Diagnostic medical ultrasound requires psychomotor skill in image acquisition and cognitive skill in image interpretation. There is a lack of an objective, quantitative, and standardized method to assess psychomotor skill. In traditional radiology and cardiology training programs, physicians are certified based on duration of training and number of procedures performed. However, assessment of trainee skill is now mandated by the worldwide shift toward competencebased medical education. The benefits of skill assessment are: 1) it provides trainees with feedback on their progress, 2) faculty can identify which trainees require additional help, and 3) hospitals have a standard for credentialing.

The invention described here is an ultrasound simulator that measures psychomotor (manual or technical) skill in acquiring medical ultrasound images in addition to cognitive skill in image interpretation; the simulator can also be employed for training. The simulator’s hardware comprises a mannequin, a mock transducer with tracking device, and a computer. The simulator’s software displays two-dimensional (2D) ultrasound images cut from 3D images in an Image Library in correspondence with the position and orientation of the mock transducer as the testee manipulates it over the mannequin. When the testee “acquires” an image, a 2D image is saved for scoring and review. The simulator’s Image Library has cases prepared from previously recorded 3D images of live patients with various diagnoses.

Psychomotor skill is measured in terms of the angle of deviation between the plane of the anatomically correct view and the plane of the image acquired by the user. The anatomically correct view is defined from the plane specified from the anatomic landmarks of the specified view of the target anatomy. For example, the apical four-chamber view of the heart is defined as the plane that contains the apex of the left ventricle, the centroid of the mitral annulus, and the centroid of the tricuspid annulus. Definition of the anatomically correct view plane is enabled by 3D reconstruction using the Piecewise Smooth Subdivision Surface Method, because this method carries anatomic labeling through the reconstruction process to every vertex, edge, and face on the 3D surface. The distance between the plane of the anatomically defined plane and the plane of the image acquired by the user is also calculated based upon a defined anatomic structure in each plane such as, for the heart, the centroid of the left ventricle.

The simulator employs Visual Guidance, a graphic display of the 3D reconstruction of the anatomy being imaged. Visual Guidance displays the position and orientation of the planes of the acquired and correct views to provide learners with immediate feedback during practice. Visual Guidance is disabled during testing, but feedback can still be provided after the learner has completed the case study and submitted his/her an swers, so that the test can still be a learning experience.

This work was done by Florence Sheehan, Edward Bolson, Catherine Otto, and Mark Anderson of the University of Washington for Johnson Space Center. MSC-25165-1