Modelling of an ultrasound-based system for cataract detection and classification

Abstract

Cataract is a cloudiness in the lens of the eye affecting its transparency. Surgical cataract removal is the only therapeutic approach. Phacoemulsification is the most common surgical procedure, in which ultrasound energy is used to emulsify the cataractous lens. The hardness of the cataract determines the optimal energy for the phacoemulsification procedure. The objective of this work is to model an ultrasoundbased system for cataract detection and classification. It was developed a computational tool for simulating the propagation of A-scan ultrasound in the eye, using the k-Wave MATLAB toolbox. Considering the computational resources available and the dimensions of cataractous microstructures, it was determined a suitable resolution that enables the simulation of ultrasound propagation through the cataractous eye in 2D and 3D. Being so, a pixel size of 8 µm was stablished. A matrix representing the eye structures was constructed in MATLAB. Since on simulations it was considered a focalized probe only the central part of the eye matrix was used. It was observed that 2D and 3D simulated signals present some differences, mainly in the amplitudes of the echoes corresponding to the interfaces of the eye. The chosen resolution was suitable for the purpose of this work. The simulation tool shows good versatility for evaluating diverse aspects of A-scan biometry.