Inclusion mechanical property estimation using tactile images, finite element method, and artificial neural network

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Materials Science

Background:

• Accurate characterization of subsurface tissue inclusions is crucial for medical diagnosis.
• Traditional palpation methods lack quantitative precision for mechanical properties like Young's modulus.
• Tactile sensation imaging offers a promising, non-invasive approach for tissue property estimation.

Purpose of the Study:

• To develop and validate a methodology for estimating the size, depth, and Young's modulus of tissue inclusions.
• To utilize tactile data and advanced algorithms for precise mechanical property quantification.
• To provide a quantitative tool for analyzing tissue inclusion characteristics.

Main Methods:

• Development of a tactile sensation imaging system to acquire surface data.
• Implementation of a forward algorithm using finite element method (FEM) to model tactile responses.
• Creation of an artificial neural network (ANN)-based inversion algorithm to extract inclusion parameters from tactile images.

Main Results:

• The developed methodology successfully estimates tissue inclusion size, depth, and Young's modulus.
• Validation using custom tissue phantoms demonstrated the method's accuracy.
• Achieved root mean squared errors of 1.25 mm for size, 2.09 mm for depth, and 28.65 kPa for Young's modulus.

Conclusions:

• The proposed tactile imaging and computational approach provides accurate estimation of key tissue inclusion parameters.
• This method holds potential for improving diagnostic capabilities in medical applications, particularly for breast tissue analysis.
• The combination of FEM and ANN offers a robust framework for quantitative tactile analysis.