Visualizing Brain Activity Using Near-Infrared Light

Biological tissue absorbs visible light, which can be perceived by the human eye, making it impossible to see inside the body with the naked eye. X-rays, discovered by Röntgen, are a type of electromagnetic wave like visible light. However, they penetrate biological tissue well, and their absorption varies depending on the type of tissue, such as fat, muscle, and bone. This allows us to see the internal structures of the body without making an incision. For this reason, X-ray imaging has become indispensable for medical diagnosis. Near-infrared light, which has a longer wavelength than visible light, also penetrates biological tissue, although not as well as X-rays. Figure 1 shows a transmission image of a hand taken by irradiating it with near-infrared light. The central part of the finger appears bright, indicating that near-infrared light is passing through the bone. On the other hand, the black lines of various thicknesses are blood vessels. This is because near-infrared light is absorbed by blood, causing the intensity of the transmitted light to be attenuated in the areas of the blood vessels.

Brain activity consists of electrical changes in nerve cells, and the parts of the brain that are active differ depending on the function being performed, such as motor control, sensation, vision, hearing, language, and thought. It is also known that when the brain is active, the amount of blood in the surrounding area increases locally. Since near-infrared light penetrates bone and is absorbed by blood, by attaching multiple fiber probes to the surface of the head to irradiate and receive near-infrared light, as shown in the photograph in Figure 2, it is possible to detect the attenuation of light corresponding to changes in blood volume in various parts of the brain tissue. Brain functional imaging with near-infrared light is the process of imaging the spatial distribution of these blood volume changes as the active areas of the brain.

In the medical field, brain functional imaging with near-infrared light is used for pre-surgical examinations for epilepsy, evaluation of functional recovery in rehabilitation for patients with brain damage, and as an aid in the differential diagnosis of psychiatric disorders such as depression. It is also applied to research in a wide range of fields related to brain function, such as cognitive development in newborns and infants, brain-machine interfaces, and brain activity during social interaction and communication with others.

X-rays can accurately image the internal structures of the body, but they cannot measure functional information such as local changes in blood volume due to brain activity. On the other hand, near-infrared light is strongly scattered by biological tissue, so it cannot travel in a straight line within the tissue like X-rays. Therefore, brain functional imaging with near-infrared light is equivalent to seeing an object in the fog, and the areas of brain activity in the image shown in Figure 2 appear more spread out than the areas where the brain was actually active. To solve this problem, it is necessary to reconstruct the image using information about the path through which the near-infrared light, irradiated and received by the fiber probes on the head, propagates within the brain tissue. However, it is not possible to directly measure light propagation within the head tissue. Therefore, we are conducting research to estimate this by creating a model on a computer that faithfully simulates the anatomical structure of the head, as shown in Figure 3(a), and simulating the propagation of near-infrared light. The simulation results have revealed that near-infrared light propagates over a wide area of brain tissue due to scattering. Furthermore, since light propagation in brain tissue is also affected by factors such as the thickness of the scalp and skull, it is important to create an accurate model and perform high-precision analysis of light propagation. By performing image reconstruction based on light propagation simulations using a head model, it becomes possible to accurately obtain a brain functional image overlaid on a structural image of the brain tissue, as shown in Figure 3(b).