The absence of effective screening methods and the subtlety of the early stage signs makes pancreatic cancer a difficult illness to identify. Improved early diagnosis and treatment results are the goals of this research, which seeks to build a state-of-the-art deep learning model utilizing CNN and ANN. This model provides accurate risk predictions for pancreatic cancer based on an analysis of disease trajectories and the incorporation of large datasets including genetic, proteomic, clinical, and lifestyle data. Raising the bar on data quality and using cutting-edge feature extraction methods both do wonders for the model’s precision. The findings show that our deep learning method can help with early diagnosis, which in turn improves patient outcomes and survival rates.