BMJ open · 2025
Sex bias consideration in healthcare machine-learning research: a systematic review in rheumatoid arthritis.
СтатьяСистематический обзорНазвание 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.
Talwar Anahita, Turner Shruti, Maw Claudia, Quayle Georgina, Watt Thomas N, Gohil Sunir, Duckworth Emma, Ciurtin Coziana
Аннотация
OBJECTIVE: To assess the acknowledgement and mitigation of sex bias within studies using supervised machine learning (ML) for improving clinical outcomes in rheumatoid arthritis (RA).
DESIGN: A systematic review of original studies published in English between 2018 and November 2023.
DATA SOURCES: PUBMED and EMBASE databases.
STUDY SELECTION: Studies were selected based on their use of supervised ML in RA and their publication within the specified date range.
DATA EXTRACTION AND SYNTHESIS: Papers were scored on whether they reported, attempted to mitigate or successfully mitigated various types of bias: training data bias, test data bias, input variable bias, output variable bias and analysis bias. The quality of ML research in all papers was also assessed.
RESULTS: Out of 52 papers included in the review, 51 had a female skew in their study participants. However, 42 papers did not acknowledge any potential sex bias. Only three papers assessed bias in model performance by sex disaggregating their results. Potential sex bias in input variables was acknowledged in one paper, while six papers commented on sex bias in their output variables, predominantly disease activity scores. No paper attempted to mitigate any type of sex bias.
CONCLUSIONS: The findings demonstrate the need for increased promotion of inclusive and equitable ML practices in healthcare to address unchecked sex bias in ML algorithms.
PROSPERO REGISTRATION NUMBER: CRD42023431754.
DESIGN: A systematic review of original studies published in English between 2018 and November 2023.
DATA SOURCES: PUBMED and EMBASE databases.
STUDY SELECTION: Studies were selected based on their use of supervised ML in RA and their publication within the specified date range.
DATA EXTRACTION AND SYNTHESIS: Papers were scored on whether they reported, attempted to mitigate or successfully mitigated various types of bias: training data bias, test data bias, input variable bias, output variable bias and analysis bias. The quality of ML research in all papers was also assessed.
RESULTS: Out of 52 papers included in the review, 51 had a female skew in their study participants. However, 42 papers did not acknowledge any potential sex bias. Only three papers assessed bias in model performance by sex disaggregating their results. Potential sex bias in input variables was acknowledged in one paper, while six papers commented on sex bias in their output variables, predominantly disease activity scores. No paper attempted to mitigate any type of sex bias.
CONCLUSIONS: The findings demonstrate the need for increased promotion of inclusive and equitable ML practices in healthcare to address unchecked sex bias in ML algorithms.
PROSPERO REGISTRATION NUMBER: CRD42023431754.