The automatic detection of diabetic kidney disease from retinal vascular parameters combined with clinical variables using artificial intelligence in type-2 diabetes patients - PubMed (original) (raw)

The automatic detection of diabetic kidney disease from retinal vascular parameters combined with clinical variables using artificial intelligence in type-2 diabetes patients

Shaomin Shi et al. BMC Med Inform Decis Mak. 2023.

Abstract

Background: Diabetic kidney disease (DKD) has become the largest cause of end-stage kidney disease. Early and accurate detection of DKD is beneficial for patients. The present detection depends on the measurement of albuminuria or the estimated glomerular filtration rate, which is invasive and not optimal; therefore, new detection tools are urgently needed. Meanwhile, a close relationship between diabetic retinopathy and DKD has been reported; thus, we aimed to develop a novel detection algorithm for DKD using artificial intelligence technology based on retinal vascular parameters combined with several easily available clinical parameters in patients with type-2 diabetes.

Methods: A total of 515 consecutive patients with type-2 diabetes mellitus from Xiangyang Central Hospital were included. Patients were stratified by DKD diagnosis and split randomly into either the training set (70%, N = 360) or the testing set (30%, N = 155) (random seed = 1). Data from the training set were used to develop the machine learning algorithm (MLA), while those from the testing set were used to validate the MLA. Model performances were evaluated.

Results: The MLA using the random forest classifier presented optimal performance compared with other classifiers. When validated, the accuracy, sensitivity, specificity, F1 score, and AUC for the optimal model were 84.5%(95% CI 83.3-85.7), 84.5%(82.3-86.7), 84.5%(82.7-86.3), 0.845(0.831-0.859), and 0.914(0.903-0.925), respectively.

Conclusions: A new machine learning algorithm for DKD diagnosis based on fundus images and 8 easily available clinical parameters was developed, which indicated that retinal vascular changes can assist in DKD screening and detection.

Keywords: Artificial intelligence; Diabetic kidney disease; Diabetic retinopathy; Fundus photography; Type 2 diabetes.

© 2023. The Author(s).

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1

Fig. 1

The workflow for developing machine learning models to detect diabetic kidney disease in this study

Fig. 2

Fig. 2

ROC curve of the best model using Random Forest classifier with SMOTE correction for data set imbalance in validation

Fig. 3

Fig. 3

Relative variable importance for the accuracy of detecting diabetic kidney disease using the random Forest classifier with SMOTE correction for data set imbalance. Abbreviations: BMI indicates body mass index; HbA1c, glycosylated hemoglobin; NVArea, non-vascular area; Tor-All, total vessel tortuosity; FD-All, total fractal dimension; Width-PA, peripheral arterial width; Width-PV, the peripheral vein width; History, history of cardiovascular and cerebrovascular disease (myocardial infarction, angina, heart failure or stroke); SMOTE, synthetic minority over-sampling technique

References

    1. Xu X, Sun F, Wang Q, et al. Comprehensive retinal vascular measurements: a novel association with renal function in type 2 diabetic patients in China. Sci Rep. 2020;10:13737. doi: 10.1038/s41598-020-70408-0. - DOI - PMC - PubMed
    1. Khitan Z, Nath T, Santhanam P. Machine learning approach to predicting albuminuria in persons with type 2 diabetes: an analysis of the LOOK AHEAD Cohort. J Clin Hypertens (Greenwich) 2021;23:2137–2145. doi: 10.1111/jch.14397. - DOI - PMC - PubMed
    1. Shi S, Ni L, Gao L, Wu X. Comparison of Nonalbuminuric and Albuminuric Diabetic Kidney Disease Among Patients With Type 2 Diabetes: A Systematic Review and Meta-Analysis. Front Endocrinol. 2022;13:871272. doi: 10.3389/fendo.2022.871272. - DOI - PMC - PubMed
    1. Hayashi Y. Detection of Lower Albuminuria Levels and Early Development of Diabetic Kidney Disease Using an Artificial Intelligence-Based Rule Extraction Approach. Diagnostics (Basel, Switzerland). 2019;9(4):133. - PMC - PubMed
    1. UcgulAtilgan C, Atilgan KG, Kosekahya P, et al. Retinal microcirculation alterations in Microalbuminuric diabetic patients with and without retinopathy. Seminars Ophthalmology. 2021;36:406–412. doi: 10.1080/08820538.2021.1896745. - DOI - PubMed

MeSH terms

LinkOut - more resources