Endoscopic photoacoustic and ultrasound imaging

Abstract

The photoacoustic (PA) effect is a physical phenomenon where the light energy irradiated to the material is converted into the form of acoustic energy. PA imaging (PAI) is an innovative imaging modality that acquires ultrasound (US) with the optical contrast image from the acoustic depth. PA microscopy (PAM) is a standard form of PAI technique enabling high-resolution imaging through optical focusing or acoustic focusing. In the last decade, PAM has made great strides in biomedical imaging based on high contrast and safety. In particular, PAM is characterized by the advantage of being able to acquire functional information (e.g., hemoglobin oxygen, metabolism) and molecular images in an in-vivo environment that could not be provided with conventional US images alone. Based on these capabilities, it is seamlessly integrated into the field where the existing US is used, contributing to improving the accuracy of diagnosis or surgical decisions. Many studies are being conducted to evaluate the accuracy of diagnosis by providing additional and important indicators in the differentiation of tumors such as breast, stomach, rectum, and thyroid, where the US modality was in charge in the past. As a result, PA imaging technology is expanding its scope as a clinical diagnostic device. Endoscopic ultrasound (EUS) is a non-invasive diagnostic device that is inserted using high-frequency (7.5 – 12 MHz) ultrasound. It is used to observe the tomographic structure of the lining of the digestive system such as the esophagus, stomach, duodenum, pancreas, large intestine, and rectum. In general, EUS is considered to have advantages in high resolution, deep imaging depth, and safety. It is used when a differential diagnosis is required for small lesions that are difficult to distinguish by X-ray or CT and enables real-time biopsy during medical procedures. However, EUS has suffered from low contrast and the absence of functional imaging capability as EUS acquired the contrast only from the physical property called acoustic impedance. To overcome these shortcomings, the PA/US endoscopy (ePAUS) incorporating PAI has been proposed. The ePAUS is a diagnostic device that adds the PA imaging capability to the traditional EUS, enabling high-resolution and optical contrast-based imaging without sacrificing US imaging function. The ePAUS can provide functional information based on a molecular investigation by PAI capability as well as structural information by the US in an in vivo environment. This dissertation covers a wide range of topics, from a feasibility study of whether ePAUS can be helpful in practical clinical diagnosis of the digestive system, to improving the performance of ePAUS using transparent transducers (TUT). The first part is about the feasibility study for PAI of the digestive system of large animals and the usefulness of diagnosing early gastric cancer (EGC) human-like porcine stomach ex vivo. Although research on the ePAUS system is being actively developed, imaging studies have been limited to small animals such as rodents and rabbits. In some studies, porcine stomach imaging has already been performed with the PAM system, but the spatial resolution is low and it is difficult to identify the stratified gastric structures and vasculature. Therefore, to investigate the PA/US imaging potential of gastric cancer, we developed a PA and ultrasound imaging system capable of imaging both histology and vasculature simultaneously. This system allowed us to obtain PA/US 3D volume data in a single imaging procedure on porcine stomachs ex vivo. The gastric wall was clearly distinguished from the tissue layer and blood vessels in the PA and US images, and these imaging results were histologically validated. In this study, we were able to obtain images of blood vessels from the gastric mucosal surface to a depth of 1.9 mm, demonstrating the potential of imaging down to the submucosal layer. Considering that the average thickness of the human mucosal layer is 1.26 mm, the mucosa and submucosal layer can be completely visualized, which is expected to enable differentiation of Tis, T1a, and T1b tumors in EGC stage evaluation. Distinguishing T1a from T1b is very important as an indicator to determine the clinical scenario of endoscopic submucosal dissection (ESD) or surgical resection. In addition, since there is a high correlation between angiogenesis in the submucosal layer and metastasis to lymph nodes of cancer, it may be useful to determine whether the tumor has invaded the lymphatic vessels through submucosal vascular imaging. The second part is the development of endoscopic photoacoustic and ultrasound using TUT. As mentioned before, ePAUS is one of the most promising applications of PAI, which can reflect the advantages of functional and high-resolution imaging while maintaining the traditional EUS imaging method. However, inherently, the existing ePAUS requires a complex beam combiner to create an overlapping region of a light beam and an acoustic beam in a limited probe space, otherwise, there is a problem of using two beams in an off-axis state. The off-axis between the optical and acoustic axes causes the low signal-to-noise ratio (SNR) and limited depth of field (DOF) problems. In this study, a small TUT that can be inserted into a probe was fabricated and applied to ePAUS to achieve coaxial alignment of the beam and acoustic beam. We also combined multimode fiber (MMF) and gradient index (GRIN) lenses to implement a quasi-OR mode with a deeper DOF to maximize the benefits of TUT applications. In conclusion, we have successfully developed a compact TUT-based ePAUS called ePAUS-TUT. The developed ePAUS-TUT enables PA imaging at close range and maintains resolution performance with a radial resolution of 30 µm and an azimuth resolution of 50 µm. Imaging capabilities were demonstrated by imaging the leaf skeleton phantom and the colon and vagina of rats in vivo. In the in vivo imaging, we were able to clearly visualize the vasculature of the lumen. By applying TUT to ePAUS, miniaturization, high-resolution imaging capability, deep DOF, and a simple production process have been achieved. This not only provides better SNR and clear image but is also expected to be useful in the process of production and mass production.