Keywords

1 Introduction

Near-infrared spectroscopy (NIRS) technologies are used for the non-invasive determination of the amount of absorbing molecular species in tissue, especially oxyhemoglobin and deoxyhemoglobin [1–5]. NIR light in the wavelength range of 700–1000 nm, which is called the “optical window for tissues”, is able to penetrate several centimeters into a variety of tissues because of the low absorbance of hemoglobin and water at these wavelengths. There are three fundamental approaches to measuring oxygenation in tissue non-invasively: the continuous wave (CW) , frequency resolved and time resolved techniques. NIRS using CW is the simplest and most practical of these technologies, although it is difficult to acquire the absolute hemoglobin concentration with this method.

Several CW NIRS systems are commercially available. However, these systems exhibit poor portability owing to their size and weight, fragile optical fibers that are easy to break, and slow data acquisition rates, all of which can make their use inconvenient. We have developed a portable, fiber-free, battery-operated, low noise, highly sensitive, wireless NIRS system with a fast sampling rate and two-channel probe operation. The product is called the “PocketNIRS Duo”. We expect that our wireless and portable NIRS system will prove to be a useful device in a number of fields including the healthcare industry, brain research, and brain–machine interfaces.

2 Instrumentation

2.1 PocketNIRS Duo

The PocketNIRS Duo consists of three main parts, as shown in Fig. 50.1: the optical probes, the controller, and the interface device (smartphone or personal computer). The optical probe has three built-in NIR light emitting diodes (LEDs) as light sources and a single silicon PIN photodiode as an optical detector which is positioned 3 cm from the LEDs. The analog signal from the photodiode is amplified and converted to a digital signal within the optical probe and then transferred to the controller via a 1 m long electrical cable. Digitization of the analogue signal prior to data transfer helps to preserve signal quality. The programmable gain amplifier (PGA) enables the measurement of signals over a wide dynamic range. The smartphone acquires and analyzes the data transmitted from the controller by Bluetooth® wireless technology, and then displays the results.

Fig. 50.1
figure 1

Our wireless NIRS system “PocketNIRS Duo” (a) Schematic diagram (b) External view

Full size image

2.2 PocketNIRS HM

The PocketNIRS Duo has been designed for the analysis of oxygenation and hemodynamics in a number of tissues including brain and muscle.

The requirement for measuring the oxygenation state of the brain is increasing. We have therefore designed a variant of the PocketNIRS, the “PocketNIRS HM”, specifically for the measurement of brain oxygenation (Fig. 50.2). All specifications are the same as those described for the PocketNIRS Duo. The control circuit, two optical probes, Bluetooth wireless data transmitter and battery are all integrated into a plastic helmet. This device does not require the use of double-sided tape on the optical probe. The PocketNIRS HM is simply mounted onto the head of the subject, the power turned on, and the start button on the smartphone pressed. The system then starts measurements immediately.

Fig. 50.2
figure 2

PocketNIRS HM (a) External view (b) Internal view

Full size image

3 The Principles of the PocketNIRS Duo

The changes in tissue concentrations of oxyhemoglobin and deoxyhemoglobin can be determined using the modified Beer-Lambert law. When the NIR light sources illuminate a tissue, the photodetector receives backscattered photons from the tissue. The penetration depth of detection is approximately one half the distance between the light sources and the photodetector. For example, if the light source and photodetector are separated by a distance of 3 cm, then the penetration depth is expected to be approximately 1.5 cm.

The attenuation of incident light intensity can be formulated as follows:

$$ OD\left(\lambda, t\right)={ \log}_{10}\frac{I_{in}\left(\lambda, t\right)}{I_{out}\left(\lambda, t\right)}=A\left(\lambda, t\right)+S\left(\lambda, t\right), $$
(50.1)

where I in (λ, t) is the incident light intensity, I out (λ, t) is the detected intensity, and OD(λ, t) is the optical density for wavelength λ at time t and is defined as the attenuation in light intensity as a function of the wavelength λ at time t. This attenuation can also be defined as the superposition of absorption A(λ, t) and scattering S(λ, t) of NIR light. Oxyhemoglobin and deoxyhemoglobin are the main absorbers of light in the NIR region in human tissue and their concentrations at a given point in time can be used to derive the light absorbance. Therefore, light absorbance can be formulated as follows:

