Light-scattering–assisted near-infrared spectroscopy for improved quantitative analysis of tissue chromophores and oxygenation

VladimirHovhannisyan1

YvonneYulingHu2

Liang-WeiChen1

Chun-YuLin1

Hsin-HungChen3

Shin-TzuChang4,5

Shean‑JenChen1✉Emailsheanjen@nycu.edu.tw

College of PhotonicsNational Yang Ming Chiao Tung University711TainanTaiwan

2Department of PhotonicsNational Cheng Kung University701TainanTaiwan

3Department of Medical Education and ResearchKaohsiung Veterans General HospitalKaohsiungTaiwan

4Department of Physical Medicine and RehabilitationKaohsiung Veterans General Hospital813KaohsiungTaiwan

5Department of Physical Medicine and Rehabilitation, Tri-Service General HospitalNational Defense Medical Center114TaipeiTaiwan

Vladimir Hovhannisyan,¹ Yvonne Yuling Hu,² Liang-Wei Chen,¹ Chun-Yu Lin¹, Hsin-Hung Chen,³ Shin-Tzu Chang,^4,5 and Shean‑Jen Chen¹

¹ College of Photonics, National Yang Ming Chiao Tung University, Tainan 711, Taiwan.

² Department of Photonics, National Cheng Kung University, Tainan 701, Taiwan

³ Department of Medical Education and Research, Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan

⁴ Department of Physical Medicine and Rehabilitation, Kaohsiung Veterans General Hospital, Kaohsiung 813, Taiwan

⁵ Department of Physical Medicine and Rehabilitation, Tri-Service General Hospital, National Defense Medical Center, Taipei 114, Taiwan

Correspondence: S.-J. C. (sheanjen@nycu.edu.tw)

Abstract

Near-infrared spectroscopy (NIRS) is a noninvasive and quantitative technique for analyzing biological tissues, particularly effective within the 670–1,000 nm spectral range due to the high optical transparency of this window. NIRS enables real-time assessment of key chromophores—deoxyhemoglobin (Hb), oxyhemoglobin (HbO₂), and oxidized cytochrome c oxidase (CCO)—thereby providing valuable insights into tissue oxygenation and metabolic activity in vivo. However, conventional NIRS approaches often overlook the influence of tissue light scattering, leading to inaccuracies in chromophore quantification. To address this limitation, we developed a portable, four-wavelength enhanced NIRS system capable of simultaneously monitoring fluctuations in Hb, HbO₂, and CCO concentrations while capturing dynamic variations in tissue light-scattering properties. Experiments performed under diverse respiratory and circulatory conditions demonstrate that integrating light-scattering analysis substantially improves the accuracy of chromophore quantification. Our findings reveal that incorporating scattering information refines absorption evaluation and significantly enhances the precision of NIRS-based oxygenation monitoring. This light-scattering–assisted approach advances the analytical performance of NIRS and extends its applicability for comprehensive physiological assessment in vivo, bridging the gap between scattering effects and quantitative tissue spectroscopy.

Introduction

In living tissues, variations in the concentrations of deoxyhemoglobin (Hb), oxyhemoglobin (HbO₂), and oxidized cytochrome c oxidase (CCO) serve as key indicators of oxygen transport and metabolic activity. Hb is responsible for oxygen delivery from the lungs to peripheral tissues, while CCO—the terminal enzyme of the mitochondrial electron transport chain—catalyzes the reduction of oxygen to water and drives adenosine triphosphate (ATP) synthesis essential for cellular energy production. Near-infrared spectroscopy (NIRS) provides a noninvasive and cost-effective optical approach for assessing cerebral and peripheral tissue oxygenation and metabolism. Broadband NIRS has proven effective in simultaneously monitoring Hb, HbO₂, and CCO during procedures such as low-level laser therapy (LLLT), where increases in HbO₂ and CCO correlate with laser energy dosage, suggesting a dose-dependent photobiomodulation effect.¹ Moreover, integrating NIRS with arterial and venous occlusion techniques allows real-time quantification of muscle blood flow and oxygen consumption during exercise and clinical testing.²^–⁵ Recent advances have further broadened these applications: NIRS has been employed for continuous cerebral oxygenation monitoring during stroke rehabilitation, enhancing sensitivity to metabolic recovery;⁶ theoretical modeling has improved chromophore quantification accuracy by addressing spectral and instrumental limitations;⁷ and validation studies have confirmed the reliability of NIRSS for assessing muscle and tissue oxygenation under dynamic physiological conditions.⁸ Collectively, these developments establish NIRS as a versatile tool for investigating photobiomodulation, hemodynamics, and mitochondrial metabolism in vivo.

