There are existing methods for simulating ultrasound images based on transducer parameters and digital 3D tissue models. These simulation models are useful for designing transducers, training, and inter-modality image registration. However, validating the accuracy of ultrasound images produced by simulation remains difficult due to lack of standardized data sets in which both volumetric medium properties and corresponding empirical ultrasound images are available.
The work presented on this page was done in the process of developing such an experimental data set. This required the development of a suite of tools to assist in capturing raw, spatially-tracked ultrasound data from a standardised model, with corresponding cross-modality (CT, MRI) data. This work was done in 2012 in the context of Engineering Honours.
DataUS is a spatially-tracked ultrasound data acquisition tool. It coordinates synchronised data acquisition from an ultrasound device, a magnetic 3D spatial tracking system, and a hardware data acquisition module that provides access to raw ultrasound channel data.
It allows custom acquisition sequences and data processing to be configured according the needs of the study.
LibUltrasound was an object oriented C++ library to configure and receive data from an Ultrasound system and a special data acquisition module. This library was built to support the development of DataUS.
LibTracker was an object oriented C++ library to configure and receive data from with a magnetic 3D spatial tracking hardware solution. This library was built to support the development of DataUS.
The DataUS tool captures experimental data in the following format.
DataUS data is used to validate simulation models in the following way.
