Microneedle (MN) is a relatively recent invention and an efficient technology for transdermal 8 drug delivery (TDD). Conventionally, the mathematical models of MNs drug delivery define the 9 shape of the holes created by the MNs in the skin as the same as their actual geometry. 10 Furthermore, the size of the MN holes in the skin is considered to be either the same or a 11 certain fraction of the lengths of the MNs. However, the histological images of the MN treated 12 skin indicate that the real insertion depth is much shorter than the length of the MNs and the 13 shapes may vary significantly from one case to another. In addressing these points, we propose 14 a new approach for modelling MN based drug delivery, which incorporate the histology of MN 15 pierced skin using a number of concepts borrowed from image processing tools. It is expected 16 that the developed approach will provide better accuracy of the drug diffusion profile. A new 17 computer program is developed to automatically obtain the outline of the MNs treated holes and 18 import these images into computer software for simulation of drug diffusion from MN systems. 19This method can provide a simple and fast way to test the quality of MNs design and modelling, 20 as well as simulate experimental studies, e.g., permeation experiments on MN pierced skin 21 using diffusion cell. The developed methodology is demonstrated using two dimensional (2D) 22 numerical modelling of flat MNs (2D). However, the methodology is general and can be 23 implemented for 3D MNs if there is sufficient number of images for reconstructing a 3D image 24 for numerical simulation. Numerical modelling for 3D geometry is demonstrated by using 25 images of an ideal 3D MN. The methodology is not demonstrated for real 3D MN as there are 26 not sufficient numbers of images for the purpose of this paper. 27