Developing digital tissue phantoms for hyperspectral imaging of ischemic wounds
David W. Allen, Jeeseong Hwang, Maritoni A. Litorja, Joseph P. Rice
Biomedical optical imaging has the potential to achieve high spatial resolution and high functional sensitivity for non-invasive assessment of ischemic wounds. However, clinical acceptance of many optical imaging devices is hampered by poor reproducibility, low accuracy, and misinterpretation of biology. These limitations are partially caused by the lack of traceable calibration standards and the lack of effective validation methods. We developed numerical, benchtop, and in vivo phantoms to calibrate, validate, and optimize the performance of optical imaging devices for non-invasive assessment of ischemic wounds. The numerical phantom simulated the light transport in a semi-infinite turbid medium. The benchtop phantom simulated biological tissue at different oxygenation levels and with different absorption/scattering backgrounds. The in vivo phantom simulated ischemic wounds by introducing bipedicle skin flaps in a domestic pig. A wide-gap 2nd derivative algorithm was used for oxygenation reconstruction from hyperspectral data cubes. The acquired hyperspectral data cubes were further regenerated by a Digital Light Processing (DLP) hyperspectral projector and projected to other imaging systems. Our numerical and experimental results not only demonstrated the technical feasibility of quantitative oxygenation assessment in ischemic wounds but also provided the potential for a digital phantom platform to facilitate standardization and calibration of many medical optical imaging devices.