Automatic Image Analyzer to Assess Retinal Vessel Caliber

Overview

The objective of this work is the development and validation (reliability and validity) of the Automatic image analyzer to assess it retinal vessel caliber (ALTAIR) platform to analyze and validate its utility in different clinical settings. Methods / Design Design : In the first phase , will be a cross- sectional study and a second phase will be a prospective observational study with annual follow-up for 4 years. The study will be conducted in primary care including 386 subjects. Main measurements: blood pressure, carotid intima-media thickness using carotid ultrasound , central blood pressure and wave velocity pulse by SphygmoCor system, Cardio-ankle vascular index (CAVI) by VASERA® VS1500 , evaluation of hypertrophy ventricular using a digital ECG and renal injury . Retinal vascular evaluation will be done by a non-mydriatic TOPCON TRCNW200 getting focused papilla, nasal and temporal and developed by software images automatically calculate the thickness of the retinal vessels , the arteriovenous index, the vascularized area and branching pattern . For validation software retina intra and inter - observer variability , the correlation of the tool measurement with AVindex , and concurrent validity with parameters of vascular structure and function and target organ damage and association analysis will analyze the different parameters of the estimated retina and evolution or appearance of new lesions in the target organs. Discussion The validation of the tool contribute to the analysis of retinal vessels improved reliability by reducing the intervention of the observer and further validate the use of more information than the previously used , especially surface vascularization and vessel branching patterns .

Design and setting In the first phase, will be an observational descriptive study, with the primary objective of validating the instrument developed. Later, in a second phase will be a prospective observational study, with annual follow-up for 4 years. The study will be conducted in the field of primary care. The study population will be subject 30 to 75 years with a cardiovascular risk factor defined by the European Hypertension Guide 2013 and does not meet any of the defined exclusion criteria: patients who can not comply with the requirements of the protocol due to mental and / or cognitive impairment, no partners, educational and understanding limitations and severe comorbidity that would endanger his life in the next 12 months. By consecutive sampling of all patients referred to the research unit for assessment of cardiovascular risk, their participation is requested from those who meet the inclusion criteria and have no reason for exclusion to the size of the estimated sample. Sample size was estimated to detect a correlation coefficient between velocity and the pulse wave parameters retinal 0.15, alpha risk of 0.05 and a beta risk of 0.20, and an estimated loss by difficulty technical or abandonment of monitoring 10%, so 386 subjects, who will be included in the study required. Software Validation retina For validation of the retina software Follow these steps after a previous training of evaluators who will make the evaluation of the images . 1 - intra- observer variability . Assessing the repeatability of the measurement, an operator must measure the same picture in the same individual, at least twice. To do this, an operator will evaluate 100 images of a random subsample of 50 patients with a gap of one week between the two embodiments . In this case , the operator and the images analyzed are the same on both days and will not have information on the previous assessment . 2 - inter-observer variability . To assess the reproducibility of the measurement system , a new operator, different from phase 1, evaluated the same 100 images previously analyzed . This operator will ignore the results that have been obtained in the previous phase. The two operators have the same experience in the field and in the use of software , in addition, both receive the same previous training. 3.- And with retina developed software program provides the results on the evaluation of arterial / venous , thickness of the vein and the artery index. To assess the degree of agreement between the two methodologies (AVindex ) valuation of 100 images were made with the two tools . In this way, we can show that the new method and provides the same results improves daily practice having more objective and faster processing of results. 4.- The validity of the measurement was analyzed in the total sample, 386 and 772 subjects fundus , analyzing the relationship of the results to the carotid intima-media thickness as a measure of vascular structure , the speed of the pulse wave as measure of vascular function and arterial stiffness gold standard (arterial stiffness ) in Cardio Ankle vascular Index , renal function , and electrocardiographic parameters estimated with the Framingham cardiovascular risk. 5.- also the association of the different parameters of the estimated retina and evolution or appearance of new lesions in target organs and cardiovascular monitoring of patients during the four years of the second phase of this project will be analyzed events