NIRS technology has also become increasingly valuable for monitoring cerebral hemodynamics—particularly in patients undergoing repetitive hemodialysis—providing critical insights into vascular responses and tissue perfusion.⁹ Its evolution into a sensitive tool for evaluating skeletal muscle oxygenation during physical activity is equally noteworthy.¹⁰ Gender-specific differences observed in vascular occlusion tests further underscore the influence of physiological characteristics such as skeletal muscle mass, adipose tissue thickness, and microvascular function on ischemic responses.¹¹ For instance, studies examining reproducibility and sex-based differences in NIRS measurements during brachial artery occlusion have revealed distinct response patterns between men and women.¹² In addition, NIRS signal analyses have elucidated the complex interplay between hemoglobin and myoglobin—particularly under conditions of reduced oxygen delivery—highlighting myoglobin’s significant yet often underappreciated role in muscular oxygen metabolism.¹³ Importantly, recent studies demonstrate that NIRS-derived assessments of skeletal muscle oxidative capacity remain reliable across different exercise protocols and diverse populations, reinforcing the utility of NIRS in both research and clinical contexts.¹⁴

For the quantitative analysis of Hb, HbO₂, and CCO concentration changes, numerous studies¹,¹⁵^–¹⁷ have applied matrix-based computational approaches to extract chromophore dynamics from NIRS data. In most of these investigations, tissue scattering properties were characterized separately—often at the start of the measurement session—using time-domain or frequency-domain NIRS systems, or complementary optical scattering techniques.¹⁶^,¹⁸^,¹⁹ However, recent research has shown that red blood cell (RBC) aggregation and disaggregation, which are strongly influenced by blood flow dynamics, markedly affect tissue light-scattering behavior during optical measurements.²⁰^,²¹ RBCs function not only as absorptive chromophores but also as dynamic scattering centers. Under conditions of low shear stress or venous stasis, aggregated erythrocytes enhance refractive index mismatches, increasing scattering and altering the apparent optical path length and absorption estimates.²²^,²³ In venous blood, this scattering contributes substantially to NIRS signal variability, with activation-related changes closely linked to flow-induced dispersion of aggregated RBCs. Moreover, recent analyses of whole-blood optical properties have demonstrated that reduced scattering coefficients are strongly dependent on hematocrit, shear rate, and aggregation state,²⁰^,²¹ underscoring the necessity of incorporating these factors into chromophore quantification models. Accounting for such scattering effects is therefore essential for improving the accuracy of chromophore concentration estimation and enhancing the physiological interpretability of NIRS data in vivo.

To advance the precision of in vivo oxygenation assessment, we developed a portable, four-wavelength NIRS system capable of simultaneously capturing changes in the absorption of Hb, HbO₂, and CCO, as well as the scattering properties of biological tissues (Tiss). This integrated approach allows concurrent monitoring of both chromophore dynamics and light-scattering variations, providing a more comprehensive characterization of tissue oxygenation. The system was evaluated under four controlled, short-term physiological perturbations—inhalation and breath-holding (IBH), exhalation and breath-holding (EBH), arterial occlusion (AO), and venous occlusion (VO)—each known to induce pronounced alterations in oxidation states and scattering behavior. Through these controlled perturbations, the developed system demonstrates enhanced sensitivity and accuracy in distinguishing absorption- and scattering-driven components of NIRS signals, thereby improving the reliability of in vivo chromophore quantification and tissue oxygenation analysis.

