Abstract |
Over the last decades, extensive research has been conducted to develop novel optoacoustic (OA)
diagnostic systems for biomedical applications. OA imaging and sensing is a hybrid technique relying
on optical excitation and acoustic detection. It is based on the formation of acoustic waves following
a material's absorption of intensity-modulated optical radiation. The amplitude of OA waves is directly
proportional to the medium's absorption coefficient for the employed excitation wavelength.
Therefore, the technique provides molecular specificity based on optical absorption by a material or
biological tissue. Optoacoustic imaging and spectroscopy represent one of the most dynamic sectors
of biomedical engineering, fostering biomedical research for preclinical and clinical applications.
Nevertheless, in most cases, OA setups are bulky, complex, and costly, as they typically require the
integration of expensive nanosecond lasers with limited wavelength availability. Therefore, to extend
the applicability of OA applications in both preclinical and clinical directions, it is necessary to increase
portability, improve multispectral capabilities, and drastically reduce the cost of OA modalities. Highpower
light-emitting diodes (LED) can facilitate this goal since they come in various wavelengths, are
remarkably cost-efficient compared to lasers, and can generate OA signals in tissue-mimicking
phantoms or biological tissues.
Within this framework, in this study, the spectroscopic potential of high-power LEDs in OA
configurations has been systematically investigated, leading to the development of a portable LEDbased
OA sensing system for quantitative measurements and biomarker monitoring. OA sensing
techniques and LEDs’ spectral unmixing potential were evaluated with tissue-mimicking phantoms
and in vivo. By employing low-cost and commercially available electronic components, two highpower
LED sources, and a single-element piezoelectric ultrasonic transducer for signal detection, the
developed prototype has the potential to provide quantitative OA measurements. Initially, the LEDs'
performance in pulsed mode operation and their OA generation efficiency were investigated. The
performance of the OA sensing system in terms of optical spectral emission and pulse characterization
was also evaluated.
Furthermore, the system’s spectral unmixing capabilities were explored using gelatin-based tissuemimicking
phantoms demonstrating both optical absorption and scattering properties. Two Series of tissue-mimicking phantom samples were generated using gelatin as a buffer medium, Indian inks as
absorption agents, and Intralipid solution to introduce optical scattering. The first tissue-mimicking
phantom series (Series I) involved phantoms demonstrating predominant optical absorption
properties with negligible scattering. Two different types of Indian inks were added to varying amounts
in the gelatin solution to form mixtures of various relative ink concentrations. The second phantom
series (Series II) was generated by preparing a gelatin-Intralipid buffer aqueous solution to introduce
significant optical scattering. Cylindrical tubes of gelatin-ink mixtures were embedded in the buffer
solution to mitigate blood vessels inside scattering tissue media. Series I phantoms were used for a
parametric study of the OA response as a function of a) absorber (ink) concentration and b) applied
energy fluence in order to evaluate the degree of linearity among the involved physical quantities, as
predicted by the standard theoretical treatment of OA effect. Series I and Series II phantom samples
were developed to test linear spectral unmixing methodology in optoacoustic measurements. Six
different relative ink concentrations were selected to realize the unmixing experiments in each
phantom series. The absolute differences from the reference values of the OA spectral unmixing
estimations in Series I phantoms ranged from 0.7 to 17 %. In Series II, phantoms exhibiting both
optical scattering and absorption properties have been sufficient to determine the relative
concentrations of the absorbing inks with absolute differences from the reference values ranging from
0.4 to 12.3 % at maximum.
The research findings of the phantom study led to the development of an upgraded portable LEDbased
OA sensing system capable of monitoring hemodynamic changes and critical biomarkers in
vivo. The upgraded system incorporated high-power LEDs that were overdriven and operated in
pulsed mode by a custom-developed, inexpensive, printed circuit board (PCB). OA signal detection,
amplification, and data acquisition are also based on affordable, custom-developed, or commercially
available electronic devices. A specially 3D designed and 3D printed finger probe was utilized for in
vivo OA measurements in transmission mode, i.e., light excitation and ultrasound OA detection are
on opposite sides of the finger. The system effectively monitored hemodynamic changes and blood
microcirculation in human index fingers with high sensitivity and accuracy. Its capability to
simultaneously record the amplitude and the detection time of flight of the generated OA signals
enables concurrent estimation of essential biomarkers. It is able to monitor in vivo several biomarkers
related to hemodynamic changes and tissue vasodilation changes under vascular occlusion conditions The system was tested in real conditions by engaging eight healthy volunteers in a series of noninvasive
measurements on their right hand index fingers. Heart rate pulsation affects the amount of blood in
the finger and, consequently, induces a periodical tissue dilation. Heart rate was systematically detected
by analyzing OA signal peak detection time of flight (TOF) oscillations due to finger tissue dilation.
Heart rate measurements were systematically validated with simultaneous measurements by a medicalgrade
pulse oximeter.
Venous occlusion (VO) challenge tests were utilized to estimate blood flow and finger tissue
vasodilation. Blood flow (BF) could be assessed in two different ways. Firstly, the linear OA signal
increase rate during venous occlusion was employed to estimate BF in a manner similar to that
proposed in other studies. Moreover, BF was calculated using a new methodological approach
proposed by this study, which is based on directly estimating the rate of blood volume changes during
venous occlusion. The approach utilizes TOF variations and geometrical modeling of the OA-excited
part of the finger. This approach allows for modeling the finger’s physical dimensions and monitoring
tissue vasodilation during hemodynamic changes. BF calculations based on this approach were in
accordance with the physiological blood flow range reported in the literature for human fingers.
Furthermore, a linear spectral unmixing methodology was used to calculate the finger’s skin tissue
oxygen saturation based on OA measurements. This methodology was utilized to monitor the
percentage of oxygen saturation difference %ΔSO2 during hemodynamic changes imposed by venous
or arterial occlusion conditions during vascular occlusion tests. In VO tests, oxygen saturation change
from the average %SO2 baseline values demonstrated an apparent increase for all volunteers. This
increase was promptly succeeded by a rapid decline after the cessation of the occlusion. In arterial
occlusion (AO) tests, a decrease in oxygen saturation is observed during occlusion conditions. Oxygen
saturation decrease is followed by a rapid increase immediately after the cessation of the occlusion
during the so-called reactive hyperemia phase. During AO experiments, recovery slope (RS) and
recovery time (RT) commonly used biomarkers related to the reactive hyperemia phase that follows
an arterial occlusion challenge test were also estimated. These measurements were consistent with the
results presented in previous studies, demonstrating the promising clinical potential of the system for
skin microcirculation and vascular assessment.
In conclusion, this study presents, for the first time, a compact and inexpensive LED-based OA
sensing system for OA spectroscopic in vivo applications. Analysis of the amplitude and the TOF of
optoacoustic signals enabled monitoring of essential biomarkers, including heart rate, local oxygen saturation, tissue vasodilation, blood flow, recovery slope, and recovery time of the arterial occlusion
reactive hyperemia phase. The system could become a valuable and reliable tool for clinical studies
assessing arteriovascular diseases and skin microcirculation. Furthermore, its applicability could be
extended to clinical studies to evaluate hemodynamic changes associated with injuries, tissue
inflammatory responses, and vasodilator drug assessment
|