Respiratory medicine · 2025
Characterisation of patients with Alpha-1 antitrypsin deficiency using unsupervised machine learning tools.
СтатьяПоддержка исследования: вне правительства СШАНазвание и аннотация: Перевод пока не подготовлен
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.
Villar-Aguilar Laura, Casal-Guisande Manuel, Fernández-Villar Alberto, García-Rodríguez Esmeralda, Priegue-Carrera Ana, Miravitlles Marc, Torres-Durán María
Аннотация
BACKGROUND: Alpha-1 antitrypsin deficiency (AATD) is an underdiagnosed and clinically heterogeneous genetic disorder that increases the risk of pulmonary emphysema and liver disease. Its management is complex due to high individual variability, which hinders risk stratification and personalised treatment. Artificial Intelligence (AI), particularly Machine Learning (ML) techniques, may offer novel approaches to the clinical characterisation of AATD.
MATERIALS AND METHODS: A pilot, single-centre cross-sectional study was conducted, including 210 AATD patients from the EARCO registry at Álvaro Cunqueiro Hospital (Vigo, Spain), between February 2020 and June 2023. The unsupervised k-prototypes algorithm was applied to identify clusters based on clinical and demographic variables. Functional and clinical differences among clusters were analysed using Mann-Whitney U and Chi-square tests, with p < 0.05 considered significant. This pilot study, based on a regional cohort with high Pi∗SZ prevalence, explores the feasibility of applying AI in clinical practice without aiming to represent the full EARCO registry.
RESULTS: Five clinically distinct clusters were identified: A) Pi∗ZZ patients with emphysema and severe functional impairment; B) women with bronchiectasis and preserved pulmonary function; C) young asymptomatic individuals with the Pi∗SZ genotype; D) older patients with COPD and cardiovascular comorbidities; E) individuals with altered hepatic profiles, alcohol consumption, and moderate AATD. Significant differences in pulmonary function, AAT levels, and comorbidities were observed.
CONCLUSIONS: This pilot study shows the feasibility of using ML to segment AATD patients into meaningful clusters, adding value for personalised medicine. This approach may guide therapeutic decisions, improve follow-up, and support the design of cluster-based multicentre trials.
MATERIALS AND METHODS: A pilot, single-centre cross-sectional study was conducted, including 210 AATD patients from the EARCO registry at Álvaro Cunqueiro Hospital (Vigo, Spain), between February 2020 and June 2023. The unsupervised k-prototypes algorithm was applied to identify clusters based on clinical and demographic variables. Functional and clinical differences among clusters were analysed using Mann-Whitney U and Chi-square tests, with p < 0.05 considered significant. This pilot study, based on a regional cohort with high Pi∗SZ prevalence, explores the feasibility of applying AI in clinical practice without aiming to represent the full EARCO registry.
RESULTS: Five clinically distinct clusters were identified: A) Pi∗ZZ patients with emphysema and severe functional impairment; B) women with bronchiectasis and preserved pulmonary function; C) young asymptomatic individuals with the Pi∗SZ genotype; D) older patients with COPD and cardiovascular comorbidities; E) individuals with altered hepatic profiles, alcohol consumption, and moderate AATD. Significant differences in pulmonary function, AAT levels, and comorbidities were observed.
CONCLUSIONS: This pilot study shows the feasibility of using ML to segment AATD patients into meaningful clusters, adding value for personalised medicine. This approach may guide therapeutic decisions, improve follow-up, and support the design of cluster-based multicentre trials.