Background <p>Cancer remains a major health challenge globally, with significant regional variations in incidence and mortality. In India, the incidence of cancer is projected to increase over the next decade, with marked regional disparities. Despite the availability of data from cancer registries, there is a gap in the use of real-time data and advanced forecasting techniques to understand spatial and temporal trends in cancer incidence. This study aims to fill this gap by employing real-time data and advanced forecasting methodologies to analyze the spatio-temporal patterns of cancer incidence in India.</p> Methods <p>Secondary data on cancer incidence from 36 regions (states and union territories) in India, obtained from the Open Government Data (OGD) Platform India, was used. The study focused on calculating the Cancer Crude Incidence Rate (CCIR) for the years 2018 to 2024. A Bayesian Spatio-Temporal Conditional Autoregressive model was used to model the spatio-temporal dynamics of CCIRs. Additionally, an Artificial Neural Network (ANN) was employed to project cancer incidence rates for 2025 and 2026. Spatial clustering techniques, including Local Moran’s I statistic, were used to identify cancer hotspots and cold spots across regions.</p> Results <p>From 2018 to 2024, temporal trends revealed an increasing CCIR across multiple regions, with notable spatial heterogeneity. The Bayesian model highlighted distinct spatio-temporal dependencies, while the ANN model demonstrated robust predictive accuracy for CCIR projections in 2025 and 2026. Emerging hotspots and cold spots were identified, providing actionable insights for targeted cancer control interventions.</p> Conclusion <p>The study highlights the increasing cancer burden in India and the significant regional variations in incidence. The findings emphasize the need for region-specific strategies and highlight the utility of real-time data and advanced models for forecasting and spatial analysis in public health.</p>