System and Methods

System configuration. Figure 1 illustrates the experimental setup of the developed NIRS system. The configuration comprises four near-infrared light-emitting diodes (NIR-LEDs) positioned on the outer surface of the palm and four optical sensors aligned on the inner surface. These components are connected to a data acquisition unit and a computer via a USB interface for real-time processing. The NIR-LEDs operate at wavelengths of 670, 805, 848, and 905 nm, each delivering an optical power of 5–8 mW through an illumination head measuring 3 × 5 mm². The diodes are embedded within a flexible black sponge to ensure mechanical stability and minimize ambient light interference. Light transmitted through the tissue is detected by TSL250R light-to-voltage optical sensors (Texas Instruments), each with a 2 mm active area. The sensor output voltage is linearly proportional to the incident light intensity. Depending on the experimental protocol, one to four sensors can be used to record the transmitted optical signals across the palm.

Fig. 1

Schematic of the experimental setup showing four near-infrared (NIR) LEDs positioned on the outer surface of the palm, four optical sensors on the inner surface, a data-acquisition device, and a computer for real-time data processing connected via USB.

System control and data acquisition are performed using a National Instruments NI USB-6001 interface, which provides eight input channels, 14-bit resolution, and a sampling rate of 20 kS/s. A custom-designed software platform governs sequential LED activation, with adjustable pulse durations between 1–10 ms and inter-pulse intervals ranging from 5 ms (200 Hz) to 10 s (0.1 Hz). The recorded optical signals are processed to compute temporal changes in the concentrations of four key tissue components—Hb, HbO₂, CCO, and Tiss—relative to baseline values over a 5–10 s window. These results are displayed in real time and stored for subsequent analysis. Because CCO concentration remains relatively stable over short time intervals, the measurement focuses on the differential absorption coefficient between its oxidized and reduced states, expressed as Cyt = oxidized CCO – reduced CCO, which serves as an indicator of the CCO redox state.

Signal acquisition and processing. A control program developed in LabWindows/CVI manages LED pulses, including frequency, duration, and sequence. The software also digitizes the signals from the photodiodes and converts the recorded spectral data into concentration changes of Hb (Δ[Hb]), HbO₂ (Δ[HbO₂]), Cyt (Δ[Cyt]), and tissue scattering (Δ[Tiss]). The changes in light attenuation (ΔA(λ_i)) measured at different wavelengths are calculated in real time according to:

$\Delta {\text{A(}}{\lambda _i})=L \cdot \left( {{\varepsilon _{{\text{Hb}}}}({\lambda _i}) \cdot \Delta [{\text{Hb}}]+{\varepsilon _{{\text{Hb}}{{\text{O}}_2}}}({\lambda _i}) \cdot \Delta [{\text{Hb}}{{\text{O}}_2}]+{\varepsilon _{{\text{Cyt}}}}({\lambda _i}) \cdot \Delta [{\text{Cyt}}]+{\mu _{{\text{Tiss}}}}({\lambda _i}) \cdot \Delta [{\text{Tiss}}]} \right)$

where L is the optical pathlength,

${\varepsilon _{{\text{Hb}}}}({\lambda _i})$

${\varepsilon _{{\text{Hb}}{{\text{O}}_{\text{2}}}}}({\lambda _i})$

, and

${\varepsilon _{{\text{Cyt}}}}({\lambda _i})$

are the extinction coefficients of Hb, HbO₂, and Cyt, respectively, and

${\mu _{{\text{Tiss}}}}({\lambda _i})$

is the reduced scattering coefficient at a specific wavelength λ. For a four-chromophore system (Hb, HbO₂, Cyt, and Tiss) measured at four wavelengths (i = 4), Eq. (1) can be expressed in matrix form as:

