This project is part of the DEPRIMAP research (Unravelling the dynamics of deprived urban areas in the Majority World using AI and Earth Observation to foster evidence-based sustainable planning), a FORMAS-funded doctoral project (grant no. 2023-01210) hosted at Karlstad University. The PhD candidate’s main supervisor is Stefanos Georganos in Geomatics at Karlstad University. DEPRIMAP aims to improve understanding of urban deprivation by integrating geospatial data, spatial modelling, and machine learning techniques, thereby supporting local governments and stakeholders in risk reduction and resilience strategies. In the previous project period, we developed and applied the City Segment Morphological Deprivation (CSMD) model, a globally consistent, neighbourhood-scale framework for mapping morphological deprivation across 5,132 cities in 103 countries across Africa, Asia, and Latin America and the Caribbean. This work is now in press at Nature Cities. Building on this foundation, the current project phase shifts focus to population modelling of deprived urban areas. This involves large-scale geospatial data processing and the development of bottom-up predictive models, including Random Forest, XGBoost, and Support Vector Machine approaches, as well as experimental work with satellite embedding-based models. The analysis requires extensive computation of morphometric indicators (building form, density, layout) and road network metrics (connectivity, accessibility), among other metrics, across thousands of cities, both CPU- and memory-intensive operations given the scale and resolution of the underlying vector and raster data. Outputs will include geospatial indicators of population distribution within deprived and non-deprived urban areas, comparisons across modelling approaches, and cross-country datasets to inform both scientific understanding and policy decision-making. The significance of this work lies in its scientific contribution to urban studies and population/deprivation modelling, as well as its operational potential: enabling standardised, scalable, and reproducible assessments of population distribution in deprived urban areas using open-source data and methods. The project will generate openly accessible datasets and documented processing pipelines to benefit researchers and practitioners working on urban resilience, poverty mapping, and sustainable development.