Dataset Information

Interpretable machine learning-derived nomogram model for early detection of diabetic retinopathy in type 2 diabetes mellitus: a widely targeted metabolomics study.

ABSTRACT:

Objective

Early identification of diabetic retinopathy (DR) is key to prioritizing therapy and preventing permanent blindness. This study aims to propose a machine learning model for DR early diagnosis using metabolomics and clinical indicators.

Methods

From 2017 to 2018, 950 participants were enrolled from two affiliated hospitals of Wenzhou Medical University and Anhui Medical University. A total of 69 matched blocks including healthy volunteers, type 2 diabetes, and DR patients were obtained from a propensity score matching-based metabolomics study. UPLC-ESI-MS/MS system was utilized for serum metabolic fingerprint data. CART decision trees (DT) were used to identify the potential biomarkers. Finally, the nomogram model was developed using the multivariable conditional logistic regression models. The calibration curve, Hosmer-Lemeshow test, receiver operating characteristic curve, and decision curve analysis were applied to evaluate the performance of this predictive model.

Results

The mean age of enrolled subjects was 56.7 years with a standard deviation of 9.2, and 61.4% were males. Based on the DT model, 2-pyrrolidone completely separated healthy controls from diabetic patients, and thiamine triphosphate (ThTP) might be a principal metabolite for DR detection. The developed nomogram model (including diabetes duration, systolic blood pressure and ThTP) shows an excellent quality of classification, with AUCs (95% CI) of 0.99 (0.97-1.00) and 0.99 (0.95-1.00) in training and testing sets, respectively. Furthermore, the predictive model also has a reasonable degree of calibration.

Conclusions

The nomogram presents an accurate and favorable prediction for DR detection. Further research with larger study populations is needed to confirm our findings.

SUBMITTER: Li J

PROVIDER: S-EPMC9355962 | biostudies-literature | 2022 Aug

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Interpretable machine learning-derived nomogram model for early detection of diabetic retinopathy in type 2 diabetes mellitus: a widely targeted metabolomics study.

Li Jushuang J Guo Chengnan C Wang Tao T Xu Yixi Y Peng Fang F Zhao Shuzhen S Li Huihui H Jin Dongzhen D Xia Zhezheng Z Che Mingzhu M Zuo Jingjing J Zheng Chao C Hu Honglin H Mao Guangyun G

Nutrition & diabetes 20220805 1

<h4>Objective</h4>Early identification of diabetic retinopathy (DR) is key to prioritizing therapy and preventing permanent blindness. This study aims to propose a machine learning model for DR early diagnosis using metabolomics and clinical indicators.<h4>Methods</h4>From 2017 to 2018, 950 participants were enrolled from two affiliated hospitals of Wenzhou Medical University and Anhui Medical University. A total of 69 matched blocks including healthy volunteers, type 2 diabetes, and DR patients ...[more]

PMID: 35931671

Similar Datasets

Project description:BackgroundThe proportion of patients with diabetic retinopathy (DR) has grown with increasing number of diabetes mellitus patients in the world. It is among the major causes of blindness worldwide. The main objective of this study was to identify contributing risk factors of DR among people with type II diabetes mellitus.MethodA sample of 191 people with type II diabetes mellitus was selected from the Black Lion Specialized Hospital diabetic unit from 1 March 2018 to 1 April 2018. A multivariate stochastic regression imputation technique was applied to impute the missing values. The response variable, DR is a categorical variable with two outcomes. Based on the relationship derived from the exploratory analysis, the odds of having DR were not necessarily linearly related to the continuous predictors for this sample of patients. Therefore, a semiparametric model was proposed to identify the risk factors of DR.ResultFrom the sample of 191 people with type II diabetes mellitus, 98 (51.3%) of them had DR. The results of semiparametric regression model revealed that being male, hypertension, insulin treatment, and frequency of clinical visits had a significant linear relationships with the odds of having DR. In addition, the log- odds of having DR has a significant nonlinear relation with the interaction of age by gender (for female patients), duration of diabetes, interaction of cholesterol level by gender (for female patients), haemoglobin A1c, and interaction of haemoglobin A1c by fasting blood glucose with degrees of freedom [Formula: see text], respectively. The interaction of age by gender and cholesterol level by gender appear non significant for male patients. The result from the interaction of haemoglobin A1c (HbA1c) by fasting blood glucose (FBG) showed that the risk of DR is high when the level of HbA1c and FBG were simultaneously high.ConclusionClinical variables related to people with type II diabetes mellitus were strong predictive factors of DR. Hence, health professionals should be cautious about the possible nonlinear effects of clinical variables, interaction of clinical variables, and interaction of clinical variables with sociodemographic variables on the log odds of having DR. Furthermore, to improve intervention strategies similar studies should be conducted across the country.

Dataset Information

Interpretable machine learning-derived nomogram model for early detection of diabetic retinopathy in type 2 diabetes mellitus: a widely targeted metabolomics study.

Objective

Methods

Results

Conclusions

Publications

Interpretable machine learning-derived nomogram model for early detection of diabetic retinopathy in type 2 diabetes mellitus: a widely targeted metabolomics study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets