Frontiers in endocrinology · 2025
Construction and verification of prediction model for Alzheimer's disease in diabetic patients.
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.
Wang Li, Sun Qing
Аннотация
OBJECTIVE: This study aimed to construct a prediction model for Alzheimer's disease (AD) in diabetic patients and evaluate its clinical application value.
METHODS: A total of 322 patients was included and randomly divided into a training set (n=225) and a validation set (n=97) at a ratio of 7:3. Clinical characteristic data of the patients were collected. In the training set, univariate analysis and multivariate logistic regression analysis were used to identify the relevant risk factors for AD onset, and a nomogram prediction model was constructed accordingly. The receiver operating characteristic (ROC) curve and calibration curve were plotted and validated in an independent validation dataset. In addition, decision curve analysis (DCA) was used to further evaluate the application value and significance of the nomogram model in clinical practice.
RESULTS: The incidence of AD in the training set was 18.67% (42/225), and that in the validation set was 18.56% (18/97). Multivariate regression analysis showed that age, duration of diabetes, fasting plasma glucose (FPG), glycosylated hemoglobin (HbA1c), triglyceride (TG), and homeostasis model assessment of insulin resistance (HOMA-IR) were all independent risk factors for AD onset (all P < 0.05). In the training set and validation set, the nomogram prediction model showed good predictive performance, with the concordance index (C-index) reaching 0.868 and 0.710 respectively. Calibration curve analysis showed a high degree of agreement between the predicted values and the observed values. The mean absolute errors in the training set and validation set were 0.103 and 0.116 respectively. The results of the Hosmer-Lemeshow test were χ² = 10.515, P = 0.230 and χ² = 5.987, P = 0.648 respectively. The ROC curve showed that the AUCs of the nomogram model for predicting occurrence of AD in the training set and validation set were 0.866 (95% CI: 0.794- 0.939) and 0.718 (95% CI: 0.517-0.920) respectively.
CONCLUSION: The prediction model for AD in diabetic patients can assist in the early prediction of the risk of AD onset, laying a solid foundation for formulating effective clinical intervention strategies. This is crucial for delaying the progression of AD and significantly improving the quality of life of patients.
METHODS: A total of 322 patients was included and randomly divided into a training set (n=225) and a validation set (n=97) at a ratio of 7:3. Clinical characteristic data of the patients were collected. In the training set, univariate analysis and multivariate logistic regression analysis were used to identify the relevant risk factors for AD onset, and a nomogram prediction model was constructed accordingly. The receiver operating characteristic (ROC) curve and calibration curve were plotted and validated in an independent validation dataset. In addition, decision curve analysis (DCA) was used to further evaluate the application value and significance of the nomogram model in clinical practice.
RESULTS: The incidence of AD in the training set was 18.67% (42/225), and that in the validation set was 18.56% (18/97). Multivariate regression analysis showed that age, duration of diabetes, fasting plasma glucose (FPG), glycosylated hemoglobin (HbA1c), triglyceride (TG), and homeostasis model assessment of insulin resistance (HOMA-IR) were all independent risk factors for AD onset (all P < 0.05). In the training set and validation set, the nomogram prediction model showed good predictive performance, with the concordance index (C-index) reaching 0.868 and 0.710 respectively. Calibration curve analysis showed a high degree of agreement between the predicted values and the observed values. The mean absolute errors in the training set and validation set were 0.103 and 0.116 respectively. The results of the Hosmer-Lemeshow test were χ² = 10.515, P = 0.230 and χ² = 5.987, P = 0.648 respectively. The ROC curve showed that the AUCs of the nomogram model for predicting occurrence of AD in the training set and validation set were 0.866 (95% CI: 0.794- 0.939) and 0.718 (95% CI: 0.517-0.920) respectively.
CONCLUSION: The prediction model for AD in diabetic patients can assist in the early prediction of the risk of AD onset, laying a solid foundation for formulating effective clinical intervention strategies. This is crucial for delaying the progression of AD and significantly improving the quality of life of patients.