Journal of diabetes research · 2026
Exploration of Early Biomarkers for Gestational Diabetes Mellitus in Pregnant Women Living Without Obesity.
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.
Wu Dan, Xu Meiyi, Zhang Cunling, Li Shanshan, Yao Liying, Geng Mengyuan, Guo Yuling, Wei Zhuo, Li Wen
Аннотация
BACKGROUND: Early prediction of gestational diabetes mellitus (GDM) enables timely interventions but remains challenging, particularly in women living without obesity who are often overlooked by standard screening methods. This exploratory analysis identifies early pregnancy biomarkers that show associations with GDM up to 8-12 weeks earlier than the established diagnostic timeframe.
METHODS: A retrospective nested case-control study was conducted at Tianjin Central Hospital of Gynecology Obstetrics. After applying the predefined inclusion and exclusion criteria and removing cases with missing blood samples, a total of 136 women were included in the final analysis, comprising 56 women with GDM and 80 non-GDM controls. These data were used to develop multiple machine learning models, including random forest, logistic regression, SVM, XGBoost, and KNN. The performance of these models was assessed by AUC, sensitivity, specificity, positive predictive value, and negative predictive value in an independent validation set.
RESULTS: The random forest classifier achieved the best performance with an AUC of 0.992 in the validation set, indicating excellent sensitivity and specificity. The model incorporated six measurable variables: age, BMI, and four cytokines (IL-1β, IL-10, IL-17A, and IL-4) participating in immune regulation in pregnancy.
CONCLUSIONS: GDM can be predicted with high accuracy early in pregnancy, even among women living without obesity. This study highlights the value of integrating immunological biomarkers with clinical characteristics and machine learning for proactive risk stratification. Although promising, our findings warrant validation in larger, multiethnic cohorts to ensure generalizability.
METHODS: A retrospective nested case-control study was conducted at Tianjin Central Hospital of Gynecology Obstetrics. After applying the predefined inclusion and exclusion criteria and removing cases with missing blood samples, a total of 136 women were included in the final analysis, comprising 56 women with GDM and 80 non-GDM controls. These data were used to develop multiple machine learning models, including random forest, logistic regression, SVM, XGBoost, and KNN. The performance of these models was assessed by AUC, sensitivity, specificity, positive predictive value, and negative predictive value in an independent validation set.
RESULTS: The random forest classifier achieved the best performance with an AUC of 0.992 in the validation set, indicating excellent sensitivity and specificity. The model incorporated six measurable variables: age, BMI, and four cytokines (IL-1β, IL-10, IL-17A, and IL-4) participating in immune regulation in pregnancy.
CONCLUSIONS: GDM can be predicted with high accuracy early in pregnancy, even among women living without obesity. This study highlights the value of integrating immunological biomarkers with clinical characteristics and machine learning for proactive risk stratification. Although promising, our findings warrant validation in larger, multiethnic cohorts to ensure generalizability.