Download presentation
Presentation is loading. Please wait.
1
醫療影像處理在診斷上之應用 嘉義大學資工系 教授 柯建全 時間 : 2009 年 11 月 13 日
2
Outline Introduction Object of medical image processing Imaging devices applications Related techniques for Medical imaging Research Results Future works
3
Introduction What is Medical imaging? Why do we need digital image processing? What kind of problems are often caused in medical images? Blurring caused by respiratory or motion Low contrast caused by imaging device or resolution Complicated textures Research trends have been transferred from 2-D to 3-D reconstruction
4
Introduction (continue) Integrate all possible methods in the filed of DIP, pattern recognition, and computer graphics Qualitative Quantitative Three categories of imaging in different modalities Structural image Functional image Molecular image
5
Object Help physicians diagnose Reduce inter- and intra-variability Produce qualitative and quantitative assessment by computer technologies Determine appropriate treatments according to the analyses Surgical simulation or skills to reduce possible errors
6
Medical Imaging Modalities X-ray Ultrasound: non-invasive Computed tomography Magnetic resonance imaging SPECT (Single photon emission tomography) PET( Positron emission tomography) Microscopy
8
X-ray
9
Ultrasound 2-D sonography 3-D sonography Doppler color sonography A series of 2-D projection Reconstruction 4-D sonography
12
Computed tomography
13
MRI 可以觀察活體三度空間的斷層影像 磁振影像取影像時可以適當控制而得到不 同參數的影像,如溫度、流場 (flow) 、水 含量、分子擴散 ( diffusion) 、 灌流 (perfusion) 、化學位移 (chemical shift) 、 功能性 (functional MRI) 及不同核種如 氫、碳、磷
14
MRI-structural and functional image
16
Related techniques Image processing Segmentation Registration Feature Extraction Shape feature Texture Motion tracking Pattern recognition Supervised learning Un-supervised learning Neuro network Fuzzy Support vector machine(SVM) Genetic algorithm
17
Related techniques 3-D graphic Virtual diagnose or visualization Fusion between different modalities Bio-medical visualization
18
SPECT-functional image
19
PET(Positron Emission Tomography ) PET 以分子細胞學為基礎,將帶有特殊標記的葡 萄糖合成藥劑注入受檢者體內,利用 PET 掃瞄儀 的高解析度與靈敏度作全身的掃描,藉由癌細胞 分裂迅速,新陳代謝特別旺盛,攝取葡萄糖達到 正常細胞二至十倍,造成掃描圖像上出現明顯的 「光點」 能於癌細胞的早期 ( 約 0.5 公分 ) 準確地判定癌細 胞,提供醫師作為診斷及治療的依據,診斷率高 達 87-91 %, 30 歲以上的成年人及有癌症家族史 的民眾,建議每隔 1 ~ 2 年做一次 PET 檢查。
20
PET (Positron emission tomography)
22
Cell identification via microscope Tools Traditional optical microscope Stained specimen Fluorescent microscope Identification for nuclear and gene expression Laser confocal microscope Identification from 2-D to 3-D Multi-photon microscope Identification from 2-D to 3-D
23
Applications in a hospital Assist surgeon plan surgical operation or diagnose Picture archiving system (PACS) 將醫療系統中所有的影像,以數位化的方式儲存,並經 由網路傳遞至同系統中,供使用者於遠側電腦螢幕閱讀 影像並判讀。 Telemedicine Surgical simulation: Medical Visualization , Surgical augmented Reality, Medical- purpose robot, Surgery Simulation , Image Guided Surgery , Computer Aided Surgery Estimate the location, size and shape of tumor
24
PACS System
25
Virtual Surgery
27
Related techniques Classification of normal or abnormal tissues such as carcinoma Pre-processing: Contrast enhancement, noise removal, and edge detection Lesion segmentation: extract contours of interest thresholding 2-D segmentation 3-D segmentation based on voxel data Color image processing
28
Our study Virtual colonoscopy Bone tumor segmentation with MRI and virtual display Breast carcinoma based on histology and cytology Visualization of cell activities using confocal laser scanning microscope