$\left[ {\begin{array}{*{20}{c}} {\Delta [{\text{Hb}}]} \\ {\Delta [{\text{Hb}}{{\text{O}}_2}]} \\ {\Delta [{\text{Cyt}}]} \\ {\Delta [{\text{Tiss]}}} \end{array}} \right]={1 \mathord{\left/ {\vphantom {1 {L \cdot }}} \right. \kern-0pt} {L \cdot }}{\left[ {\begin{array}{*{20}{c}} {{\varepsilon _{{\text{Hb}}}}({\lambda _1})}&{{\varepsilon _{{\text{Hb}}{{\text{O}}_{\text{2}}}}}({\lambda _1})}&{{\varepsilon _{{\text{Cyt}}}}({\lambda _1})}&{{\mu _{{\text{Tiss}}}}({\lambda _1})} \\ {{\varepsilon _{{\text{Hb}}}}({\lambda _2})}&{{\varepsilon _{{\text{Hb}}{{\text{O}}_2}}}({\lambda _2})}&{{\varepsilon _{{\text{Cyt}}}}({\lambda _2})}&{{\mu _{{\text{Tiss}}}}({\lambda _2})} \\ {{\varepsilon _{{\text{Hb}}}}({\lambda _3})}&{{\varepsilon _{{\text{Hb}}{{\text{O}}_2}}}({\lambda _3})}&{{\varepsilon _{{\text{Cyt}}}}({\lambda _3})}&{{\mu _{{\text{Tiss}}}}({\lambda _3})} \\ {{\varepsilon _{{\text{Hb}}}}({\lambda _4})}&{{\varepsilon _{{\text{Hb}}{{\text{O}}_2}}}({\lambda _4})}&{{\varepsilon _{{\text{Cyt}}}}({\lambda _4})}&{{\mu _{{\text{Tiss}}}}({\lambda _4})} \end{array}} \right]^{ - 1}} \cdot \left[ {\begin{array}{*{20}{c}} {\Delta {\text{A(}}{\lambda _1})} \\ {\Delta {\text{A(}}{\lambda _2})} \\ {\Delta {\text{A(}}{\lambda _3})} \\ {\Delta {\text{A(}}{\lambda _4})} \end{array}} \right]$

Based on published tables of absorption coefficients for Hb, HbO₂, and Cyt,²² and scattering coefficients of forearm tissue,²³ the spectral dependences of four parameters—ε_Hb(λ), ε_HbO₂(λ), ε_Cyt(λ), and µ_Tiss(λ)—were plotted (Fig. 2). Specifically, for λ₁ = 670 nm, λ₂ = 805 nm, λ₃ = 848 nm, and λ₄ = 905 nm, Eq. (2) takes the explicit form:

Fig. 2

Absorption spectra of Hb, HbO₂, Cyt, and the reduced scattering coefficient of forearm tissue within the near-infrared range, used for quantitative characterization of the four biomarkers.

$\left[ {\begin{array}{*{20}{c}} {\Delta [{\text{Hb}}{{\text{O}}_2}]} \\ {\Delta [{\text{Hb}}]} \\ {\Delta [{\text{Cyt}}]} \\ {\Delta [{\text{Tiss]}}} \end{array}} \right]={1 \mathord{\left/ {\vphantom {1 {L \cdot }}} \right. \kern-0pt} {L \cdot }}{\left[ {\begin{array}{*{20}{c}} {3.04}&{0.43}&{4.91}&{10.1} \\ {0.82}&{0.89}&{4.62}&{7.6} \\ {0.78}&{1.08}&{4.57}&{6.5} \\ {0.89}&{1.25}&{2.17}&{5.2} \end{array}} \right]^{ - 1}} \cdot \left[ {\begin{array}{*{20}{c}} {\Delta {\text{A(}}{\lambda _1})} \\ {\Delta {\text{A(}}{\lambda _2})} \\ {\Delta {\text{A(}}{\lambda _3})} \\ {\Delta {\text{A(}}{\lambda _4})} \end{array}} \right]$

Data representation. In Figs. 3–5, the red, brown, blue, and green dots represent raw experimental data corresponding to measured light intensities at wavelengths of 670 nm, 805 nm, 848 nm, and 905 nm, respectively. The red, brown, blue, and green lines in the processed data plots illustrate the calculated concentration changes of Hb, HbO₂, Cyt, and Tiss. In addition, the total hemoglobin concentration (HbT), computed as Hb + HbO₂, is shown in magenta. Each processed dataset represents the average of three repeated measurements conducted under identical external conditions.

Fig. 3

Inhalation and breath-holding (IBH). (A) Raw data acquired with the four-wavelength NIRS system. (B) Computed changes in chromophore concentrations. The onset of breath-holding is indicated by a green arrow, and the resumption of normal breathing by a red arrow.