$$ A\left(\lambda, t\right)={\varepsilon}_{Hb{O}_2}\left(\lambda \right)\left[Hb{O}_2\right](t)L\left(\lambda, t\right)+{\varepsilon}_{Hb}\left(\lambda \right)\left[Hb\right](t)L\left(\lambda, t\right), $$
(50.2)

where \( {\varepsilon}_{Hb{O}_2}\left(\lambda \right) \) and ε Hb (λ) are the respective extinction coefficients of oxyhemoglobin and deoxyhemoglobin at wavelength λ [6], [HbO 2 ] (t) and [Hb] (t) are the respective concentrations of oxyhemoglobin and deoxyhemoglobin, and L(λ, t) is the optical path length of the wavelength λ at time t. The optical path length can be expressed in terms of the source–detector separation distance as follows:

$$ L\left(\lambda, t\right)=DPF\left(\lambda, t\right)\cdot d, $$
(50.3)

where d is the distance between the light sources and the photodetector and DPF(λ, t) is the differential path length factor [7–9].

When scattering S(λ, t) is assumed to be constant with respect to both wavelength and time, the differential OD value at time t can be expressed as:

$$ \varDelta OD\left(\lambda, t\right)={\varepsilon}_{Hb{O}_2}\left(\lambda \right)\varDelta \left\{\left[Hb{O}_2\right]L\left(\lambda, t\right)\right\}+{\varepsilon}_{Hb}\left(\lambda \right)\varDelta \left\{\left[Hb\right]L\left(\lambda, t\right)\right\}, $$
(50.4)

using Eqs. (50.1) and (50.2), and the measurement of the OD value at initial time t 0.

In addition, the measurement of the differential OD value can be obtained as follows:

$$ \varDelta OD\left(\lambda, t\right)={ \log}_{10}\frac{I_{in}\left(\lambda, t\right)}{I_{out}\left(\lambda, t\right)}-{ \log}_{10}\frac{I_{in}\left(\lambda, {t}_0\right)}{I_{out}\left(\lambda, {t}_0\right)}={ \log}_{10}\frac{I_{out}\left(\lambda, {t}_0\right)}{I_{out}\left(\lambda, t\right)}, $$
(50.5)

In the above, I in (λ, t) at time t is equal to I in (λ, t 0) at initial time t 0 because I in (λ, t) is kept constant during measurement.

The relationship between changes in the product of the concentration and the optical path length and the differential OD values is derived from Eqs. (50.4) and (50.5), and least squares method as follows:

$$ \begin{array}{l}\left({}{}^t\mathbf{A}\mathbf{A}\right)\mathbf{C}={}{}^t\mathbf{A}\mathbf{M},\ \mathrm{with}\mathbf{A}=\left[\begin{array}{l}\begin{array}{cc}{\varepsilon}_{Hb{O}_2}(735)& {\varepsilon}_{Hb}(735)\end{array}\hfill \\ {}\begin{array}{cc}{\varepsilon}_{Hb{O}_2}(810)& {\varepsilon}_{Hb}(810)\end{array}\hfill \\ {}\begin{array}{cc}{\varepsilon}_{Hb{O}_2}(850)& {\varepsilon}_{Hb}(850)\end{array}\hfill \end{array}\right],\\ {}\kern0.85em \mathbf{C}=\left[\begin{array}{c}\varDelta \left\{\left[Hb{O}_2\right](t)L(t)\right\}\\ {}\varDelta \left\{\left[Hb\right](t)L(t)\right\}\end{array}\right]\ \mathrm{and}\ \mathbf{M}=\left[\begin{array}{c}\varDelta OD\left(735,t\right)\\ {}\varDelta OD\left(810,t\right)\\ {}\varDelta OD\left(850,t\right)\end{array}\right],\end{array} $$
(50.6)

where L(t) is the optical path length assuming, for simplicity, that the path length of all wavelengths are approximately equal. Changes in the product of the concentration and the optical path length, Δ{[HbO 2](t)L(t)} and Δ{[Hb](t)L(t)}, can be determined by solving Eq. (50.6).

