Your browser doesn't support javascript.
A naked eye 3D display and interaction system for medical education and training.
J Biomed Inform ; 100: 103319, 2019 Dec.
Artigo em Inglês | MEDLINE | ID: mdl-31655272
ABSTRACT
To provide natural simulated objects and intuitive user interaction in medical education and training, we propose a naked eye 3D display and interaction system. The current 3D rendering algorithms for naked eye 3D displays are not suitable for medical use, due to the requirements of displaying and interacting with high quality medical images and simulating soft tissues. Because the traditional 3D rendering procedure and vertex indexing in collision detection require substantial computing power when using a naked eye 3D display, the current method cannot achieve fluent displays and interactions. Thus, we develop a novel octree-based 3D rendering and interaction algorithm for high quality medical models to improve the rendering rate and obtain smooth human machine interactions when using the naked eye 3D display device. We also valuate the soft-body phantom simulation of the naked eye 3D display device by combining the traditional 3D rendering algorithm with the elastic 3D simulation to simulate deformable tissues. We integrate an incremental interaction method and a Kalman filter-based hand tracking method to achieve a larger user interaction range and robust hand tracking. We used the proposed system to perform human-computer interactions with rigid phantoms and soft-body phantoms. The experimental results showed that the proposed rendering algorithm for rigid phantoms could achieve higher rendering performance (50 FPS) than the traditional rendering algorithm (9.8 FPS). The user experiments showed that the 3D simulation system equipped with the enhanced rendering algorithm could achieve fluent interactions when using the naked eye 3D display, thus promoting education experiences and reducing task completion times.

Similares

MEDLINE

...
LILACS

LIS

Texto completo: Disponível Coleções: Bases de dados internacionais Base de dados: MEDLINE Idioma: Inglês Revista: J Biomed Inform Assunto da revista: Informática Médica Ano de publicação: 2019 Tipo de documento: Artigo País de afiliação: China