CC BY
21CLINICAL APPLICATIONS OF OCT IN GASTROENTEROLOGY
NIJAS MOHAMED1, PAWAN KUMAR1, RENU JOHN1*, S. ANURADHA SEKARAN2, R. PRADEEP3, D.
NAGESHWAR REDDY4
1Medical Optics and Sensors Laboratory, Department of Biomedical Engineering, Indian Institute of Technology
Hyderabad, Telangana-502285, 2Pathology & Lab. Medicine, Asian Institute of Gastroenterology Hyderabad 3Surgical Gastroenterology, Asian Institute of Gastroenterology Hyderabad 4Medical Gastroenterology, Asian Institute of Gastroenterology Hyderabad *Email address for correspondence: renujohn@bme. iith. ac. in
ABSTRACT
Optical coherence tomography (OCT) has emerged as an important tool in bio-imaging. Being a non-invasive and realtime imaging technique, it offers many promising applications in the medical field. OCT is very effective for imaging micro-scale internal structures of the various biological samples up to a depth of 2-3 mm. The invention of this near-infrared light-based, low coherence interferometric imaging technique has revolutionized the field of three-dimensional and cross-sectional tissue imaging by bridging the resolution gap between microscopy and ultrasound imaging techniques. In recent times, OCT has been established as a reliable standard procedure for diagnosing ophthalmic anomalies like Age-related macular degeneration (AMD), diabetic retinopathy, corneal edema etc. The highly sensitive subsurface imaging capability of the OCT system makes it a potential tool for the diagnosis of gastrointestinal (GI) tract cancer at the very beginning stage. The micro-scale lateral and axial resolution of the OCT system provides an edge to it on the other GI tract imaging modalities in detecting early neoplastic changes in the mucosa of the GI tract. We report this study with the prime focus on morphological differentiation and clear margin assessment of the tissue to exhibit the potential of the OCT system for early diagnosis and guided surgery in the case of gastric cancer.
Keywords: Optical coherence tomography, Cancer imaging, Gastric cancer, Gastrointestinal tract imaging.
Introduction:
OCT was derived from the low coherence interferometry imaging concept based on a Michelson interferometer and shown for the first time in 1990 by Huang et al. At the Massachusetts Institute of Technology in Prof James Fujimoto's laboratory [1]. It was quickly discovered to be a useful tool for retinal imaging. The field of OCT imaging is now much more diversified, with systems used in medical imaging, art restoration, non-destructive testing, thin-film analysis, and other applications. Real-time processing of OCT data is now possible thanks to advances in parallel computing and GPU architecture. As a result, OCT has the potential to become a standard imaging modality in the near future [2,3].
Tomography is an imaging technology that uses numerous cross-sectional pictures to create 2D or 3D representations of the sample. Optical tomography is a type of tomography that uses photons to create images. Optical coherence tomography (OCT), Optical Diffraction Tomography (ODT), and Diffuse Optical Tomography (DOT) are the three primary forms of optical tomography [4]. OCT is one of the three that is particularly important for the following reasons:
• In OCT, the axial and lateral resolutions are decoupled or independent of one another.
• OCT provides millimetre-level depth resolution for in-vivo and ex-vivo whole-body imaging.
• OCT uses IR on near IR light for imaging, which has less scattering and has a higher penetration depth in biological tissues and is the only technology that delivers micrometre scale axial resolution for microscopic imaging structures.
• It is a non-invasive, label-free approach that can also be utilized for functional/multi-modal imaging, albeit exogenous contrast ants may be used if necessary
Low coherence sources like superluminescent diodes (SLD), wideband, or supercontinuum laser sources are used in OCT. This is due to the fact that the lateral resolution is inversely related to the light bandwidth [5]. A source with low spatial and temporal coherence is ideal for a high-resolution OCT device. However, there is typically an intrinsic trade-off between temporal and spatial coherence in most sources, as systems with low spatial coherence also have low temporal coherence. SLDs, which act in a similar way to laser diodes, are frequently used as sources in commercial systems. Because of their broad emission bandwidths and moderate temporal coherence, SLDs are a viable choice for OCT sources. Combining two or more SLDs with different emission bandwidths can result in extremely high
bandwidths and ultra-high-resolution imaging [6]. Broadband supercontinuum laser sources, on the other hand, are emerging as viable sources for OCT due to their massive emission bandwidths and better temporal coherence characteristic of laser sources.
