Mass cytometry imaging is a technique that utilizes mass spectrometry instead of fluorescent flow cytometry to detect metal-conjugated antibodies binding to single cell antigens. This method generates multi-dimensional images, a complex data type with significant computational and interpretation challenges due to the absence of a standardized analysis workflow. To address this, our project evaluated the quality and efficacy of a previous study’s proposed workflow to advance the interpretability and accuracy of mass cytometry data analysis. The approach evaluated whether the previously proposed workflow could effectively analyze alternative samples classified by previous studies as “mixed” immune structure. The methods included an in-depth examination of cell labeling based on spatial proximity, which emerged as a secondary analytical outcome. This research advances spatial labeling by enhancing the spatial information related to physical connections between interacting cells and incorporating biological interactions, such as protein involvement. By integrating these factors, this research proposes a method for more accurate labeling, contributing to a more robust analysis workflow for this complex data type. Unstructured “mixed” samples remained distinguishable, though less effectively than structured “compartmentalized” samples. Future applications involve exploring additional datasets to further develop means of distinguishing between contact and non-contact interactions among cancer and immune cells.