Fig. 5

Arterial occlusion (AO). (A) Raw data acquired with the four-wavelength NIRS system. (B) Computed changes in chromophore concentrations. Cuff inflation to 200 mmHg, initiating arterial occlusion, is marked by a green arrow, and cuff release is denoted by a red arrow.

Experimental Results

In this study, a four-wavelength NIRS system operating at 670, 805, 848, and 905 nm was employed to monitor concentration changes in Hb, HbO₂, Cyt, and Tiss in the palms of human volunteers. The system comprised four NIR LEDs positioned on the outer surface of the palm and up to four optical sensors aligned on the inner surface, as illustrated in Fig. 1. Representative results obtained from the measurements are shown in Figs. 3–6. In the upper panels of each figure, experimental light-absorption data at the four wavelengths are displayed: red dots (670 nm), brown dots (805 nm), blue dots (848 nm), and green dots (905 nm). The lower panels show the corresponding calculated concentration changes: Hb (red), HbO₂ (brown), Cyt (blue), Tiss (green), and total hemoglobin (HbT = Hb + HbO₂, magenta).

Fig. 6

Venous occlusion (VO). (A) Raw data acquired with the four-wavelength NIRS system. (B) Computed changes in chromophore concentrations. Cuff inflation to 100 mmHg, causing venous outflow blockage, is indicated by a green arrow. A blue arrow marks the phase of sustained cuff inflation associated with partial rebound in HbO₂ and Cyt levels. The red arrow indicates cuff deflation and rapid normalization of physiological parameters.

Experiments were conducted on five healthy volunteers (three men and two woman, aged 23–67 years), in full compliance with ethical standards.

All participants provided written informed consent after being briefed on the study’s purpose, procedures, and potential risks.

The experimental protocol was approved by the Institutional Review Board (IRB) of National Yang Ming Chiao Tung University (NYCU) and conducted in accordance with the Declaration of Helsinki and the ethical guidelines of the National Health Research Institutes (NHRI), Taiwan.

Inhalation and breath-holding (IBH). During IBH, temporary cessation of airflow—and thus oxygen intake—induces a short period of tissue hypoxia, reflected by a decline in HbO₂ concentration and a concomitant rise in Hb levels. This reduction in oxygen availability also decreases the oxidized form of Cyt, indicating reduced mitochondrial oxidative activity. As shown in Fig. 3, Hb concentrations increased steadily following the onset of breath-holding (green arrow), while HbO₂ and Cyt levels declined until normal respiration resumed (red arrow). The system consistently detected variations in tissue absorption–scattering properties (green line), confirming its ability to monitor concurrent optical property changes. Furthermore, total hemoglobin (HbT = Hb + HbO₂; magenta line) exhibited a distinct rise from baseline within the first 30 s of the IBH phase, reflecting transient vasodilation and blood pooling in response to reduced oxygen supply.

Exhalation and breath-holding (EBH). The physiological responses during EBH were broadly consistent with those observed in IBH (Fig. 4). The onset of breath-holding and resumption of normal respiration are indicated by the green and red arrows, respectively. The key distinction lies in the lower initial lung oxygen volume at the start of EBH, which limited breath-holding duration to roughly one-third that of IBH. Consequently, reductions in HbO₂ and Cyt, along with the corresponding increase in Hb, occurred at approximately twice the rate observed during IBH. This accelerated dynamic reflects the faster onset of tissue hypoxia due to the smaller oxygen reserve, underscoring the system’s sensitivity to short-term physiological fluctuations in oxygenation and metabolism.

Fig. 4

Exhalation and breath-holding (EBH). Analysis of chromophore concentration changes. The onset of exhalation and breath-holding is indicated by a green arrow, and the resumption of normal respiration by a red arrow.

Arterial occlusion (AO). In AO experiments, a medical tonometer cuff was inflated to 200 mmHg (green arrow) to induce temporary arterial occlusion. As shown in Fig. 5, arterial inflow to the monitored region was halted, leading to a progressive decline in HbO₂ due to tissue deoxygenation, accompanied by a corresponding increase in Hb. Because both arterial inflow and venous outflow were restricted, total hemoglobin (Δ[HbT]) remained approximately constant, indicating minimal net blood-volume change within the illuminated tissue.