Figure 1. Basic OCT set up using Michelson Interferometer
Time Domain OCT (TD-OCT) and Fourier Domain OCT are the primary imaging modalities used in OCT (FD-OCT). The sample is placed in one of the arms of a balanced interferometer, and the interferogram is recorded with the help of a detector in TD-OCT. The refractive index variation in the depth profile of the sample is encoded in the interferogram, which will provide the variations in-depth, also known as an axial scan image or A-Scan image. A lateral scan or B -scan can be formed by raster scanning several A-scans along with the sample. Because each B-scan is a cross-sectional image, a volumetric image of the sample can be obtained by stacking B -scans.
On the other hand, FD-OCT records the spectrum using a dispersion element, commonly a grating. Later Fourier transform is applied on the spectrum to get the A-scan image. Hence, compared to TD-OCT, this gives higher scan rates and a significant SNR advantage to FD-OCT. OCT has a wide range of functional modalities. They get functional information by utilizing one or more of the sample's optical characteristics. Spectroscopic OCT, for example, employs spectral features such as wavelength-dependent absorption or scattering to acquire functional information from a sample, such as the concentration of a particular component [7].
Although OCT has a wide range of applications, it used to be an expensive optical setup, which was one of its main drawbacks and is the key reason why OCT is still not widely used in medical imaging. However, the cost of OCT systems is decreasing now as new technologies develop, such as fibre-based supercontinuum sources, fibre-based systems, integrated optical components, etc. Another issue is the OCT size of the OCT system. With the advent of small, fibre-based OCT devices [8], even this problem has an optimal solution. The development of integrated optical devices such as fibre mirrors, circulators, and other components is projected to reduce OCT system size further. OCT modules the size of a briefcase have previously been demonstrated, including a battery-operated source that makes the device portable [9].
Stomach cancer is the world's fifth most prevalent cancer and the third most significant cause of cancer-related death. Males are twice as likely as females to have it. Eastern Asia has the highest incidence rates of all geographic locations, and the bacterium Helicobacter Pilori is the primary etiological factor in nearly 90% of cases [10]. The Epstein-Barr virus is another pathogen linked to stomach cancer. This pathogen is present in the malignant cells of 80 per cent of stomach carcinomas with lymphoid stroma but not in the normal epithelial cells. However, its significance in carcinogenesis is unknown [11]. Other proven risk factors for stomach cancer include poor eating habits, heavy alcohol usage, persistent smoking, and a lack of fruit intake. Early detection of stomach cancer can dramatically enhance survival chances. Stomach cancers are divided into two types based on where they occur: (i) non-cardia gastric cancer, which affects the distal region of the stomach, and (ii) cancers of the gastric cardia [12].
Because most patients with early-stage stomach cancer are asymptomatic, diagnosis is typically delayed until the disease has progressed. Anorexia, dyspepsia, weight loss, and stomach pain are the most typical symptoms at the time of diagnosis. Dysphagia may be seen in patients with tumours at the gastro-oesophageal junction or in the proximal stomach. Endoscopy and biopsy are commonly used to diagnose stomach cancer. The chief staging modalities for locally advanced gastric cancer are endoscopic ultrasonography and chest and abdomen CT scans. Laparoscopy is performed to
rule out the metastatic peritoneal disease with a limited volume. PET-CT and MRI aren't commonly used to diagnose and stage stomach cancer. However, there is mounting evidence that PET-CT can help with staging by detecting affected lymph nodes and metastatic disease. On the other hand, these tests are not always accurate, especially in individuals with mucinous tumours, as they may underestimate the severity of the disease [13,14].
In this paper, we report using a Spectral Domain OCT system to detect malignancies in the Gastrointestinal tract. We demonstrate the application of OCT for the demarcation of tumours in stomach tissues and the identification of malignancies, and the diagnoses were later reaffirmed by comparing with results of gold standard histopathology
Experimental Methods:
Surgical tissue samples were taken from patients who had been diagnosed with a gastrointestinal tumour. The omentum was connected to the gastrectomy specimen, which measured 30x22x4cms. One end of the stomach was stapled (distal resected end), while the other was opened and everted. The specimen was promptly moved to a standard saline medium to avoid tissue breakdown. For imaging, the same was quickly kept under SD-OCT. B scan imaging was used to identify normal tissues and tumorous sections of the samples in various places. The sample was thoroughly inspected to determine the various stages of the tumour. The sites were then marked with India ink to be easily identified during histology. After the images are obtained, the specimen is put to a formaldehyde solution. Then the sample is sent for histopathology, and images are compared with OCT images.
Results:
OCT images were captured and processed to analyse different stages of tumours in the obtained sample. Figure 2 shows the sample and the processed OCT images.
e.
f.
g.
