Cited 0 time in webofscience Cited 0 time in scopus

Multimodal Imaging and Analysis System for Cancer Diagnosis

Title
Multimodal Imaging and Analysis System for Cancer Diagnosis
Authors
Jihun Kim
DGIST Authors
Kim, Jihun; Lee, KijoonHwang, Jaeyoun
Advisor(s)
황재윤
Co-Advisor(s)
Kijoon Lee
Issue Date
2019
Available Date
2019-08-23
Degree Date
2019-08
Type
Thesis
Description
Multimodal imaging and analysis system, cancer diagnosis, optical imaging, high-frequency ultra-sound imaging
Abstract
Examination of tumor distribution and invasion depth in pre-operative or resected tissues is one of the crucial indicators used in cancer diagnosis to determine the therapy method. For this, biopsies are conventionally per-formed, to check for the presence and progress of the cancerhowever, they are time-consuming and are also subjective owing to observer differences and various environmental factors. To overcome these shortcomings, various imaging techniques have been developed, but they are primarily focused on improving a single imag-ing technique, providing only surface or depth information about the lesion, depending on the technique ap-plied. Recently, a few researchers have developed multimodal imaging systeMaster, but these systeMaster have demonstrated limited ability to deliver information on both tumor distribution and invasion depth simultane-ously. Despite many technical advances, cancer detection rates have not improved significantly because of these limitations. To address existing cancer diagnostic imaging system shortcomings, we developed a novel, multimodal imaging and analysis system, based on optical and high-frequency ultrasound imaging methods, which combines a form of microscope or a forward-looking endoscope. The system has been developed as a multimodal imaging system, based on a combination of autofluorescence-based multispectral, photometric stereo, high-frequency B-mode, acoustic radiation force impulse, and integrated backscattering coefficient imaging. The system has demonstrated the ability to provide information on tumor surface distribution and invasion depth, in addition to 3-D topography information—which are the critical information related to the diagnosis and treatment of cancer—simultaneously, using colorectal cancers excised from humans and mice. These results have shown that a multimodal imaging and analysis system, based on optical and high-frequency ultrasound imaging technology, can be an innovative diagnostic tool for cancer diagnosis, as well as for various other clinical applications.|수술 전 또는 절제 된 조직의 종양의 표면 분포 및 침습 깊이를 검사하는 것은 암 진단에서 치료 방법을 결정하는 데 아주 중요하다. 이를 위해, 암의 분포 및 침습 깊이를 검사 할 수 있는 생체 조직을 병변 부위에서 일부분 채취 하여 검사하는 생체 검사가 일반적으로 수행 된다. 그러나 관찰자의 경험 및 다양한 환경 요인에 따라 생체 검사는 시간이 많이 소요되고 주관적이라는 단점이 존재 한다고 보고되고 있다. 이러한 단점을 극복하기 위해 다양한 단일 영상 시스템이 개발되었다. 하지만 단일 이미징 시스템은 적용된 이미징 기법에 따라 병변에 대한 표면 정보 또는 깊이 정보만을 제공하는 한계점을 가지고 있다. 최근 소수의 연구 그룹에서 다중 모달 이미징 시스템을 개발 하고 있지만, 이들 시스템은 종양 분포와 침습 깊이에 대한 정보를 동시에 제공 할 수 없는 제한적인 능력을 보여주었다. 위와 같이 이미징 시스템에 대한 많은 기술적 진보가 나타났음에도 불구하고 제한된 이미징 기법에 의해 암 진단율이 크게 향상되지 않았다. 하여, 우리는 기존의 암 진단 영상 시스템의 단점을 극복하기 위해 광학 및 고주파 초음파 이미징 기법 기반 다중 모달 현미경 또는 전방 내시경 시스템을 개발 하였다. 이 시스템은 자가 형광 기반 다중 분광 (Autofluorescence based Multispectral), 측광입체시법(Photometric Stereo), 고주파 초음파 밝기 (High-frequency Ultrasound B-mode), 음향 방사력 (Acoustic Radiation Force Impulse) 및 후방 산란 계수 (Backscattering Coeffi-cient) 이미징 기법을 포함한다. 개발 된 시스템은 조직 모방 팬텀을 이용해 성능 평가를 하였으며, 이 후 암환자에게서 절제 한 종양 조직 검사에 적용하였다. 결과적으로 암의 진단 및 치료와 관련된 중요한 정보인 종양의 표면 분포 및 침습 깊이에 대한 정보를 제공 할 수 있는 능력을 보여주었다. 이러한 결과는 광학 및 고주파 초음파 영상 기법 기반 다중 모달 이미징 및 분석 시스템이 암을 진단하는데 있어 혁신 적인 진단 도구가 될 수 있음을 시사하였으며, 암 진단 이외에도 다양한 병변에 대한 진단에 적용 될 수 있음을 보여주었다.
Table Of Contents
Abstract ················································································ i Contents ··············································································· ii List of Figures········································································ vi List of Tables ······································································· viii Ⅰ. Introduction 1.1 Colorectal Cancer ····························································· 1 1.2 Examinations of Colorectal Cancer with Single Imaging Modality ··· 2 1.3 Necessity of Multimodal Imaging System ································· 4 1.4 Significance and Hypothesis ················································· 6 1.5 Overview of Thesis ··························································· 8 Ⅱ. Background 2.1 Optical Properties in Biological Tissues ··································· 9 2.1.1 Basic Physics of Light-Mater Interactions ·························· 9 2.1.2 Fluorescence ··························································· 12 2.2 Ultrasound Properties in Biological Tissues ···························· 13 2.2.1 Attenuation ····························································· 13 2.2.2 Scattering ······························································· 14 2.3 Principles of Imaging Modalities ········································· 18 2.3.1 Autofluorescence