Automated Immunophenotyping Assessment for Diagnosing Childhood Acute Leukemia using Set-Transformers

TL;DR

This paper introduces FCM-Former, a novel self-attention based machine learning tool that automates immunophenotyping in childhood acute leukemia, achieving high accuracy and reducing manual diagnostic efforts.

Contribution

The paper presents the first automated immunophenotyping method for pediatric leukemia diagnosis using FCM data with a self-attention model, improving speed and objectivity.

Findings

Achieved 96.5% accuracy in classifying leukemia types.

Automates the traditionally manual immunophenotyping process.

First to apply self-attention models to FCM data for leukemia diagnosis.

Abstract

Acute Leukemia is the most common hematologic malignancy in children and adolescents. A key methodology in the diagnostic evaluation of this malignancy is immunophenotyping based on Multiparameter Flow Cytometry (FCM). However, this approach is manual, and thus time-consuming and subjective. To alleviate this situation, we propose in this paper the FCM-Former, a machine learning, self-attention based FCM-diagnostic tool, automating the immunophenotyping assessment in Childhood Acute Leukemia. The FCM-Former is trained in a supervised manner, by directly using flow cytometric data. Our FCM-Former achieves an accuracy of 96.5% assigning lineage to each sample among 960 cases of either acute B-cell, T-cell lymphoblastic, and acute myeloid leukemia (B-ALL, T-ALL, AML). To the best of our knowledge, the FCM-Former is the first work that automates the immunophenotyping assessment with FCM data in diagnosing pediatric Acute Leukemia.