Figure 2 (a) Shows the sample (oesophagal tumour tissue) cross-section where OCT is performed (b) Normal tissue OCT image (c) OCT image of the tissue in transition stage (d) OCT image of a fully developed tumour (e) Histopathology result of normal tissue f) histopathology result of marginal tissue (g) Histopathology shows
infiltrating tumour to the muscularis mucosa
Evident morphological differences between cancer and normal tissue could be observed. The presence of the gastric pits in the epithelium lining of the tissue indicates healthy mucosa. Neoplastic changes occur with the extinction of the gastric pits and entirely disappear in malignancy conditions. However, the morphological features observed in the cross-sectional image produced by the OCT system could not differentiate between dysplasia and cancer. After comparing the OCT images and histopathology results, we saw that the histopathology results confirm the OCT findings. The study reaffirms that the OCT can be used to detect gastrointestinal tumours in an early stage, as it can differentiate between normal, malignant, and marginally tumorous tissues. This can save a lot of time and effort as the OCT imaging can be done in-vivo compared to histopathology. Moreover, the system is real-time and can be used for quick imaging.
Conclusions:
The study reaffirms that the OCT can be used to detect gastrointestinal tumours in an early stage, as it can differentiate between normal, malignant, and marginally tumorous tissues. This can save a lot of time and effort as the OCT imaging can be done in-vivo compared to histopathology. Moreover, the system is real-time and can be used for quick imaging. The incorporation of artificial intelligence (AI) can further improve the diagnostic capability of OCT such that it can differentiate between dysplasia and cancer.
Reference:
[1] Fujimoto, James G., et al. "Optical biopsy and imaging using optical coherence tomography." Nature medicine 1.9 (1995): 970-972.
[2] Li, Xiqi, Guohua Shi, and Yudong Zhang. "High-speed optical coherence tomography signal processing on GPU." Journal of Physics: Conference Series. Vol. 277. No. 1. IOP Publishing, 2011.
[3] Jian, Yifan, Kevin Wong, and Marinko V. Sarunic. "Graphics processing unit accelerated optical coherence tomography processing at megahertz axial scan rate and high resolution video rate volumetric rendering." Journal of biomedical optics 18.2 (2013): 026002.
[4] Boas, David A., Constantinos Pitris, and Nimmi Ramanujam, eds. Handbook of biomedical optics. CRC press, 2016.
[5] Fujimoto, James G., et al. "Optical coherence tomography: an emerging technology for biomedical imaging and optical biopsy." Neoplasia 2.1-2 (2000): 9-25.
[6] Consortini, Anna. Trends in Optics: Research, Developments, and Applications. Academic Press, 1996.
[7] Kim, Jina, et al. "Functional optical coherence tomography: principles and progress." Physics in Medicine & Biology 60.10 (2015): R211.
[8] Schmitt, Joseph M. "Optical coherence tomography (OCT): a review." IEEE Journal of selected topics in quantum electronics 5.4 (1999): 1205-1215.
[9] Kim, Sanghoon, et al. "Design and implementation of a low-cost, portable OCT system." Biomedical optics express 9.3 (2018): 1232-1243.
[10] Bray, Freddie, et al. "Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries (vol 68, pg 394, 2018)." Ca-a Cancer Journal for Clinicians 70.4 (2020): 313-313.
[11] Wu, Ming-Shiang, Chia-Tung Shun, Chung-Chun Wu, Tsuey-Ying Hsu, Ming-Tsan Lin, Ming-Chu Chang, Hsiu-Po Wang, and Jaw-Town Lin. "Epstein-Barr virus—associated gastric carcinomas: relation to H. pylori infection and genetic alterations." Gastroenterology 118, no. 6 (2000): 1031-1038.
[12] Layke, John C., and Peter P. Lopez. "Gastric cancer: diagnosis and treatment options." American family physician 69.5 (2004): 1133-1140.
[13]Altini, Corinna, Artor Niccoli Asabella, Alessandra Di Palo, Margherita Fanelli, Cristina Ferrari, Marco Moschetta, and Giuseppe Rubini. "18F-FDG PET/CT role in staging of gastric carcinomas: comparison with conventional contrast enhancement computed tomography." Medicine 94, no. 20 (2015).
[14] Waddell, T., M. Verheij, W. Allum, D. Cunningham, A. Cervantes, and D. Arnold. "Gastric cancer: ESMO-ESSO-ESTRO Clinical Practice Guidelines for diagnosis, treatment and follow-up." Annals of Oncology 24 (2013): vi57-vi63.
