Page 85 - AI for Good Innovate for Impact

AI for Good Innovate for Impact

loss dataset to bolster geospatial analysis and modelling efforts. Acknowledging that these
datasets reflect conditions as of 2021, they are considered valid for analysis from 2021 to 2025.
This assumption rests on the understanding that significant structural determinants, particularly
housing infrastructure in remote communities, are unlikely to undergo substantial change 4.1-Healthcare
within this timeframe. A known limitation is the ABS's public acknowledgement of persistent
undercounting, with a net undercount of Aboriginal and Torres Strait Islander people estimated
at approximately 17.4% in the 2021 Post Enumeration Survey [2]. To mitigate potential issues
arising from mismatched data vintages across layers, the project will flag the year of origin for
each dataset, apply cautious interpretation when comparing data, and commit to updating
models and overlays once the 2026 Census data becomes available. Public environment
datasets from sources such as [7] and [8] related to water, sanitation conditions, and food
security are also integrated. Recognised limitations for these datasets include typically broader
geographical levels, necessitating careful consideration of spatial resolution and infrequent
updates. To address this, only data corresponding to the specific year of the hearing loss
dataset will be used [3]. The project aims to generate data-driven policy recommendations
for targeted healthcare interventions in communities grappling with high rates of otitis media.
Furthermore, it will produce clear, actionable reports for stakeholders, including policymakers,
healthcare providers, and community leaders, highlighting the crucial correlations between
social factors and otitis media prevalence.

Use-case status: The use case is part of a larger research project�
Partner(s): Charles Darwin University [6]

2�2 Benefits of use case

Aboriginal and Torres Strait Islander children in Australia have the highest rates of ear disease
(otitis media) in the world. It can lead to lifelong impacts making it a critical area for intervention.
The Geospatial Mapping of Otitis Media (OM) and its correlation with social determinants in
remote Northern Territory communities presents a groundbreaking approach to addressing
a critical public health issue for Aboriginal Australians.

This proposal directly targets Good Health and Well-being, as it aims to reduce the health
burden of otitis media. By integrating AI-powered geospatial analysis, this project will enable
policymakers and health professionals to identify at-risk areas for more targeted service
delivery and interventions, effectively improving access and outcomes in underserved regions.
Therefore, this use case focuses on addressing health disparities and systemic inequalities faced
by Aboriginal communities. By using AI to create heatmaps and spatial clusters, the project will
generate actionable insights that can facilitate targeted health interventions. The innovative use
of AI and geospatial visualisation holds the potential to evolve into a standard for health policy
applications, particularly in remote areas that face similar public health challenges globally. The
project offers scalability by integrating up-to-date census data, with future-proofing measures
to keep the system evolving in line with changing socio-economic indicators. Future work
includes extending this model for broader application across Indigenous populations globally,
creating a sustainable and impactful solution to address inequalities in healthcare delivery.

49