Machine Learning Approaches for Detecting Diabetic Retinopathy from Clinical and Public Health Records

INTRODUCTION Annual eye examinations are recommended for diabetic patients in order to detect diabetic retinopathy and other eye conditions that arise from diabetes. Medically underserved urban communities in the US have annual screening rates that are much lower than the national average and could benefit from informatics approaches to identify unscreened patients most at risk of developing retinopathy. METHODS Using clinical data from urban safety net clinics as well as public health data from the CDC's National Health and Nutrition Examination Survey, we examined different machine learning approaches for predicting retinopathy from clinical or public health data. All datasets utilized exhibited a class imbalance. RESULTS Classifiers learned on the clinical data were modestly predictive of retinopathy with the best model having an AUC of 0.72, sensitivity of 69.2% and specificity of 55.9%. Classifiers learned on public health data were not predictive of retinopathy. DISCUSSION Successful approaches to detecting latent retinopathy using machine learning could help safety net and other clinics identify unscreened patients who are most at risk of developing retinopathy and the use of ensemble classifiers on clinical data shows promise for this purpose.

[1]  Taghi M. Khoshgoftaar,et al.  RUSBoost: A Hybrid Approach to Alleviating Class Imbalance , 2010, IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans.

[2]  Yoav Freund,et al.  A decision-theoretic generalization of on-line learning and an application to boosting , 1995, EuroCOLT.

[3]  Nathan Congdon,et al.  The Prevalence of Diabetic Retinopathy in the United States , 2002 .

[4]  B. Klein,et al.  Global Prevalence and Major Risk Factors of Diabetic Retinopathy , 2012, Diabetes Care.

[5]  G. Rubin,et al.  Eye care utilization by older Americans: the SEE Project. Salisbury Eye Evaluation. , 1999, Ophthalmology.

[6]  L. Deeb,et al.  Interventions Among Primary-Care Practitioners to Improve Care for Preventable Complications of Diabetes , 1988, Diabetes Care.

[7]  Tien Yin Wong,et al.  Current Epidemiology of Diabetic Retinopathy and Diabetic Macular Edema , 2012, Current Diabetes Reports.

[8]  R. Klein,et al.  Cigarette smoking and ten-year progression of diabetic retinopathy. , 1996, Ophthalmology.

[9]  L. Howie,et al.  Ophthalmic examination among adults with diagnosed diabetes mellitus. , 1993, JAMA.

[10]  Ian H. Witten,et al.  The WEKA data mining software: an update , 2009, SKDD.

[11]  T H Payne,et al.  Preventive Care in Diabetes Mellitus: Current Practice in Urban Health-Care System , 1989, Diabetes Care.

[12]  Xiuqing Guo,et al.  Systemic soluble tumor necrosis factor receptors 1 and 2 are associated with severity of diabetic retinopathy in Hispanics. , 2012, Ophthalmology.

[13]  Romero-Aroca Pedro,et al.  Prevalence and Relationship Between Diabetic Retinopathy and Nephropathy, and its Risk Factors in the North-East of Spain, a Population-Based Study , 2010, Ophthalmic epidemiology.

[14]  R. Tibshirani Regression Shrinkage and Selection via the Lasso , 1996 .

[15]  J Wylie-Rosett,et al.  Ophthalmic referral rates for patients with diabetes in primary-care clinics located in disadvantaged urban communities. , 1995, Journal of diabetes and its complications.

[16]  Omolola Ogunyemi,et al.  Teleretinal Screening for Diabetic Retinopathy in Six Los Angeles Urban Safety-Net Clinics: Final Study Results , 2013, AMIA.

[17]  M. C. Leske,et al.  Patterns of adherence to diabetes vision care guidelines: baseline findings from the Diabetic Retinopathy Awareness Program. , 2001, Ophthalmology.

[18]  I. Mühlhauser,et al.  Cigarette Smoking and Diabetes: An Update , 1994, Diabetic medicine : a journal of the British Diabetic Association.

[19]  Simon Lebech Cichosz,et al.  A classification model for predicting eye disease in newly diagnosed people with type 2 diabetes. , 2015, Diabetes research and clinical practice.

[20]  Omolola Ogunyemi,et al.  Autonomy versus automation: perceptions of nonmydriatic camera choice for teleretinal screening in an urban safety net clinic. , 2013, Telemedicine journal and e-health : the official journal of the American Telemedicine Association.

[21]  Lutz Heinemann,et al.  Smoking Is Associated With Progression of Diabetic Nephropathy , 1994, Diabetes Care.

[22]  Omolola Ogunyemi,et al.  Evaluating Predictive Modeling's Potential to Improve Teleretinal Screening Participation in Urban Safety Net Clinics , 2013, MedInfo.

[23]  Ann L. Albright,et al.  Prevalence of diabetic retinopathy in the United States, 2005-2008. , 2010, JAMA.

[24]  Omolola Ogunyemi,et al.  Teleretinal screening for diabetic retinopathy in six Los Angeles urban safety-net clinics: initial findings. , 2011, AMIA ... Annual Symposium proceedings. AMIA Symposium.

[25]  R. Klein,et al.  Association of Cigarette Smoking With Diabetic Retinopathy , 1991, Diabetes Care.

[26]  Tae Keun Yoo,et al.  Diabetic retinopathy risk prediction for fundus examination using sparse learning: a cross-sectional study , 2013, BMC Medical Informatics and Decision Making.

[27]  S. Garg,et al.  Diabetic Retinopathy Screening Update , 2009, Clinical Diabetes.

[28]  Francisco Herrera,et al.  A Review on Ensembles for the Class Imbalance Problem: Bagging-, Boosting-, and Hybrid-Based Approaches , 2012, IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews).

[29]  R. Klein,et al.  Global Prevalence and Major Risk Factors of Diabetic Retinopathy , 2012, Diabetes Care.

[30]  R. Klein,et al.  Factors associated with having eye examinations in persons with diabetes. , 1995, Archives of family medicine.

[31]  Barbara E K Klein,et al.  Epidemiology of Proliferative Diabetic Retinopathy , 1992, Diabetes Care.

[32]  A. Cavallerano,et al.  Teleretinal Imaging to Screen for Diabetic Retinopathy in the Veterans Health Administration , 2008, Journal of diabetes science and technology.

[33]  Omolola Ogunyemi,et al.  Workflow concerns and workarounds of readers in an urban safety net teleretinal screening study. , 2011, AMIA ... Annual Symposium proceedings. AMIA Symposium.