DWI影像與CT影像之融合技術於輔助早期肺癌診斷之應用

Application of Fusion Technique in DWI and CT Image to Assist Diagnosis Lung Cancer in Early Stage

張育培(Yu-Pei Chang)

指導教授 ： 蘇振隆

中原大學/工學院/生物醫學工程研究所/碩士(2013年)

https://doi.org/10.6840/cycu201301054

摘要

肺癌是國人最容易罹患的癌症之一，其中早期發現及早治療能有效降低死亡率，並有效提升五年存活率。臨床上常用正子攝影結合電腦斷層(PET-CT)來進行早期肺癌偵測，然PET含有較高之輻射性。故本研究目的為結合核磁共振擴散加權影像(DWI)與電腦斷層影像(CT)，同時提供解剖與功能性影像資訊，輔助醫師進行肺癌早期診斷，並有效減少病人輻射劑量。
本研究利用10組病人之DWI與CT影像，並針對DWI影像進行中值濾波、直方圖等化等影像前處理，提升DWI影像在邊緣結構上的呈現，根據解剖位置在DWI與CT影像上選取三點參考點，利用三點將DWI影像進行雙線性內插法放大與CT影像進行對位和校正，並在定位完加入人為調整及未調整使用Alpha-Blending的方法使兩張影像融合。本研究利用假體和10組實體影像的評估及驗證。
結果顯示前處理部分本系統能容忍30%以下的高斯雜訊，且以灰階25-255為範圍之直方圖等化為最佳呈現DWI影像邊緣。在初步定位結果後，無論假體和實體影像的三點重心參考點誤差皆為0 mm，定位完後自動調整及半自動調整的影像融合誤差為2.343 mm及0.333 mm。
本系統利用半自動調整融合DWI與CT影像，結合兩者影像資訊，提供臨床醫師早期診斷肺癌及治療，並有效減少病人之劑量，未來能結合呼吸監控擷取影像與影像上之參考點，提升影像對位及融合系統準確性，幫助醫師在早期偵測肺癌之效果。

關鍵字

影像融合 ； 影像對位 ； 擴散加權影像

並列摘要

Lung cancer is one of the most common types of cancer in Taiwan. It is the most effective treatment when detecting lung cancer in early stage. Currently, PET-CT which generated high radiation dose is used in clinical diagnosis. In this study, the purpose in this study is to fusion the optimized diffusion weighted image (DWI) and computed tomography (CT) image that provides anatomical and functional information to assist doctor to diagnosis and reduce radiation dose for patient.
We used 10 patient’s DWI and CT images in this study. First, we used median filter and histogram equalization as pre-process method to enhance the structure and edge of DWI image. Then depend on the 3 anatomical points on DWI and CT images to use bilinear interpolation method to enlarge the DWI image and register the two images. After register, we use the Alpha-Blending to fusion the two images and compare auto and semi-adjust on the fusion of DWI and CT images. This study used phantom and 10 patient images to evaluate registration and fusion system.
The preliminary results show that this system can allow Gaussian noise less than 30% when using median filter, and the best edge of DWI image by using histogram equalization in gray level 25-255. The center of gravity’s error are all 0mm for phantom and patient image. After registration the errors of auto-fusion method and semi-fusion method were compared, and the result in auto and semi-fusion error are 2.343mm and 0.333mm.
In this study, we use semi-control to fusion DWI and CT image to provide anatomical and functional information to assist doctor to diagnosis and reduce radiation dose for patient. In the future, we can combine the breath monitor and mark point to get the DWI image that can promote the registration and fusion accuracy. And promote the effective to assist doctor to diagnosis lung cancer in early stage.

並列關鍵字

Diffusion Weighted imaging ( DWI ) ； Image registration ； Image fusion

參考文獻

[12] 黎尚昆，杜文鈞，安治宇，曾清秀，電腦斷層影像與超音波影像融合應用於肝臟穿刺導引 ，中央大學生物醫學工程研究所，2011年。

[30] 陳國銘，電腦斷層攝影與核磁共振之三維腰椎影像登錄與融合之研究，國立成功大學工程科學系碩士論文，2003年。

[3] Nomori H, Watanabe K, Ohtsuka T et al, “The size of metastatic foci and lymph nodes yielding false-negative and false-positive lymph node staging with positron emission tomography in patients with lung cancer,” J Thorac Cardiovasc Surg, 127: 1087-92, 2004.

[4] Roberts PF, Follette DM, Haag D et al, “Factors associated with false positive staging of lung cancer by positron emission tomography,” Ann Thorac Surg, 70: 1154-60, 2000.

[5] Ichikawa T, Erturk SM, Motosugi U et al, “High b-value diffusion-weighted MRI in colorectal cancer,” A.J.R., 187: 181-4, 2006.