
Despite many years of basic and clinical research, cancer is a major health problem worldwide.1 Early diagnosis of growing cancer and rapid treatment is key to extend the life and improve the quality of life. Different imaging modalities such as computed tomography (CT), magnetic resonance imaging (MRI), positron emission tomography (PET), single-photon emission computed tomography (SPECT), and ultrasound imaging (UI) have been used to aid clinicians in the diagnosis, staging, and treatment of human cancers.2 However, although this method is excellent for diagnosing cancer, it is difficult to detect early cancer due to its low resolution, and it is difficult to accurately remove the lesion during surgery, resulting in many side effects.
Therefore, many research groups have showed a lot of interest in research for early detection of cancer and real-time identification of lesions during surgery, and studied optical imaging as an alternative to solve them. Compared established cancer imaging modalities, optical imaging has high resolution to probe functional and structural changes in real time and is a
Among them, OCT and fluorescence imaging are representative methods that can identify the boundaries of cancer in real time and identify lesions during surgery. OCT is well known for retinal imaging in ophthalmology5 and studies have shown that OCT has improved diagnostic efficacy in various structures such as the skin, gastrointestinal, respiratory, genitourinary tracts and the oral cavity.5-13 Recently, OCT has been proposed for the early detection of cancer by distinguishing tumorous and non-tumorous tissues with its high resolution at large (up to 2 to 3 mm) penetration depths. OCT can be used guided biopsy and can aid in the intraoperative imaging of cancer by providing real-time feedback to surgeons. It can also be used for monitoring of tumor responses to treatments (e.g., radiotherapy, chemotherapy, photodynamic therapy (PDT), and ablative (thermal- or cryo-) therapy.2,14-16
Fluorescence imaging (FI) is one of the most popular method for the visualization of cells, tissues, and biological processes taking place in a living organism. Fluorescence images can be produced from microscopy, imaging probes, and spectroscopy.17,18 In particular, near-infrared (NIR) fluorescence imaging has been playing a significant role in the field of tumor-specific fluorescence-guided surgery (TS-FGS).19 NIR fluorescence provides high-resolution images and penetrates into blood and tissues more than a few hundred microns since the optical window of 650 to 1350 nm has maximum depth of penetration in tissues and relatively low scattering and autofluorescence of biomolecules. With these advantages, NIR fluorescence imaging can provide real-time image guidance to surgeons for finding structures that need to be resected, such as tumor margins and sentinel lymph nodes.20
The purpose of this review is to introduce the operating principle of the OCT and fluorescence imaging systems mentioned above in the optical imaging system, which is the core for image guided surgery, and the applicability of the system by making an animal model. First, an introduction of the implemented OCT and an animal model injected with human malignant melanoma cell were evaluated, and the effectiveness of the OCT was evaluated and compared with pathological data. Second, the construction and operation of the fabricated fluorescence imaging system were introduced. Using the implemented system, indocyanine green (ICG) was injected intravenously into a mouse model and the performance of the system was confirmed through real-time images moving along the blood vessels. In addition, using a fluorescence imaging system, the possibility of use in surgery was confirmed in real time through the detection of sentinel lymph nodes, which can be very important for cancer staging. Finally, we would like to confirm the importance of the research group’s image guided surgery by introducing the latest research trends in optical imaging.
Fig. 1 depicts the swept-source optical coherence tomography (SS-OCT) system, which was constructed based on a Mach-Zehnder interferometer. The system consists of five modules: a wavelength swept source, an optical fiber interferometer, a scanner, a controller, and a digitizer, used for acquiring and displaying OCT volume images. In order to use the OCT system to identify lesions during surgery, a fast scanning speed is required. Therefore, in this system, a wavelength-swept laser source (SL132120; Thorlabs Inc., Newton, NJ, USA) with a center wavelength of 1310 nm, a maximum half width of 100 nm, and a scan rate of 200 kHz was used as the light source. The laser beam from the swept source was divided into a sample arm and reference arm through a 75:25 fiber coupler (TW1300R3F2; Thorlabs), which was guided to the optical fiber Mach–Zehnder interferometer. Fig. 1 shows a sample arm irradiated light to the sample, and the sample arm was fabricated with a Galvo-mirror and scan lens (LSM54-1310; Thorlabs) using a 3-D printer. The reference arm consists of a collimator, lens, iris, and mirror, and this was used as the reference beam. And the iris was used to control to have the appropriate optical power in the reference arm. After the light is reflected from the sample and the reference arm, the lights interfered with each other through a 50:50 coupler. This interference signal was detected by the balanced photodetector. The spectral interference signal which was converted into a voltage was transmitted to a 12-bit high-speed digitizer. To implement the fast Fourier transform, the spectrum interpolation was previously performed on all A-scans for resampling k-space. The number of A-scans samples was 500 and the scan range of the system was 12 mm × 12 mm. The 2-D cross-sectional image was constructed by using 512 B-scans and the 3-D volume images utilized 512 cross-sectional images. The axial resolution and lateral resolution are 11 μm and 22 μm in the air.
