This review will explore the integration of machine learning (ML) techniques to enhance the analysis of increasingly complex and voluminous flow cytometry data, as traditional manual methods are insufficient for handling this data. We attempt to provide a comprehensive introduction to ML in flow cytometry, detailing the transition from manual gating to computational methods and emphasizing the importance of data quality. Key ML techniques are discussed, including supervised learning methods like logistic regression, support vector machines, and neural networks, which rely on labeled data to classify disease states. Unsupervised methods, such as k-means clustering, FlowSOM, UMAP, and t-SNE, are highlighted for their ability to identify novel cell populations without predefined labels. We also delve into newer semi-supervised and weakly supervised methods, which leverage partial labeling to improve model performance. Practical aspects of implementing ML in clinical settings are addressed, including regulatory considerations, data preprocessing, model training, validation, and the importance of generalizability, and we underscore the collaborative effort required among pathologists, data scientists, and laboratory professionals to ensure robust model development and deployment. Finally, we show the transformative potential of ML in flow cytometry in uncovering new biological insights through advanced computational techniques.

Background/Objectives: Castleman disease is a rare cytokine-driven lymphoproliferative disorder in which lymph node histopathology provides key diagnostic information. Morphologic features are graded semiquantitatively and often show substantial interobserver variability. Methods: We developed an automated approach to grade six Castleman disease-associated histologic features on hematoxylin and eosin (H&E) whole slide images (WSIs) using an attention-based multiple instance learning (MIL) model built on a large pathology foundation model encoder. A multi-institution cohort comprised 544 lymph node WSIs, including 397 from cases with Castleman disease or Castleman-like histology and 147 from cases without suspicion for Castleman disease. These case ascertainment categories were not model prediction targets. Slides were assigned ordinal grades (0–3) for regressed germinal centers, follicular dendritic cell prominence, increased vascularity, hyperplastic germinal centers, plasmacytosis, and follicular twinning by hematopathologists. Model performance was assessed on a held-out evaluation set of 142 WSIs. Results: Across the six features, accuracy ranged from 0.52 to 0.68 (mean 0.60). Disagreements were predominantly minor: 96% of predictions were within one grade of the reference. Feature-specific tile contribution heatmaps showed qualitative spatial correspondence with selected plasma cell-rich, vessel-rich, and follicular regions but were not evaluated as quantitative feature localization maps or causal explanations. In a preliminary reader study, concordance with the reference varied widely among hematopathologists (Krippendorff’s alpha 0.20–0.95, mean 0.50), with the model demonstrating concordance in the mid-range (0.56). Conclusions: These findings support the feasibility of automated grading of Castleman disease-associated histologic features for research standardization. The model does not classify Castleman disease, and its potential use as an input to diagnostic or clinical trial workflows requires evaluation in separately designed studies with adjudicated diagnostic labels and integrated clinical data.