This study aims to analyze the protein aggregates spatial distribution for different cataract degrees, and correlate this information with the lens acoustical parameters and by this way, assess the cataract regional hardness. Different cataract degrees were induced ex vivo in porcine lenses. A 25 MHz ultrasonic transducer was used to obtain the acoustical parameters (velocity, attenuation, and backscattering signals). B-scan and Nakagami images were constructed. Also, lenses with different cataract degrees were sliced in two regions (nucleus and cortex), for fibers and collagen detection. A significant increase with cataract formation was found for the velocity, attenuation, and brightness intensity of the B-scan images and Nakagami m parameter ( ). The acoustical parameters showed a good to moderate correlation with the m parameter for the different stages of cataract formation. A strong correlation was found between the protein aggregates in the cortex and the m parameter. Lenses without cataract are characterized using a classification and regression tree, by a mean brightness intensity ≤0.351, a variance of the B-scan brightness intensity ≤0.070, a velocity ≤1625 m/s, and an attenuation ≤0.415 dB/mm·MHz (sensitivity: 100% and specificity: 72.6%). To characterize different cataract degrees, the m parameter should be considered. Initial stages of cataract are characterized by a mean brightness intensity >0.351 and a variance of the m parameter >0.110. Advanced stages of cataract are characterized by a mean brightness intensity >0.351, a variance of the m parameter ≤0.110, and a mean m parameter >0.374. For initial and advanced stages of cataract, a sensitivity of 78.4% and a specificity of 86.5% are obtained.

This paper addresses the use of computer-aided diagnosis (CAD) system for the cataract classification based on ultrasound technique. Ultrasound A-scan signals were acquired in 220 porcine lenses. B-mode and Nakagami images were constructed. Ninety-seven parameters were extracted from acoustical, spectral and image textural analyses and were subjected to feature selection by Principal Component Analysis (PCA). Bayes, K Nearest-Neighbors (KNN), Fisher Linear Discriminant (FLD) and Support Vector Machine (SVM) classifiers were tested. The classification of healthy and cataractous lenses shows a good performance for the four classifiers (F-measure ≥92.68%) with SVM showing the highest performance (90.62%) for initial versus severe cataract classification.

No abstract

No abstract

Cataract is a clouding or opacity of the normally transparent crystalline lens of the eye. The cataract formation is associated with the increase of both inner fiber compaction and protein aggregation, which can be characterized by ultrasound backscattering. In this study, the tissue scatterers changing with cataract formation was investigated, and their influence in the frequency dependent attenuation such as in the Nakagami distribution was analyzed. For this purpose, cataracts were induced in twenty porcine lenses. A 25 MHz focused transducer was used to estimate the ultrasound attenuation considering the spectral ratio between echo signals from a reflector with and without lenses inserted. A power-law frequency dependence model was used to study the frequency dependent attenuation. The analyzed signals showed high backscattering and also a variation of the Nakagami parameter with cataract formation, indicating a scatter size increase. This conclusion could be important to assess the cataract hardness and to provide the correct information about its type and severity.