A schematic diagram and image of the fluorescence imaging system with LED lightening is shown in Fig. 2. The fluorescence imaging system consisted of LED lightening, camera module with lens and band pass filter. As the wavelength of the LED, an LED having a wavelength of 780 nm, which is the absorption wavelength of ICG, was used. In addition, as the band pass filter, a long pass filter that passes over 800 nm, the fluorescence wavelength of ICG, was used. To secure the entire image of the mouse model, a lens with a focal length of about 200 mm was used and mounted on the front of the camera module. In order to deliver a constant light to the sample, a reflector was manufactured and a lens was attached to each LED light to prevent light from spreading. Images acquired from the camera using a USB 3.0 port were signal acquired using a graphical user interface (GUI).
A total of 5 male Balb/c nude mice (aged 7 weeks) used in the experiment were approved by the Animal Protection Committee of Dankook University. To confirm real-time fluorescence and OCT images of lesions, mice were inoculated at two spots on the dorsal flank with 1×106 human malignant melanoma A-375 cells, diluted in 100 μL phosphate-buffered saline. The A-375 cells were grown and maintained in minimum essential medium (MEM) supplemented with 10% fetal bovine serum and 1% penicillin and streptomycin. Subcutaneous tumor growth and overall health were monitored twice a week.
Indocyanine green is a cyanine dye having a molecular weight of 774.97 g/mol. It is a negatively charged ion that belongs to the cyanine dyes (Fig. 3B).21 ICG (BioActs, RFP0815) was dissolved in dimethyl sulfoxide (DMSO) and diluted with PBS to the final concentration 10 μM for the experiments. To accurately measure the fluorescence wavelength for the absorption spectrum of ICG presented in the specification, 780 nm laser diode (LD) as shown in Fig. 3C was used, and the fluorescence spectrum was measured using the implemented fluorescence spectroscopy system. The wavelength range at full width half maximum of the measured fluorescence spectrum was 810 nm to 860 nm (Fig. 3D), confirming that it was a fluorescence signal suitable for the characteristics of the 800 nm long pass filter in the fabricated system.
Subcutaneous tumor was imaged with OCT system 12 days after injection. 3D volume and 2D cross-sectional images of cancerous region (Fig. 4D, F, G) were compared with the images of normal region (Fig. 4A-C). The cancerous tissue appears darker in the OCT image in part because it contains more blood than normal tissue. The OCT system used in this study employs a 1310 nm light source. At this wavelength, light absorption by blood is considerable, which produces the darker appearance of the malignant tissue.22 Additional differences between normal and cancerous regions was shown through statistical analysis of OCT images and corresponding histological sections (Fig. 5).