Simultaneously, Cyt levels decreased, reflecting diminished mitochondrial oxidation caused by reduced oxygen delivery. Upon cuff deflation (red arrow), arterial perfusion and oxygen supply were gradually restored. The reperfusion phase was characterized by transient overshoots in HbO₂ and Cyt—rising about 10% above baseline—before returning to steady-state levels within 1–1.5 min. Subsequently, both HbO₂ and Cyt exhibited a secondary dip, reaching minima approximately 55 s after reperfusion onset, followed by gradual recovery. Throughout this process, Hb and Tiss signals closely tracked hemodynamic changes, eventually stabilizing as normal circulation resumed.

Venous occlusion (VO). During VO experiments, the cuff was inflated to 100 mmHg to obstruct venous outflow while allowing arterial inflow for approximately 70 s. As illustrated in Fig. 6, a distinct hemodynamic pattern emerged. Within the first 20 s of occlusion, Hb, HbO₂, and Cyt concentrations all increased, reflecting blood pooling and enhanced local oxygen delivery. A notable inflection appeared near the 60 s mark (blue arrow), when Hb levels began to decline while HbO₂ and Cyt continued to rise. This transition coincided with the cessation of signal attenuation across all wavelengths and the onset of an upward trend in the absorption data. The observed shift likely represents a physiological adaptation of the venous system to sustained pressure, possibly facilitating partial venous drainage or redistribution of blood toward the central circulation. Following cuff release (red arrow), normal blood flow was rapidly re-established. All measured parameters—Hb, HbO₂, Cyt, and Tiss—returned sharply to baseline within about 10 s, confirming the system’s capacity to resolve fast transient hemodynamic changes in vivo.

Discussions

A persistent challenge in applying in vivo NIRS lies in the pronounced attenuation of light intensity caused by the strong scattering properties of biological tissues.²⁴ Such scattering not only lengthens the photon path through tissue—thereby increasing the probability of absorption—but also reduces detected signal amplitude and complicates accurate chromophore quantification.²⁵ It is well established that increases in tissue oxygenation signals during localized or whole-body warming correlate strongly with enhanced cutaneous blood flow.²⁶ Recent advances further indicate that the tissue scattering coefficient is not constant: dynamic changes in scattering have been observed during exercise and thermal perturbation, suggesting that assuming a fixed scattering coefficient can lead to substantial overestimation of physiological parameters in NIRS analyses.²⁷^,²⁸ Recognizing and compensating for such scattering variability is therefore critical for improving the fidelity of chromophore concentration estimation and the interpretation of oxygenation dynamics in both research and clinical contexts.

Recent investigations have elucidated the link between NIRS-derived signals and microvascular blood-flow dynamics, offering deeper insight into the often-overlooked role of tissue scattering in hemoglobin quantification.²⁰ Even subtle alterations in erythrocyte flow can meaningfully modify tissue scattering profiles through changes in RBC morphology and aggregation–disaggregation dynamics at the membrane level. Consequently, shifts in erythrocyte flow affect not only the detected concentrations of Hb and HbO₂ but also the measured scattering characteristics of the tissue matrix. Thus, NIRS signal variations represent combined effects of absorption and scattering changes, rather than pure chromophore fluctuations.

Interferometric NIRS studies have further confirmed the wavelength-dependent nature of dynamic scattering in human forearm tissue, demonstrating that flow-related scattering fluctuations can significantly influence blood-flow index estimates and confound absorption-only models.²⁸ Additionally, comprehensive reviews of whole-blood optical properties reveal that scattering coefficients are strongly dependent on hematocrit, shear rate, and RBC aggregation state—parameters closely linked to microvascular flow.¹⁹ These findings collectively emphasize the need to integrate dynamic scattering corrections into chromophore quantification algorithms to minimize bias and enhance physiological interpretability of NIRS measurements in vivo.

