One event to occur during the development of type one diabetes is inflammation and destruction of the islets due to the presence of immune cells, known as insulitis. A better way to understand insulitis is through Multiplex Immunohistochemistry (IHC) whole slide scans. IHC analysis is based on a mixture of qualitative analysis made by expert pathologists and manual tracing of the structures for further quantitative data extraction. However, a great limitation of the current analysis of IHC is the loss of useful information that can facilitate the understanding of insulitis and its later effects. To overcome these limitations, we developed a computer vision algorithm that can extract structural and morphological descriptors such as the total area of the islet, glucagon, insulin and the distribution across the tissue from a single cell to the entire islet structure. In addition, we provide the capability of quantifying CD3+ cells to determine the presence of inflammation from the islet and single cells. This algorithm has demonstrated a 98% accuracy to identify islets above 300-micron square. Moreover, we were able to accelerate the time to process a 3-stain (glucagon, insulin, and CD3) IHC pancreas image of 250Mb in size to an approximate time of 30 minutes. This algorithm was developed on Python using a mixture of image processing and machine learning algorithms. It is envisioned that this algorithm can be improved by introducing immunofluorescence whole slide scan analysis while increasing the number of stains per image.