The red and yellow rectangles marked on the Fig. 5A correspond to histological images (Fig. 5B, C) and compared with each other. Histological images were taken using cross-sectional slicing and hematoxylin & eosin (H&E) staining to confirm the correlation with the OCT image taken previously. Generally normal tissue has well-organized tissue structure, while the structure of cancer tissue is often disorganized.2Fig. 5 shows the different characteristic organization and thickness of skin layers. Hypodermis, containing adipose tissue and muscle was below the dermis in normal region (Fig. 5C). In contrast, hypodermis except the muscle layer has become too thin to be seen in cancerous region (Fig.5B). These results suggest that OCT can provide diagnostic information for precancerous lesions by detecting morphological changes.22
However, the intensity image of OCT has a limitation in being able to identify the clear boundaries of cancer when compared with images obtained using pathology slides. Therefore, recently, many research groups are actively conducting research on diagnosing cancer boundaries in real time using polarization-sensitive OCT using polarization technique.23 Polarization-sensitive (PS) OCT is an imaging modality that provides large volumetric views of tissue microstructure with high resolution while simultaneously measuring birefringence of organized tissues such as collagen.24 PS-OCT clearly shows a distinction between solid carcinoma and surrounding fibrosis compared to structural OCT (Fig. 6A, B). The boundary between solid carcinoma and fibrosis clearly visualized with PS-OCT is confirmed with corresponding pathological slide images (Fig. 6C). These results suggest that PS-OCT could serve as a powerful imaging technology for assessing tissue acquisition sites within lung nodules by drastically enhancing differentiation between tumor and fibrosis. This work demonstrates the significant potential of using PS-OCT for biopsy guidance and enhanced cancer detection in the operation room.25
Fig. 7 depicts a fluorescence signal distribution in the mouse body after intravenous injection of ICG (150 μl of 10 μM ICG). From 5 minutes to 6 hours, the signals seems to gradually accumulate towards the liver, intestine, and after 6 hours, started to accumulate to the kidneys (Fig. 7A). 6 hours after the injection of ICG, we extracted organs from a mouse model and checked the intensity of ICG respectively. The level of ICG emitted from liver was the most highly detected, followed by small intestine and kidneys (Fig. 7B). Xenograft tumors in mice could hardly be detected by free ICG because of its rapid clearance and low accumulation in the tumor.26
To overcome these limitations, incorporating ICG into nanoparticle (NP) platforms is extensively studied. Various types of ICG-incorporated NPs have been developed and functionalized to embrace imaging and therapeutic techniques for cancer diagnosis and treatment. These ICG NPs show enhanced photostability, biocompatibility and tumor accumulation, low self-aggregation, and usually brighter fluorescence signal compared to free ICG dye due to the protective architecture.27 Zheng et al.28 prepared biodegradable folic acid-targeted NPs encapsulating ICG (FA-INPs) with intrinsic FA-targeting ligands. The FA-INP showed enhanced ICG stability, produced stronger temperature response than free ICG and exhibited significant targeting to breast tumors in
One example of intraoperative fluorescence imaging is sentinel lymph node (SLN) biopsy (SLNB) using fluorescent probes. So we generate xenograft models of malignant melanoma cancer by dorsal flank injection of A-375 cells to confirm the possibility of detection of sentinel lymph nodes in real time. 3 weeks after tumor cell injection,
We presented an overview of the optical imaging method capable of real-time monitoring of lesions and their applications. In order to acquire a real-time image of the lesion, a high-speed OCT having a speed of 400 fps was implemented using a high-speed swept source. In consideration of high-speed scanning of samples and convenience, a scanner was implemented using Galvo-mirror and 3D printer. In addition, a fluorescence imaging system optimized for the absorption or fluorescence wavelength of ICG, a representative contrast agent, was prepared for real-time fluorescence imaging. Considering user convenience, it is implemented as a hand-held type, and 780 nm LED is used as the light source.
To confirm the performance and surgical application of both systems, we prepared xenograft mouse models injected with human malignant melanoma cell and successfully acquired images using both imaging systems. Through real-time OCT images of lesions acquired in the xenograft mouse model, we saw the possibility of distinguishing between normal and cancerous tissues. However, it is difficult to determine the exact boundary of cancer based only on the intensity images of OCT, so many other research groups are conducting research on the boundary of cancer in real time during surgery using the OCT system using polarization. Therefore, we think that the real-time image confirmation of the lesion for OCT performed by this research group and the OCT technology using the polarization technique conducted by other research groups suggested the possibility of real-time classification of cancer boundaries in three dimensions. It is expected to be used in clinical practice in the future.
In addition, using ICG, a fluorescent agent widely used in clinical practice, ICG images were checked hourly through intravenous injection, and images of sentinel lymph nodes were acquired. ICG, a fluorescent agent, moved along the blood vessels, and it could be seen from the image that it was accumulated or escaped from each organ over time. This proved to be a key technology for image guided surgery as it enables real-time imaging of fluorescent agents during surgery. However, since ICG is not a tumor-specific agent, tumors generated in the xenograft mouse model could hardly be detected. However, as many research groups have recently introduced in this review, there are many studies on the development of tumor-specific agents and their application to image guided surgery.
As suggested, optical imaging systems, especially OCT systems and fluorescence imaging systems, can check non-invasive and
This work was supported by the Korea Medical Device Development Fund (KMDF) grant funded by the Korea government (KMDF_PR_20200901_0024). All experiments with mice were performed in the animal facility at Dankook University according to the guidelines of the Dankook University Institutional Animal Care & Use Committee.
![]() |
![]() |