Our experimental results demonstrate that, under the physiological conditions modeled in this study, variations in tissue light scattering are tightly coupled with concurrent changes in hemoglobin and CCO absorption. By explicitly incorporating and quantifying the scattering component within the NIRS analytical framework, we achieved more accurate estimations of hemoglobin and CCO concentration changes in vivo. These findings highlight the importance of treating scattering not as a static background property but as an active optical variable that evolves alongside metabolic and hemodynamic processes. Incorporating wavelength-resolved scattering analysis within the near-infrared spectrum can therefore substantially enhance the precision of NIRS-based oxygenation assessments. We anticipate that such combined absorption–scattering modeling will improve the physiological interpretability of NIRS data and expand its applicability for real-time monitoring of tissue oxygenation, metabolism, and microvascular function in both clinical and experimental settings.

Conclusions

We developed a four-wavelength NIRS system capable of real-time, in vivo monitoring of dynamic changes in Hb, HbO₂, CCO, and tissue light-scattering properties within the near-infrared range. The system was rigorously evaluated under diverse physiological conditions—including arterial and venous occlusion and controlled respiratory maneuvers—to assess its ability to detect simultaneous absorption and scattering variations. The results demonstrate that integrating scattering analysis substantially improves the accuracy of chromophore quantification and enhances the interpretability of oxygenation dynamics. This advanced NIRS platform also exhibits strong potential for assessing the physiological effects of LLLT and elucidating oxygen metabolism across a range of tissues, including brain, muscle, and peripheral organs under varying functional states. Given its adaptability and sensitivity, the proposed system offers broad applicability across disciplines such as sports and rehabilitation medicine, high-altitude and aerospace physiology, emergency and critical care, organ transplantation, dermatology, and wound management. Collectively, these findings highlight its promise as a versatile, noninvasive tool for advancing both clinical diagnostics and fundamental research on tissue oxygenation and metabolic regulation.

Data availability

The datasets used and analyzed in the current study are available from the corresponding author on reasonable request.

Data Availability

The datasets used and analyzed in the current study are available from the corresponding author on reasonable request.

References

Wang, X. et al. Interplay between up-regulation of cytochrome-c-oxidase and hemoglobin oxygenation induced by near-infrared laser. Sci. Rep. 6, 30540 (2016).

Hassan, I. A. A. et al. Measurement of peripheral oxygen utilisation in neonates using near-infrared spectroscopy: comparison between arterial and venous occlusion methods. Early Hum. Dev. 57, 211–224 (2000).

Dennis, J. J. et al. Measurement of muscle blood flow and O₂ uptake via near-infrared spectroscopy using a novel occlusion protocol. Sci. Rep. 11, 918 (2021).

Niezen, C. K. et al. The use of a vascular occlusion test combined with near-infrared spectroscopy in perioperative care: a systematic review. J. Clin. Monit. Comput. 36, 933–946 (2022).

Verhaeghe, C., Lapage, K. & Moerman, A. Quantitative assessment of cytochrome c oxidase patterns in muscle tissue by near-infrared spectroscopy in healthy volunteers. J. Clin. Monit. Comput. 36, 271–278 (2022).

Zhang, Q. et al. Functional near-infrared spectroscopy monitoring of brain tissue during stroke rehabilitation. Sensors 24, 8175 (2024).

Park, J. et al. Spectral constraints on accurate estimation of ∆[oxCCO] and hemoglobin signals in near-infrared spectroscopy. J. Biophotonics. 17, e202400231 (2024).

Chen, L., Patel, S. & Ito, M. Reliability of NIRS-based muscle oxygenation measurements during dynamic exercise. J. Exerc. Sci. Fit. 22, 247–256 (2024).

Wong, A. et al. Assessment of cerebral oxygenation response to hemodialysis using near-infrared spectroscopy (NIRS): Challenges and solutions. J. Innov. Opt. Health Sci. 14, 2150016 (2021).

10.

Ellis, C. & Burns, D. All about oxygen: using near-infrared spectroscopy to understand bioenergetics. Adv. Physiol. Educ. 46, 685–692 (2022).

11.

Keller, J. L. et al. Sex differences in NIRS-derived values of reactive hyperemia persist after experimentally controlling for the ischemic vasodilatory stimulus. J. Appl. Physiol. 135, 3–14 (2023).

12.

Jeskanen, T. et al. Reproducibility and sex differences in muscle oxygenation during brachial artery occlusion in healthy participants. Clin. Physiol. Funct. Imaging. 44, 388–395 (2024).

