Computer-assisted interpretation of flow cytometry data in hematology

Abstract

A computer program has been developed for computer‐assisted diagnosis (including subclassification) of flow cytometry data of acute leukaemias and non‐Hodgkin lymphomas by means of artificial intelligence. The knowledge base for the system has been formulatedas semantic networks that describe physiological hematopoiesisas well as the pathological situation (eg., aberrant antigen expression) of hematological disorders. The semantic networks reflect the hierarchy of cells and their occurrence in diseases, the normal and pathological antigen expression patterns of cells, cell maturation, and the frequency of cell populations in normal blood and bone marrow. Using these semantic networks, the diagnosis algorithm compares the characteristic antigen expression pattern of a disease with the actual findings in the blood or bone marrow sample. The algorithm can separate mixed populations by taking double staining findings into account. Finally, a diagnosis text is generated that describes all identified cell populations and the resulting diagnosis. The validationof the program showed a correct diagnosis (disease group and subclassification) in 97% of the cases (n = 633) with slight differences between the disease groups (e.g., B‐NHL: 99%, B‐cell ALL: 84%). © 1996 Wiley‐Liss, Inc.

Automated analysis in flow cytometry

“…While the process is transparent and flexible, this approach may not uncover data relationship that has not been envisioned by the human expert. Another rule‐based system is semantic networks (12), a graphic‐based hierarchal representation of relations which show logical relations between classes. While easily understood and adaptable, the diagnostic procedure is also deterministic and limited by the state of knowledge of the expert.…”

Automated classification of patients with chronic lymphocytic leukemia and immunocytoma from flow cytometric three-color immunophenotypes

“…Rule-or knowledge-based systems (8,30) exploit only the restricted expert knowledge on the currently interpretable fraction of the total information con- tent of multiparameter cytometric measurements. Problems in cluster analysis (7,24,33) concern the meaningful definition of biologically relevant clusters in the learning data set and the handling of lost or new clusters in pathological conditions.…”

Prädiktivmedizin mittels Zytomik: Möglichkeiten der Datenmusteranalyse

“…Die bioinformatische Wissensextraktion strebt die Bestimmung des Musters derjenigen zellulären Informationsparameter an, die Auskunft über den augenblicklichen (Diagnose) oder den therapieabhängigen zukünftigen (Prädiktion) Krankheitsverlaufvon Einzelpatienten geben können. Dazu werden in derZytometrie häufig parametrische, d.h. mathematische Verfahren wie Cluster- [7,8] oder Hauptkomponentenanalyse [9], Multivarianz-und Statistikverfahren [10][11][12], wissensbasierte [13,14] Die genannten Verfahren sind bezüglich der Auswertezielsetzung überwiegend gruppenorientiert. Ihre Durchführung kann an mathematische Voraussetzungen bezüglich der Werteverteilung von Messparametem gebunden sein.…”