29
Virtual colonscopy-Browsing or navigation within a colon Helical CT – patients injected contrast medium Re-sampling — Voxel-based Interpolation Automatic segmentation (seed) threshloding Determination of the skeleton of the colon Connected-Component Labeling Surface rendering and volume rendering Extraction of suspicious sub-volumes for diagnosis
31
Automatic segmentation
32
Determination of the skeleton of the colon
33
Display and measurement
34
Bone tumor segmentation with MRI and virtual display—Contrast medium Otsu thresholding Region growing Tri-linear interpolation Morphological post-processing Morphological post-processing Surface rendering Measurement
35
Histogram of T1 weighted and T2 weighted
38
(a) 0 度 (b) 45 度
39
Classification of Breast Carcinoma
42
正常異常 系統判斷為正常 126 系統判斷為異常 111 準確性 敏感度有效性 76.67%64.71%92.31%
44
應用雷射共軛焦顯微鏡影像三維重構研究螢 光細胞之活動 - 以子宮頸癌細胞為例
45
細胞結構簡介
46
雷射共軛焦顯微鏡之成像原理 46 雷射共軛焦顯微鏡解析度 :
47
雷射共軛焦顯微鏡雜訊生成之原因 大部分的生物樣本,潛在一些特性會降低 CLSM 影像解析度: 第一個是具有散射的特性 第二個特性是折射率的不匹配 (refractive- index mismatch) 所產生的。 由於折射率的不匹配會引入球面像差,而使得縱 向與橫向解析度變差。 47
48
散射的特性 48 散射光強度 :
49
散射的特性 49
50
研究影像 50
51
實驗方法與架構 51 選取重要影像 重要影像初始輪 廓偵測 Snake 自動偵測 重要影像輪廓 利用適應性 Snake 自動偵測 整體輪廓 排除對比較差的 細胞核輪廓 重建三維細胞 分割蛋白質劇烈 活動區域 二質化細胞質 區域
52
選取重要影像 由於雷射共軛焦取像環境的限制,在細胞邊界處通常訊號 較弱且較為模糊,使得初始輪廓分割相當困難,因此本步 驟選取出對比最好的影像作為重要影像 ( 分割時的初始切 片 ) ,偵測出其初始輪廓。 52
53
重要影像初始輪廓偵測 53
54
Snake 自動偵測重要影像細胞輪廓 - 去除人工雜物 由於雷射共軛焦取像的特性,其邊緣部分通常模糊不清, 因此初始輪廓的結果偶爾會產生過小得輪廓,本研究排除 輪廓長度小於 100pixel 得輪廓 ( 可能是雜訊的輪廓 ) ,僅以較 大的輪廓作為初始輪廓。 54
55
利用適應性 Snake 自動偵測整體影像輪廓 55
56
利用適應性 Snake 自動偵測整體影像輪廓 56
57
輪廓重疊偵測 - 濾除對比較差細胞核輪廓 偵測 細胞核訊號分佈模式 : 重要影像前半部 重要影像後半部 57 = 細胞質區域 Mean = 細胞核區域 Mean
58
輪廓重疊偵測 - 濾除對比較差細胞核 輪廓偵測 58
59
59
60
60
61
細胞內蛋白質反應劇烈區域的分割 61
62
細胞內蛋白質反應劇烈區域的分割 初始點決定與 K-Means 群法偵測最亮區域 62
63
不同角度顯示 63
64
其他範例 (Case3) 64
65
不同角度顯示 65
66
三維重建資料比較 66
67
體積比例量測 Case 1Case 2Case3 細胞質區域 (Voxels) 224396521563629562 蛋白質活動劇烈區域 (Voxels) 381933877785 比例 1.7019%0.649%1.236% 67
68
效能評估 ProcessCase1:Time(Sec)Case2:Time(Sec)Case3:Time(Sec) 細胞質區域分割 687483 蛋白質活動劇烈區 域分割 767 整體三維重建 9109 整體時間 849099 68
69
拉普拉斯三維平滑 69
70
拉普拉斯三維平滑 70
71
Requirements for medical image processing system in clinical diagnosis Automatic and less human interaction Qualitative and quantitative measurements Stable and reliable (experiments with much more cases) Performance evaluation True positive, true negative, false positive, false negative Accuracy, sensitivity, and specificity Receiving operating characteristic curve (An index for evaluating the effectiveness of classification Optimal classification threshold Area under ROC approach 1 – better classification
72
ROC curve
73
Analyses of prognosis on breast cancer for a stained tissue Microscopy with different resolution (400 or 100) for a stained tissue Fluorescent microscopy in detecting the number of chromosome Immunohistochemistry(IHC)
74
Her-2 IHC image
75
Fish image(normal)
76
Fish image (abnormal)
77
Preliminaries or problems ? Blurring often caused by patient motion or respiration Clinical opinion or idea obtained from an experienced surgeon Non-absolute answers at some specific conditions Trade-off between complexity and performance Large variations for different image modality
78
Preliminaries or problems ? Automation is necessary so as to help physicians Prove identification accuracy — comparison between manual and image processing approaches Classification based on neural network, pattern recognition, or fuzzy,.. etc is crucial in practical applications
79
Thanks for your attention!
Similar presentations