While single-cell "omics" have been critical for the study of human disease, most of these approaches require tissue dissociation, losing important spatial information. Conversely, traditional immunohistochemical methods are only able to capture the spatial expression of only a limited set of markers. Thus, multiplexed microscopy has reshaped the study of tissue biology by measuring cell expression from +40 proteins while maintaining the spatial context. This study aims to understand the cellular and immune landscapes that underpin highly heterogeneous renal autoimmunity. As such, we collected and analyzed using our advanced computational pipeline for image analysis: 25 Lupus Nephritis (LN), 23 Renal Allograft Rejection (RAR), and six ”normal” kidney (NK) control samples. Biopsies were iteratively stained using the PhenoCycler protocol, and imaged with a spinning disk confocal microscope to acquire high-resolution, full-section images with a 42-marker panel covering a wide array of immune and non-immune markers.In this work, we applied novel algorithms for cell detection, segmentation, classification, and spatial feature extraction, to examine the interaction of immune cells within the renal microenvironment. Nuclear detection and segmentation was performed using Cellpose2.0, and the cell body was approximated by dilating the nuclear masks. We have further developed a decision-tree classifier for the multiclass annotation of renal cells that is analogous to well-established flow-cytometry-based cell analyses and immunophenotyping. In our decision tree, cells are sequentially sorted into marker-negative and marker-positive populations using their mean fluorescence intensity (MFI). Marker order is based upon well-established, hierarchical expression of immunological cell markers. Moreover, we have further created a computational tool to capture Microtubule Organizing Center (mTOC) polarization between pairs of cells, which we leverage to better direct cell-cell contacts and cognate immunity. Ultimately, we identify key immune populations involved in renal autoimmunity, in particular, CD14+MerTk+ M1 MΦ, CD14+CD163+ M2 (MΦ), and CD8+ T-cells. We detect a cohort-specific polarization of immune lineages, with LN showing a predominant myeloid- driven pathology, while RAR was characterized by a T-cell autoimmune response. The presence of these immune players in tissue was significantly associated with various features of tissue inflammation, elucidating their potential as biomarkers for disease activity and progression. CD14+MerTk+ macrophages, in concert with CD8+ T-cells, appear to drive an inflammatory response in RAR. In contrast, inflammation in our LN samples is overwhelmingly driven by CD14+MerTk+ macrophages mostly independent from T-cells. This would suggest a more nuanced cellular interplay, where T-cell modulation of macrophage activity influences the inflammatory microenvironment. In summary, our comprehensive spatial analysis offers new insights into the cellular players underpinning renal autoimmunity; highlighting the potential of immune spatial features to be leveraged as a predictive and diagnostic tool. This dissertation paves the way for future research into targeted therapies that address the unique immunological profiles of autoimmune patients; with a greater emphasis on targeting the myeloid cell compartment in LN and the myeloid:T-cell interactions in RAR. The development of computational tools for spatial analysis presents a technical advancement that allows for a novel approach to understanding the complex and heterogeneous immune responses in renal autoimmunity.