Multilayer, Dynamic, Mixed Solid/Liquid Human Head Models for the Evaluation of Near Infrared Spectroscopy Systems

Abstract

Human head-mimicking laboratory models (phantoms) play an essential role in the calibration, testing, and evaluation of near infrared spectroscopy (NIRS)-based optical brain imaging systems. Phantoms reported in the literature are generally of solid nature mimicking the optical properties of the overall human head with just a single layer. Solid phantoms can further be constructed as multilayered to mimic superficial layers of the head or with disks to mimic lesions within the brain layer. However, solid phantoms are of static nature, and hence, changing characteristics within the brain layer, such as in oxygen saturation cannot be modeled. Liquid phantoms can be used to model dynamic changes within the brain; however, existing liquid phantoms are usually of one layer and cannot mimic all the superficial layers of the head. In this article, we report the design and development of a six-layer human head-mimicking phantom that is a mixture of solid compartments modeling the superficial layers of the head and a liquid section mimicking the brain layer which provides a realistic and dynamic design. We also discuss studies which utilized this hybrid, six-layer phantom for the design and evaluation of a NIRS-based brain monitoring system, the hand-held hematoma detector. The results presented here highlight how the proposed phantom approach can be used to assess repeatability/reproducibility, as well as design and execution of agreement tests within and across different models of a NIRS-based brain monitoring device. Our new solid-liquid hybrid phantom model can guide the design and calibration of new NIRS systems, evaluation of their processing algorithms, and testing of the device capabilities under various conditions.