Thammasat University students interested in the allied health sciences, big data, artificial intelligence, economics, computer science, technology, and related subjects may find it useful to participate in a free 15 August Zoom webinar conference on how data science is being used in various sectors including public health: The Crazy Places Data Science Can Take You.

The event, on Monday 15 August 2022 at 11am Bangkok time, is organized by the University of Melbourne, Australia.

The TU Library collection includes several books about different aspects of using data science in health.

As the webinar description posted online explains:

Data science is an amazing field that brings together mathematics and computer science – along with biology, physics, chemistry, policy, finance, climate science, advertising … you name it and data scientists are probably working on it.

In this masterclass we welcome data scientists Dr Christopher Baker and Isobel Abell from the School of Mathematics and Statistics, and the , who will give us the lowdown on how data science is now pretty much everywhere, and how governments are using data scientists to protect us from foot and mouth disease, Covid outbreaks and more.

Students are invited to register for the free event at this link:

https://melbconnect.com.au/events/the-crazy-places-data-science-can-take-you/

The speakers will include data scientist Dr. Christopher Baker, a research fellow at the School of Mathematics and Statistics and the Centre of Biosecurity Risk Analysis in the School of BioSciences, University of Melbourne:

His research focusses on using mathematics to inform policy in biosecurity, ecology and epidemiology. Since joining The University of Melbourne in 2020, he has worked on COVID-19 modelling to support policy in Australia and countries in Oceania through WHO collaboration; worked with the Department of Agriculture, Water and Environment on biosecurity policy; and has worked with the Melbourne Centre of Data Science to help build collaborations with the focus of using mathematics to improve policy in biological systems.

Also speaking will be data scientist Isobel Abell, a PhD student in the School of Mathematics and Statistics at the University of Melbourne. Her work focuses on using infectious disease modelling in a decision-making context.

Last year a research article published in Public Health Reviews examined The Emergence and Future of Public Health Data Science.

It noted:

Data science is a newly‐formed and, as yet, loosely‐defined discipline that has nonetheless emerged as a critical component of successful scientific research. We seek to provide an understanding of the term “data science,” particularly as it relates to public health; to identify ways that data science methods can strengthen public health research; to propose ways to strengthen education for public health data science; and to discuss issues in data science that may benefit from a public health perspective…

Public health data science is the study of formulating and rigorously answering questions in order to advance health and well-being using a data-centric process that emphasizes clarity, reproducibility, effective communication, and ethical practices…

Public health is well situated not simply to react to the emergence of data science, but to lead in the ongoing evolution of this dynamic new field. Data have been the foundation of public health’s mission: to understand the burden of disease, disability and injury and the opportunity to improve health across the full life course, to recognize solutions to disparities, to infer causal mechanisms, and to provide evidence for the impact of interventions. Public health researchers are trained to think critically about the appropriateness of a study’s design to evaluate scientific hypotheses, and to interrogate the measurement and sampling processes that produce observations. Public health research is inherently interdisciplinary and collaborative, drawing from quantitative and qualitative expertise across domains to effect change in the health of populations. Crucially, public health is concerned with the ethical questions that surround data and prediction algorithms, and the impact these can have on exacerbating disparities in health outcomes. These long-standing public health competencies are clearly relevant to the future of data science…

Innovative Data Science Research Methods can Extend Public Health’s Reach

Data relevant to public health has grown in scale and complexity, and will continue to do so for the foreseeable future. Now-common observations on individuals that involve large quantities of data already include genomic and other information available through biosamples; exposure to mixtures of environmental pollutants; lifestyle behaviors measured continuously through wearable devices; detailed health care histories from electronic records; and the social media, search queries, digital records of on- and off-line transactions, and similar elements that make up an individual’s digital footprint. These data sources have emerged in parallel to the development of new analytical strategies and capabilities, including statistical or machine learning methods, prediction algorithms, and deep neural networks. Rich data and complex methods promise to reshape the questions public health researchers can ask and the ways in which those questions can be answered. To capitalize on this potential, it is necessary to synergistically combine data science approaches with the public health science perspectives…

Meanwhile, we recognize prediction accuracy as an important goal in itself… Clearly, data science for public health will rely on interdisciplinary teams to make advances–no single researcher, or even a research team comprised of members of a single discipline, will have the requisite breadth of expertise needed to solve problems in this environment. Successful teams will be well-versed in the behavioral, social, or biological determinants of the health outcome of interest; understand the complex systems that describe the determinants and outcomes; recognize the potential for opaque methods to propagate bias inherent in underlying data; and rely on quantitative experts to identify and implement appropriate analytic strategies. Increasingly, teams will integrate expertise in bioinformatics, computer science, engineering, and other quantitative domains that have been, to date, infrequent partners in public health research. As a consequence, institutions that adopt incentives to promote team science and actively seek to bridge silos of expertise will be leaders in public health data science; external groups, particularly those that fund research, should encourage this through initiatives that reward interdisciplinary work…

Outlook and Conclusion

We have defined and discussed public health data science, but have not precisely located this field with respect to public health or public health’s constituent disciplines. This reflects current reality–individuals who identify as public health data scientists often do so as a secondary discipline, with primary expertise in epidemiology, biostatistics, health policy, environmental health, or another area. It’s possible that this will remain true, but the present state may also be attributable to institutional organization around traditional fields or the ways that organization has shaped training in the past. Over time, public health data science may emerge as a primary domain among those for whom the implied perspective resonates, suggesting a need for institutional flexibility or reorganization. In any case, we want to be clear that public health data science does not simply rebrand an existing discipline like epidemiology or biostatistics; this view is flawed in a way that is analogous to dismissals of data science as a new term for computer science.

Whether the public health data science persists in a loose, interdisciplinary form or solidifies as a distinct field is not something we can predict, and will depend both on how existing disciplines define themselves moving forward and the ways data science itself evolves over time. While cannot predict the future, we do look forward to an ongoing, mutually transformative partnership between public health and data science that will strengthen both disciplines and the improve ability to extract from data the actionable insights that advance health.

(All images courtesy of Wikimedia Commons)