Machine Learning of Discriminative Gate Locations for Clinical Diagnosis

Abstract

High‐throughput single‐cell cytometry technologies have significantly improved our understanding of cellular phenotypes to support translational research and the clinical diagnosis of hematological and immunological diseases. However, subjective and ad hoc manual gating analysis does not adequately handle the increasing volume and heterogeneity of cytometry data for optimal diagnosis. Prior work has shown that machine learning can be applied to classify cytometry samples effectively. However, many of the machine learning classification results are either difficult to interpret without using characteristics of cell populations to make the classification, or suboptimal due to the use of inaccurate cell population characteristics derived from gating boundaries. To date, little has been done to optimize both the gating boundaries and the diagnostic accuracy simultaneously. In this work, we describe a fully discriminative machine learning approach that can simultaneously learn feature representations (e.g., combinations of coordinates of gating boundaries) and classifier parameters for optimizing clinical diagnosis from cytometry measurements. The approach starts from an initial gating position and then refines the position of the gating boundaries by gradient descent until a set of globally‐optimized gates across different samples are achieved. The learning procedure is constrained by regularization terms encoding domain knowledge that encourage the algorithm to seek interpretable results. We evaluate the proposed approach using both simulated and real data, producing classification results on par with those generated via human expertise, in terms of both the positions of the gating boundaries and the diagnostic accuracy. © 2019 The Authors. Cytometry Part A published by Wiley Periodicals, Inc. on behalf of International Society for Advancement of Cytometry.

Artificial Intelligence-Assisted Diagnostic Cytology and Genomic Testing for Hematologic Disorders

“…Recent studies explored the use of AI in other areas of flow cytometric analysis beyond disease classification. Studies indicated that an AI-based method for automated gating of cell populations in flow cytometry data achieves high accuracy and consistency across different datasets [ 68 ]. In addition, Arvaniti and colleagues developed an AI-based tool called CellCNN to identify rare cell populations in flow cytometry data, which could have significant implications in the diagnosis and monitoring of various diseases [ 69 ].…”

A cell-level discriminative neural network model for diagnosis of blood cancers

“…Addressing this problem requires the simultaneous optimization of cell population identification (biomarkers) and sample-level classification (diagnosis). In previous work [21], we showed that the simultaneous optimization can be achieved for diagnosis of chronic lymphocytic leukemia (CLL), by adapting gradient descent optimization for identifying a global optimal gate to maximize classification accuracy. In this paper, we hypothesize that a density-based discriminative point set model using backpropagation can address the simultaneous optimization, without requiring initial gating.…”

A cell-level discriminative neural network model for diagnosis of blood cancers

“…Addressing this problem requires the simultaneous optimization of cell population identification (biomarkers) and sample-level classification (diagnosis). In previous work ( Ji et al 2020 ), we showed that the simultaneous optimization can be achieved for diagnosis of chronic lymphocytic leukemia (CLL), by adapting gradient descent optimization for identifying a global optimal gate to maximize classification accuracy. In this article, we hypothesize that a density-based discriminative point set model using backpropagation can address the simultaneous optimization, without requiring initial gating.…”