13.

Koirala, B. et al. Relationship between muscle venous blood oxygenation and near-infrared spectroscopy: quantitative analysis of the Hb and Mb contributions. J. Appl. Physiol. 134, 1063–1074 (2023).

14.

Tandirerung, F. J., Jamieson, A., Hendrick, E., Hughes, A. D. & Jones, S. Near-infrared spectroscopy (NIRS) in vivo assessment of skeletal muscle oxidative capacity: a comparison of results from short versus long exercise protocols and reproducibility in non-athletic adults. Front. Physiol. 15, 1429673 (2024).

15.

Matcher, S. J. et al. Performance comparison of several published tissue near-infrared spectroscopy algorithms. Anal. Biochem. 227, 55–68 (1995).

16.

Bale, G., Elwell, C. E. & Tachtsidis, I. From Jöbsis to the present day: a review of clinical near-infrared spectroscopy measurements of cerebral cytochrome-c-oxidase. J. Biomed. Opt. 21, 091307 (2016).

17.

Kaynezhad, P. et al. Quantification of the severity of hypoxic-ischemic brain injury in a neonatal preclinical model using measurements of cytochrome-c-oxidase from a miniature broadband-near-infrared spectroscopy system. Neurophoton 6, 045009 (2019).

18.

Takeyama, R. Estimation of red blood cell aggregation size using power-law scattering spectra. Jpn J. Appl. Phys. 63, S11 (2024).

19.

Saiko, G., Sadrzadeh-Afsharazar, F., Burton, T., Prahl, S. & Douplik, A. Absorption, scattering, and refractive index of blood and its components: a review. Front. Photonics. 6, 1636398 (2025).

20.

Tomita, M., Ohtomo, M. & Suzuki, N. Contribution of the flow effect caused by shear-dependent RBC aggregation to NIR spectroscopic signals, NeuroImage 33, 1–10 (2006).

21.

Hammer, S. M. et al. Effect of assuming constant tissue scattering on measured tissue oxygenation values during tissue ischemia and vascular reperfusion. J. Appl. Physiol. 127, 22–30 (2019).

22.

Wray, S. et al. Characterization of the near infrared absorption spectra of cytochrome aa3 and haemoglobin for the non-invasive monitoring of cerebral oxygenation. Biochim. Biophys. Acta. 933, 184–192 (1988).

23.

Taroni, P. et al. vivo absorption and scattering spectroscopy of biological tissues. Photochem. Photobiol Sci. 2, 124–129 (2003).

24.

Jobsis, F. F. Noninvasive, infrared monitoring of cerebral and myocardial oxygen sufficiency and circulatory parameters. Science 198, 1264–1267 (1977).

25.

Davis, S. L. et al. Skin blood flow influences near-infrared spectroscopy-derived measurements of tissue oxygenation during heat stress. J. Appl. Physiol. 100, 221–224 (2006).

26.

Ferreira, L. F. et al. Effects of assuming constant optical scattering on measurements of muscle oxygenation by near-infrared spectroscopy during exercise. J. Appl. Physiol. 102, 358–367 (2007).

27.

Obonai, A. Modeling of scattering coefficient for biological tissues exposed to near-infrared laser irradiation. Appl. Therm. Eng. 256, 123990 (2024).

28.

Mazumder, D. et al. Interferometric near-infrared spectroscopy (iNIRS) reveals mixed static and dynamic scattering in human forearm tissue. Biomed. Opt. Express. 15, 2152–2167 (2024).

Funding

This work was supported by National Science and Technology Council of Taiwan (NSTC 110-2221-E-A49-059-MY3) and Ministry of Education (MOE) of Taiwan (VGHUST112-G3-2-2).

Author Contribution

V.H. performed the experiment, developed data acquisition algorithms, analyzed the data and prepared/wrote the manuscript. Y.Y.H and L.-W.C. analyzed and discussed the manuscript. C.-Y.L. and S.-T.C. discussed and interpreted the results. H.-H.C. discussed the method and results. S.-J.C. supervised the study, reviewed and revised the manuscript. All authors reviewed the manuscript.

Competing Interests

The authors declare no competing interests.

Figures and legends

Yes