International journal of rheumatic diseases · 2025
An Efficient Machine Learning Pipeline for Distinguishing Cancer and Non-Cancer Patients in Systemic Lupus Erythematosus.
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.
Chen You-Yue, Huang An-Fang, Yang Jing, Fu Lu, Xu Wang-Dong
Аннотация
BACKGROUND: The objective is to develop a machine learning model that is capable of effectively distinguishing between cancer and non-cancer patients among systemic lupus erythematosus (SLE).
METHODS: A total of 2811 patients with SLE were included in this study, among which 208 had concurrent cancers. Age, gender, and 95 clinical and laboratory indicators were included. Initially, data preprocessing and feature transformation were conducted. After evaluating feature importance, selected features were included in the subsequent steps for establishing the machine learning model. The optimal model chosen was assessed using a series of metrics to obtain a comprehensive performance evaluation of the model.
RESULTS: Through integrated feature selection, features such as Age, C4, WBC (urine), BASO_R, ALP, AST, and C3 were screened out for model construction. Among the various machine learning (ML) models established after parameter optimization, the logistic regression model performed the best. Furthermore, the logistic regression model still showed the best predictive capability by evaluating metrics such as receiver operating characteristic (ROC) curves, precision-recall curve (PRC) plots, as well as model prediction accuracy, F1 score, and recall.
CONCLUSION: We have developed a convenient and straightforward machine learning model for SLE patients, which can easily and usefully distinguish cancer from non-cancer patients within the SLE patients.
METHODS: A total of 2811 patients with SLE were included in this study, among which 208 had concurrent cancers. Age, gender, and 95 clinical and laboratory indicators were included. Initially, data preprocessing and feature transformation were conducted. After evaluating feature importance, selected features were included in the subsequent steps for establishing the machine learning model. The optimal model chosen was assessed using a series of metrics to obtain a comprehensive performance evaluation of the model.
RESULTS: Through integrated feature selection, features such as Age, C4, WBC (urine), BASO_R, ALP, AST, and C3 were screened out for model construction. Among the various machine learning (ML) models established after parameter optimization, the logistic regression model performed the best. Furthermore, the logistic regression model still showed the best predictive capability by evaluating metrics such as receiver operating characteristic (ROC) curves, precision-recall curve (PRC) plots, as well as model prediction accuracy, F1 score, and recall.
CONCLUSION: We have developed a convenient and straightforward machine learning model for SLE patients, which can easily and usefully distinguish cancer from non-cancer patients within the SLE patients.