Dataset Information

Parametric model fitting-based approach for retinal blood vessel caliber estimation in eye fundus images.

ABSTRACT: Changes in the retinal vessel caliber are associated with a variety of major diseases, namely diabetes, hypertension and atherosclerosis. The clinical assessment of these changes in fundus images is tiresome and prone to errors and thus automatic methods are desirable for objective and precise caliber measurement. However, the variability of blood vessel appearance, image quality and resolution make the development of these tools a non-trivial task.A method for the estimation of vessel caliber in eye fundus images via vessel cross-sectional intensity profile model fitting is herein proposed. First, the vessel centerlines are determined and individual segments are extracted and smoothed by spline approximation. Then, the corresponding cross-sectional intensity profiles are determined, post-processed and ultimately fitted by newly proposed parametric models. These models are based on Difference-of-Gaussians (DoG) curves modified through a multiplying line with varying inclination. With this, the proposed models can describe profile asymmetry, allowing a good adjustment to the most difficult profiles, namely those showing central light reflex. Finally, the parameters of the best-fit model are used to determine the vessel width using ensembles of bagged regression trees with random feature selection.The performance of our approach is evaluated on the REVIEW public dataset by comparing the vessel cross-sectional profile fitting of the proposed modified DoG models with 7 and 8 parameters against a Hermite model with 6 parameters. Results on different goodness of fitness metrics indicate that our models are constantly better at fitting the vessel profiles. Furthermore, our width measurement algorithm achieves a precision close to the observers, outperforming state-of-the art methods, and retrieving the highest precision when evaluated using cross-validation. This high performance supports the robustness of the algorithm and validates its use in retinal vessel width measurement and possible integration in a system for retinal vasculature assessment.

SUBMITTER: Araujo T

PROVIDER: S-EPMC5905988 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:PurposeTo compare three commonly used retinal vessel caliber measurement software systems, and propose an algorithm for conversion between measurement systems.MethodsWe used 120 retinal photographs to evaluate the agreement between three commonly used software (Retinal Analysis [RA], Integrative Vessel Analysis [IVAN], and Singapore I Vessel Assessment [SIVA]). Bland-Altman plots were used to evaluate agreement of retinal arteriolar (central retinal artery equivalent, CRAE) and venular (central retinal vein equivalent, CRVE) calibers. Pearson's correlation was used to assess the associations between systemic factors and retinal vessel calibers, and Z-test was used to compare the strength of the correlation coefficients across the three software systems. An algorithm was created to convert measurements, with paired t-test performed to evaluate the differences between SIVA-measured retinal calibers and SIVA-approximates converted from RA- and IVAN-measurements using the algorithm.ResultsDifferences between SIVA- and RA-measured calibers (CRAE: mean difference [MD] = -21.8 ?m, 95% limits of agreement [LOA], -47.3 to 3.7 ?m; CRVE: MD = -7.7 ?m, 95% LOA, -28.0 to 12.6 ?m), SIVA- and IVAN-measured calibers (CRAE: MD = -6.7 ?m, 95% LOA, -23.8 to 10.4 ?m; CRVE: MD = -18.2 ?m 95% LOA, -36.7 to 0.4 ?m) were large. However, the strength of correlations between systemic factors with SIVA-measured retinal calibers was not significantly different to that measured using RA and IVAN (P ? 0.332). SIVA-approximates converted from RA and IVAN measurements using the proposed algorithm was not significantly different from SIVA-measured calibers (P ? 0.20).ConclusionAbsolute measurements of retinal vessel calibers vary between three common software systems but associations with systemic factors were similar.Translational relevanceThe proposed algorithm allowed conversions of RA and IVAN measurements to SIVA-approximates. This conversion is important for future data pooling and establishment of normative values for retinal vascular caliber measurements.

Dataset Information

Parametric model fitting-based approach for retinal blood vessel caliber estimation in eye fundus images.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure