Transfusion · 2026
IrregulAB: An open-source rule-based engine for automated red cell antibody panel interpretation.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Crucitti Davide, Ballester Violeta Vidal, Fernández Jesús Gómez, Mellid Eugenia Fernandez, Perdomo Angie Lizeth Ruge, Míguez Carlos Pérez, Orgueira Adrián Mosquera
Аннотация
BACKGROUND: Interpretation of antibody identification panels remains a manual, expertise-intensive step in transfusion testing; interpretive errors and long turnaround times persist even in well-equipped hospitals. Existing commercial software is closed source and often omits enzyme-phase, cold-phase, and dosage-sensitive exclusion rules, limiting third-party verification and local adaptation.
STUDY DESIGN AND METHODS: We developed IrregulAB, an open-source (LGPL) web application that encodes serologic decision rules in two stages. First, a rule-based screening stage applies LISS, enzyme, and cold rule-out logic with safeguards for antigen dosage in heterozygotes to discard antibodies incompatible with the observed reaction pattern. Second, a minimal-set solver exhaustively searches for the smallest antibody combination(s) that reproduce all assessable reactions and reports all equally plausible solutions. The system was evaluated on 108 development antibody-identification panels and 35 held-out temporal validation panels.
RESULTS: The screening stage retained the complete reference antibody set in 93% of development panels and 91% of held-out temporal validation panels. The minimal-set stage produced at least one valid explanatory antibody combination in 87% of development panels and 89% of held-out temporal validation panels.
CONCLUSION: IrregulAB is a transparent, rule-based decision-support tool for trained immunohematology personnel; it is not an autonomous antibody-identification or reporting system. By exposing every encoded decision rule, it enables independent verification, local tuning, and more standardized review of routine antibody workups under expert supervision. Its current performance was evaluated retrospectively in a single-center Bio-Rad gel-card dataset; multicenter and multi-platform validation is required before wide clinical implementation.
STUDY DESIGN AND METHODS: We developed IrregulAB, an open-source (LGPL) web application that encodes serologic decision rules in two stages. First, a rule-based screening stage applies LISS, enzyme, and cold rule-out logic with safeguards for antigen dosage in heterozygotes to discard antibodies incompatible with the observed reaction pattern. Second, a minimal-set solver exhaustively searches for the smallest antibody combination(s) that reproduce all assessable reactions and reports all equally plausible solutions. The system was evaluated on 108 development antibody-identification panels and 35 held-out temporal validation panels.
RESULTS: The screening stage retained the complete reference antibody set in 93% of development panels and 91% of held-out temporal validation panels. The minimal-set stage produced at least one valid explanatory antibody combination in 87% of development panels and 89% of held-out temporal validation panels.
CONCLUSION: IrregulAB is a transparent, rule-based decision-support tool for trained immunohematology personnel; it is not an autonomous antibody-identification or reporting system. By exposing every encoded decision rule, it enables independent verification, local tuning, and more standardized review of routine antibody workups under expert supervision. Its current performance was evaluated retrospectively in a single-center Bio-Rad gel-card dataset; multicenter and multi-platform validation is required before wide clinical implementation.