Cardiovascular risk prediction based on Retinal Vessel Analysis using machine learning

Cardiovascular risk prediction is a vital aspect of personalized health care. In this study, retinal vascular function is assessed in asymptomatic participants who are classified into risk groups based on Framingham Risk Score. Feature selection, oversampling and state-of-the-art classification methods are applied to provide a sound individual risk prediction based on Retinal Vessel Analysis (RVA) data obtained by non-invasive methods. The results indicate that the RVA based cardiovascular risk prediction models are competitive with well established Framingham and Qrisk based models.

[1]  Igor Kononenko,et al.  Estimating Attributes: Analysis and Extensions of RELIEF , 1994, ECML.

[2]  Ž. Reiner,et al.  Relation of atherosclerotic changes in retinal arteries to the extent of coronary artery disease. , 2005, The American journal of cardiology.

[3]  Stephen R. Alty,et al.  Predicting Arterial Stiffness From the Digital Volume Pulse Waveform , 2007, IEEE Transactions on Biomedical Engineering.

[4]  R. Klein,et al.  Retinal vessel diameter and cardiovascular mortality: pooled data analysis from two older populations. , 2007, European heart journal.

[5]  K. Kotseva,et al.  EUROASPIRE III: a survey on the lifestyle, risk factors and use of cardioprotective drug therapies in coronary patients from 22 European countries , 2009, European journal of cardiovascular prevention and rehabilitation : official journal of the European Society of Cardiology, Working Groups on Epidemiology & Prevention and Cardiac Rehabilitation and Exercise Physiology.

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

[7]  C. Lau,et al.  Abnormal Vascular Function in PR‐Interval Prolongation , 2011, Clinical cardiology.

[8]  Constantinos S. Pattichis,et al.  Assessment of the Risk Factors of Coronary Heart Events Based on Data Mining With Decision Trees , 2010, IEEE Transactions on Information Technology in Biomedicine.

[9]  A. Sheikh,et al.  Predicting cardiovascular risk in England and Wales: prospective derivation and validation of QRISK2 , 2008, BMJ : British Medical Journal.

[10]  A. Ekárt,et al.  Coexistence of macro‐ and micro‐vascular abnormalities in newly diagnosed normal tension glaucoma patients , 2012, Acta ophthalmologica.

[11]  J. Shaw,et al.  Correlation of light-flicker-induced retinal vasodilation and retinal vascular caliber measurements in diabetes. , 2009, Investigative ophthalmology & visual science.

[12]  Multiscale regularity analysis of the Heart Rate Variability: stratification of cardiac death risk , 2007, 2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[13]  A. Ekárt,et al.  Ageing effect on flicker‐induced diameter changes in retinal microvessels of healthy individuals , 2015, Acta ophthalmologica.

[14]  Nitesh V. Chawla,et al.  SMOTE: Synthetic Minority Over-sampling Technique , 2002, J. Artif. Intell. Res..

[15]  R. Guthke,et al.  Generation of rules for expert systems by statistical methods of fermentation data analysis , 1994 .

[16]  R. Klein,et al.  Retinal Microvascular Signs and Risk of Stroke: The Multi-Ethnic Study of Atherosclerosis (MESA) , 2012, Stroke.

[17]  N John Bosomworth,et al.  Practical use of the Framingham risk score in primary prevention: Canadian perspective. , 2011, Canadian family physician Medecin de famille canadien.

[18]  Timo Slawinski,et al.  Test- and Rating Strategies for Data Based Rule Generation , 1998 .

[19]  Helmut Alt,et al.  Computing the Fréchet distance between two polygonal curves , 1995, Int. J. Comput. Geom. Appl..

[20]  Juan F. Ramirez-Villegas,et al.  Heart Rate Variability Dynamics for the Prognosis of Cardiovascular Risk , 2011, PloS one.

[21]  B.-U. Seifert,et al.  RETINAL VESSEL ANALYZER (RVA) - DESIGN AND FUNCTION , 2002, Biomedizinische Technik. Biomedical engineering.

[22]  Haibo He,et al.  ADASYN: Adaptive synthetic sampling approach for imbalanced learning , 2008, 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence).

[23]  Ines Lanzl,et al.  Age, blood pressure, and vessel diameter as factors influencing the arterial retinal flicker response. , 2004, Investigative ophthalmology & visual science.

[24]  JOSE F. VALENCIA,et al.  Using Entropy Rates to Improve Risk Stratification to Predict Cardiac Mortality , .

[25]  P. Mitchell,et al.  Retinal Vascular Imaging: A New Tool in Microvascular Disease Research , 2008, Circulation. Cardiovascular imaging.

[26]  Pat Langley,et al.  Selection of Relevant Features and Examples in Machine Learning , 1997, Artif. Intell..

[27]  M. Pencina,et al.  General Cardiovascular Risk Profile for Use in Primary Care: The Framingham Heart Study , 2008, Circulation.

[28]  B. Norrving,et al.  Global atlas on cardiovascular disease prevention and control. , 2011 .

[29]  F. Massey.  The Kolmogorov-Smirnov Test for Goodness of Fit , 1951 .

[30]  J. Perkiömäki.  Heart Rate Variability and Non-Linear Dynamics in Risk Stratification , 2011, Front. Physio..

[31]  Shyr-Shen Yu,et al.  Data Mining for Bioinformatics: Design with Oversampling and Performance Evaluation , 2015 .

[32]  A. Hofman,et al.  Retinal vessel diameters and risk of stroke , 2006, Neurology.

[33]  Illhoi Yoo,et al.  Data Mining in Healthcare and Biomedicine: A Survey of the Literature , 2012, Journal of Medical Systems.

[34]  M Pfaff,et al.  Prediction of cardiovascular risk in hemodialysis patients by data mining. , 2004, Methods of information in medicine.

[35]  Ian H. Witten,et al.  Generating Accurate Rule Sets Without Global Optimization , 1998, ICML.

[36]  J. Flammer,et al.  The eye and the heart , 2013, European heart journal.

[37]  Ronald Klein,et al.  Changes in retinal vessel diameter and incidence and progression of diabetic retinopathy. , 2012, Archives of ophthalmology.