Student perceptions of photogrammetric 3D cardiac models generated by use of a smartphone and high-resolution camera.

Abstract

Digital anatomy specimens are increasingly being produced using photogrammetric modelling, but student perceptions of their usability and fidelity are infrequently evaluated. This study used a high-resolution digital camera (hrDC) and smartphone camera to create photogrammetric models of a porcine heart and evaluated their usability (primary aim) and visual fidelity (secondary aim). Student evaluation (n = 52) of prototype models using a Likert-type questionnaire was undertaken. Results demonstrated that both smartphone and hrDC-based photogrammetry produced highly usable models (median usability statement agreement = 5/5 for both hrDC and smartphone models; p = 0.325). The appearance of the smartphone-based model was significantly inferior to those produced by the high-resolution camera on multiple domains including colour (median hrDC agreement = 5/5, median smartphone agreement = 4/5; p < 0.05), texture (median hrDC agreement = 5/5, median smartphone agreement = 3/5; p < 0.05), and clarity of internal structures (median hrDC agreement = 5/5, median smartphone agreement = 4/5; p < 0.05). Specific features on the smartphone-based model were less easily identified, including the pulmonary valve (p < 0.05) and papillary muscle (p < 0.05). These data indicate that a usable anatomical model can be created with smartphone images but is of a lower fidelity when compared to those produced by a high-resolution digital camera.