Contact details

• Location: 64, IC3
  Campus: Albany

Qualifications

• Doctor of Philosophy in Molecular Medicine - University of Auckland (2019)

Certifications and Registrations

• Licence, Supervisor, Massey University

Research Interests

My research examines how biological, clinical and imaging data can be combined to improve prediction, stratification and decision-making in complex health settings.

In breast cancer, I am interested in how screening performance, breast density, tumour biology, treatment, geography and access to care interact across the patient pathway. This includes work on multimodality imaging, prognosis, treatment response, linked registry analysis and variation in outcomes across population groups in Aotearoa New Zealand.

In reproductive science, I study developmental timing and image-based phenotypes in embryos and sperm. Current work includes embryo morphokinetics, cohort-relative embryo ranking, automated developmental-stage recognition, sperm morphology and fertility prediction in human and animal systems.

A recurring question across these programmes is whether a model remains useful when moved beyond the dataset in which it was developed. I therefore focus on validation across clinics, imaging platforms and patient groups, and on identifying where model outputs add information beyond existing clinical assessment.

My wider interests include survival analysis, longitudinal modelling, clinical prediction, computer vision, linked health data, model calibration, health equity and the use of quantitative evidence in clinical and service-level decision-making.

Thematics

Health and Well-being, Future Food Systems

Area of Expertise

Field of research codes
Agricultural Biotechnology (100100):
Applied Statistics (010401):
Artificial Intelligence and Image Processing (080100):
Biostatistics (010402):
Cancer Diagnosis (111202):
Computer Vision (080104): Information And Computing Sciences (080000):
Mathematical Sciences (010000):
Medical And Health Sciences (110000):
Medical Biotechnology (100400):
Oncology and Carcinogenesis (111200): Public Health and Health Services (111700):
Statistics (010400):
Technology (100000)

Keywords

Applied biostatistics; biomedical machine learning; computer vision; longitudinal and survival analysis; medical imaging; clinical prediction models; breast cancer epidemiology; reproductive biology and IVF; image-based phenotyping; model validation and calibration; explainable AI; health equity; translational data science

Teaching Statement

My teaching focuses on making statistical reasoning usable in real analytical settings. I emphasise interpretation, model criticism and decision-making rather than treating methods as isolated formulae or software procedures. Students work with realistic data, incomplete information and ambiguous analytical choices, because these are the conditions under which statistical judgement is actually developed.

I use interactive visualisations, applied coding activities and structured model comparisons to help students connect mathematical ideas with the behaviour of fitted models. This is particularly important in regression, multivariate analysis and applied data science, where students can otherwise reproduce software output without understanding the geometry, assumptions or practical consequences of the analysis.

I also place substantial emphasis on accessibility for mixed-mode and distance learners. My teaching design aims to reduce unnecessary technical barriers while preserving statistical depth. This includes cloud-based analytical environments, downloadable code, reproducible examples and activities that allow students to inspect how results change when assumptions, variables or modelling choices are altered.

My professional work in biomedical research, clinical analytics and technology translation informs my teaching. Examples are drawn from health data, medical imaging, reproductive science and commercial analytics, allowing students to see how statistical methods interact with data governance, implementation constraints, uncertainty and organisational decisions. This is particularly relevant within the Master of Analytics, where graduates need to communicate quantitative evidence to technical and non-technical decision-makers.

Graduate Supervision Statement

My supervision style is collaborative, structured and strongly oriented towards producing research that can be interpreted and used beyond the thesis. I work with students to define a clear scientific or clinical question, establish an auditable analysis plan, identify the limitations of the available data, and connect methodological choices to the decisions the research is intended to inform.

Students are expected to develop independence, but not in isolation. I use regular meetings, written action points, reproducible analytical workflows and staged manuscript development to maintain momentum. Early work usually concentrates on data provenance, outcome definitions, validation strategy and the distinction between exploratory analysis and claims that can be supported. Later supervision focuses on interpretation, writing, publication and translation.

My supervision is interdisciplinary. I work with students from statistics, data science, computer science, medicine and biological sciences, often alongside clinicians, laboratory scientists and external partners. Areas of interest include breast cancer epidemiology and imaging, clinical prediction, linked health data, survival and longitudinal analysis, reproductive medicine, IVF, embryo assessment, sperm phenotyping, biomedical machine learning and computer vision.

I also support a shared postgraduate research environment in which students discuss methods, manuscripts and translational problems across related projects. This allows students to learn from work outside their immediate thesis while developing experience in collaborative research, clinical interpretation and communicating quantitative findings to different audiences. Your current supervision model already operates as an integrated research-training environment rather than a collection of disconnected projects.

Other Links

• Personal Research Website - Research programmes, publications, teaching resources and postgraduate opportunities
• Google Scholar - Publications and citation record
• LinkedIn - Academic, clinical and enterprise profile
• GitHub - Research software, teaching resources and technical projects
• 30,000 Voices Report - National analysis of breast cancer care and outcomes in Aotearoa New Zealand