Multiscale AM-FM Methods for Diabetic Retinopathy Lesion Detection

By C. Agurto, Víctor Murray, E. Barriga, S. Murillo, Marios S. Pattichis, H. Davis, S. Russell, Michael David Abràmoff, Peter Soliz. 2010

303 citations recorded by the source corpus.

In this paper, we propose the use of multiscale amplitude-modulation-frequency-modulation (AM-FM) methods for discriminating between normal and pathological retinal images. The method presented in this paper is tested using standard images from the early treatment diabetic retinopathy study. We use 120 regions of 40 × 40 pixels containing four types of lesions commonly associated with diabetic retinopathy (DR) and two types of normal retinal regions that were manually selected by a trained analyst. The region types included microaneurysms, exudates, neovascularization on the retina, hemorrhages, normal retinal background, and normal vessels patterns. The cumulative distribution functions of the instantaneous amplitude, the instantaneous frequency magnitude, and the relative instantaneous frequency angle from multiple scales are used as texture feature vectors. We use distance metrics between the extracted feature vectors to measure interstructure similarity. Our results demonstrate a statistical differentiation of normal retinal structures and pathological lesions based on AM-FM features. We further demonstrate our AM-FM methodology by applying it to classification of retinal images from the MESSIDOR database. Overall, the proposed methodology shows significant capability for use in automatic DR screening.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

What do we need to build explainable AI systems for the medical domain? (2017, methodology): “The introduction of a multiscale approach in (Murray et al., 2010) demonstrated that the method can be used to reconstruct general images.”
Tree Image Growth Analysis Using Instantaneous Phase Modulation (2011, unknown): “Tree Image Growth Analysis Using Instantaneous”
DREAM: Diabetic Retinopathy Analysis Using Machine Learning (2014, methodology): “Another method based on the extraction of AM–FM features of fundus images, followed by dimensionality reduction and hierarchical clustering, and the use of partial least-squares method to assign DR severity grade has been presented in [8].”
Weakly-supervised localization of diabetic retinopathy lesions in retinal fundus images (2017, background): “A lot of research has been conducted to detect specific lesion areas, like blood vessel transformations, exudates, microaneurysms and hemorrhages [10, 11, 12, 13, 14].”
Automatic Diabetic Macular Edema Detection in Fundus Images Using Publicly Available Datasets (2011, methodology): “In another recent study, Agurto et al. (2010) diagnose diabetic retinopathy avoiding the usage of manually segmented lesion too, by employing AM-FM features.”
IDRiD: Diabetic Retinopathy - Segmentation and Grading Challenge. (2020, methodology): “Further, several approaches were devised for multiple lesion detection such as multiscale amplitude-modulation-frequency-modulation ( Agurto et al., 2010 ), machine learning ( Roychowdhury et al., 2014 ), a combination of Hessian multiscale analysis, variational segmentation and texture features (…”
Multiscale Amplitude-Modulation Frequency-Modulation (AM–FM) Texture Analysis of Ultrasound Images of the Intima and Media Layers of the Carotid Artery (2011, methodology): “For these cases, the user of the proposed system may run the snakes initialization procedure again in order to estimate the correct initial snake contour, and during the snakes deformation the user may interact and manually correct the contour [5], [25].”