Imaging ··········································· 18 2.3.2 Multispectral Imaging ················································ 19 2.3.3 B-mode Imaging ······················································ 20 2.3.4 Acoustic Radiation Force Impulse Imaging ······················· 22 2.3.5 Backscatter Coefficient ··············································· 23 Ⅲ. A Multimodal Biomicroscopic System based on High-frequency Acoustic Radiation Force Impulse and Multispectral Imaging Tech-niques for Tumor Characterization Ex vivo 3.1 Introduction ·································································· 26 3.2 Materials and Methods ····················································· 30 3.2.1 Multimodal Biomicroscopic System based on High-frequency Ultrasound and Optical Imaging Techniques ······················ 30 3.2.2 Construction of a Tissue-mimicking Phantom and Preparation of a Colorectal Tumor for a System Evaluation ······················ 33 3.2.3 Multimodal Imaging and Analysis of the Phantom and Colorectal Tumors ······························································· 35 3.2.4 Statistical Analysis of Spectral Signatures and Mean Displace-ments of Colorectal Tumors and Normal Tissues ················· 36 3.3 Results ········································································ 36 3.3.1 Tissue-mimicking Phantom Study for an Evaluation of the Mul-timodal Biomicroscopic System ····································· 36 3.3.2 Multimodal Imaging of Colorectal Tumors Ex Vivo ············· 41 3.4 Discussion ···································································· 44 3.5 Conclusions ·································································· 50 Ⅳ. Multimodal Endoscopic System based on Multispectral and Photo-metric Stereo Imaging and Analysis 4.1 Introduction ·································································· 52 4.2 Materials and Methods ····················································· 55 4.2.1 System and Probe Design ············································ 55 4.2.2 White Light and Multispectral Imaging and Analysis ··········· 57 4.2.3 Photometric Stereo Imaging ········································· 58 4.2.4 Preparation of the Polyp-mimicking Phantom ···················· 60 4.2.5 Preparation of the Colorectal Cancer Model ······················· 61 4.2.6 Statistical Analysis ···················································· 62 4.3 Results ········································································ 62 4.3.1 Evaluation of the Performance of a Multimodal Endoscopic ··· 62 4.3.2 Evaluation of the Proposed System using the Polyp-mimicking Phantom ································································· 66 4.3.3 Multimodal Endoscopic Imaging of Colorectal Tumors ········· 67 4.4 Discussion ···································································· 70 4.5 Conclusions ·································································· 72 Ⅴ. Multimodal Endoscopic System based on High-frequency Ultra-sound and Multispectral Imaging Techniques for Tumor Detection 5.1 Introduction ·································································· 75 5.2 Materials and Methods ····················································· 77 5.2.1 System and Probe Design ············································ 77 5.2.2 White Light and Multispectral Imaging and Analysis ··········· 79 5.2.3 Development of the Endoscopic Ultrasound Transducer ········ 81 5.2.4 Development of the Ultrasound Pulser/receiver ·················· 82 5.2.5 Golay Code for Coded Excitation and Integrated Backscattering Coefficient ······························································ 84 5.2.6 Preparation of Gelatin Phantom, and Human Colorectal Tumor ············································ 85 5.3 Results ········································································ 86 5.3.1 Evaluation of the Proposed System using a Bilateral Gelatin Phantom ································································· 87 5.3.2 Multimodal Endoscopic Imaging of Human Colon Tissues including Tumors Ex vivo ··············································· 90 5.4 Discussion ꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏ 92 5.5 Conclusions ꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏꞏ 93 Ⅵ. Conclusions 95 Bibliography 98
URI
http://dgist.dcollection.net/common/orgView/200000220573
http://hdl.handle.net/20.500.11750/10479
DOI
10.22677/thesis.200000220573
Degree
Doctor
Department
Department of Information and Communication Engineering
University
DGIST
Related Researcher
  • Author Hwang, Jae Youn MBIS(Multimodal Biomedical Imaging and System) Laboratory
  • Research Interests Multimodal Imaging; High-Frequency Ultrasound Microbeam; Ultrasound Imaging and Analysis; 스마트 헬스케어; Biomedical optical system
Files:
There are no files associated with this item.
Collection:
Department of Information and Communication EngineeringThesesPh.D.


qrcode mendeley

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

BROWSE