Nano-Tumor Database, which is recently released utilizing information, produced from a Physiologically Based Model (PBM) mode that contains 376 datasets, this study examined tumor models, the effects of NP physicochemical properties, and cancer types on NP tumor delivery efficiency. Outperforming all other machine learning techniques, the deep neural network model accurately predicted the effectiveness of various NPs in treating various tumors, consisting of Linear Regression, Support Vector Machine, Bagged model, and Random Forest techniques. To increase tumor delivery efficiency and to improve the design of cancer nanomedicine, this study offers a quantitative model. Our comprehension of the reasons for low NP tumor delivery efficiency is enhanced by these findings. This research shows that it is possible to study cancer nanomedicine by combining Artificial Intelligence with PBM modeling techniques.