximization algorithms for applications in nonstandard mixture
Dr. Ash Rahimi's research interests fall
within the fields of Natural Language Processing, Social Network Analysis and Machine
Learning. I am specifically interested in exploiting both structured and unstructured
data to help machines understand conversational language in Emergency Situations
and Health Informatics.
Dr. Aminath Shausan's
expertise focuses on statistical and probability modelling and analysis of epidemics.
Expertise in Bayesian statistical analysis of dengue modelling, Artificial intelligence
to project to understand pandemics.
Dr. Kirsty Short
is an influenza virologist by training with extensive experience in emerging viral
pathogens and pandemic preparedness. With expertise working on SARS-CoV-2 and
in particular the role of children in disease spread, the impact of this disease
on people with diabetes and the development of new antiviral therapies.
Dr. Sally Shrapnel
is an interdisciplinary scientist working at the interface of causality and machine
learning. She has both a clinical and technical background, with bachelor's degrees
in Medical Science, Medicine and Surgery, a Fellowship of the Royal Australian
College of General Practitioners, a Master's degree in Bioengineering (Imperial College, London)
and a PhD in physics and philosophy—on the topic of quantum causal machine learning.
Dr. Ian Wood's
research expertise covers classification, bioinformatics, stochastic optimisation,
machine learning and mixture models.
March 2021: The Australia Business Review, Artificial intelligence to help pinpoint COVID diseases. Sally Shrapnel,
Data scientists will use artificial intelligence to identify which COVID patients will likely experience
longer-term conditions such as kidney damage.