In addition, change in the product of the total hemoglobin concentration and the optical path length, Δ{[tHb](t)L(t)}, is obtained as follows:

$$ \varDelta \left\{\left[tHb\right](t)L(t)\right\}=\varDelta \left\{\left[Hb{O}_2\right](t)L(t)\right\}+\varDelta \left\{\left[Hb\right](t)L(t)\right\} $$
(50.7)

4 Evaluation

4.1 Solid Phantom Test

To evaluate the performances of our NIRS system, an epoxy based solid phantom (absorption coefficient \( \left[{\mu}_a\right] = 0.04\ {\mathrm{mm}}^{-1} \) and reduced scattering coefficient \( \left[{\mu}_s^{\prime}\right] = 0.8\ {\mathrm{mm}}^{-1} \)) was used. The signal-to-noise ratio (SNR) is limited by shot noise generated by the random arrival of incidence photons at the photodetector.

The solid phantom test revealed that an SNR of at least 60 dB was obtainable with our system.

4.2 Arm Cuff Occlusion Test

The ability of the PocketNIRS Duo to measure oxygenation in human subjects was evaluated using an arm cuff occlusion test. The optical probe was placed on the left forearm of the subject. The PocketNIRS Duo clearly detected changes in forearm hemodynamics following cuff occlusion at 200 mmHg for 3 min (Fig. 50.3 shows the typical plots of one subject among five subjects).

Fig. 50.3
figure 3

Result of the arm cuff occlusion test (a) Experimental setup (b) Plots of changes in the product of hemoglobin concentrations and the optical path length in the left forearm of a subject

Full size image

4.3 Breath-Holding Experiment

To evaluate the ability of the PocketNIRS HM to measure brain hemodynamics we performed a breath-holding experiment. The optical probe was placed on the left side of the subject’s forehead and a commercial pulse-oximeter sensor (8000R, Nonin Medical Inc.), which source-detector separation was approximately 7.5 mm, was attached on the right side of the forehead. The commercial pulse-oximeter sensor (SR-5C, Konica Minolta Inc.) was attached on the right forefinger. The forehead hemodynamics and forefinger SpO2 (blood oxygen saturation level) data are presented in Fig. 50.4, which shows typical plots of one subject among five subjects. The subject held their breath 30 s after the commencement of measurements. In response to breath-holding, the SpO2 value recorded by the pulse oximeter gradually decreased. The PocketNIRS HM detected a decrease in Δ[HbO 2 ] and an increase in Δ[Hb] a few tens of seconds after an increase in Δ[HbO 2 ] and Δ[Hb]. The subject restarted breathing approximately 136 s later. Following rebreathing, Δ[HbO 2 ] increased and Δ[Hb] decreased but the SpO2 of the right forehead recovered later than these signals. Especially, the recovery and the initiation of SpO2 in the right forefinger was delayed approximately 10 s than that in the right forehead. A small wave, the period of which is approximately 0.96 Hz (57 bpm), could be observed in Δ[HbO 2 ] signal, which could be attributed to the subject’s heartbeat.

Fig. 50.4
figure 4

Changes in brain oxygenation during a breath-holding experiment

Full size image

The PocketNIRS HM therefore demonstrates sufficient sensitivity to clearly monitor brain hemodynamics and has a data acquisition rate sufficient to detect pulsation due to heartbeat.

5 Conclusion

We have developed a portable wireless NIRS system, the PocketNIRS Duo, and the related PocketNIRS HM system that specifically facilitates the analysis of brain hemodynamics. The primary features of these systems are their ability to measure changes in oxyhemoglobin, deoxyhemoglobin and total hemoglobin concentration. These systems demonstrate a high SNR (greater than 60 dB), fast data acquisition rate (up to 60 Hz), and simultaneous two-channel probe operation capability. Furthermore, the small size and light weight (requiring only two AAA batteries for operation), and the wireless capabilities of these systems (which are controllable by either smartphone or personal computer) make these devices ideal tools not only for clinical applications but also for healthcare monitoring at home, sports medicine, and the field of brain research.