探討使用雙能量電腦斷層攝影於消除牙體金屬假影之影像品質評估

Reduction of dental filling metallic artifacts by using dual-energy computed tomography

李政穎(李政穎 Zheng-Ying Li)

指導教授 ： 李尚熾

中山醫學大學/醫學科技學院/醫學影像暨放射科學系/碩士(2019年)

https://doi.org/10.6834/CSMU201900104

摘要

前言：
影像對於醫療診斷是不可或缺的必要工具，提供高品質影像是專業放射師的重責大任。然而金屬物質對電腦斷層的影像品質是一大傷害，如何改善金屬假影，提升影像品質一直以來都是學者們共同努力的方向。近年電腦斷層快速進步，雙能量電腦斷層掃描儀DECT(Dual Energy Computed tomography)能有效降低雜訊並提升影像對比度，提供優異的影像品質。此外，還能透過雙能量的X光分離金屬資訊，並加以改善，是消除金屬假影的一大利器。

研究目的：
口腔部位常有牙體金屬植入物，在電腦斷層影像出現金屬假影，影響臨床診斷品質。本研究使用DECT搭配多種參數掃描，減少金屬假影，並試圖找到最佳的掃描參數。

材料與方法：
以標準頭部假體進行實驗，嵌入4顆牙體金屬，DECT掃描重組權重影像 (Weighted images)、模擬單能影像Mono (Monochromatic images)和消除金屬假影軟體MARS (Metallic Artifact Reduction Software)進行影像後處理。將各影像使用MATLAB定量分析，比較影像對比度CNR (Contrast-Noise Ratio)與金屬假影嚴重程度AI (Artifact Index)，並請兩位資深放射科醫師定性分析，對不同掃描參數影像品質進行主觀評分。

結果：
定量分析結果顯示120 kVp、0.7權重和Mono低能量影像有較佳的CNR；140 kVp、0.3權重和Mono高能量影像有較佳的AI值，而iMAR只在軟組織CNR項目中有效，其餘項目皆使影像品質下降。定性分析顯示高能量影像能消除金屬假影，而後處理軟體iMAR (iterative Metal Artifact Reduction)無法提升影像診斷品質。Mono影像在70~90 keV可同時診斷頭頸部的軟組織和硬組織，再用MDT (Metal Deletion Technique) 後處理修正法具有優異的影像品質。

結論：
雙能Mono影像在70 ~ 90 keV搭配MDT後處理修正法，同時具備診斷軟組織與高密度組織能力，有效降低金屬假影且保持優異影像對比度。臨床應同時參考原始影像，以及補充資訊用途的MARS修正影像，作為診斷依據。

關鍵字

雙能量電腦斷層 ； 模擬單能影像 ； 金屬假影消除 ； 疊代法 ； 影像品質

並列摘要

INTRODUCTION: The medical image plays an important role for clinical diagnostic of patients and acquired high-quality images is a major responsibility for professional radiologic technologists. However, some materials such as metal were produced severe streak artifacts to reduce image quality in computed tomography. In recent years, the rapidly advancement of computed tomography had been developed, which called dual-energy computed tomography (DECT), not only effectively reducing noise but also improving image contrast. In addition, metallic artifacts also could be potentially separated by using DECT.

PURPOSE: Dental metallic implants in the oral cavity are commonly produced metal artifacts in computed tomography. Therefore, this study aims to investigate the optimization protocols to reduce metallic artifact by using different scan settings of DECT.

MATERIAL AND METHODS: This study used a standard head phantom inserted into four metal implants, and reconstructed weighted images (0.3, 0.5, 0.7), Monochromatic images (Mono) of DECT, respectively. Metallic Artifact Reduction software (MARS) was also used to compare efficiency. Quantitatively analysis was measured by using contrast-noise ratio (CNR), and metal artifact index (AI). Two senior radiologists qualitatively analyzed the subjective scores of the image quality.

RESULTS: Quantitative analysis showed that use of 120 kVp or weighted images 0.7 or Mono with low-keV images had higher CNR; 140 kVp or weighted images 0.3 or Mono with high-keV images had better AI. While MARS was only effective for soft tissue in CNR, and the others shows lower image quality. Qualitative analysis showed that high-energy images could reduce metal artifacts, but MARS could not directly improve image diagnostic quality. Soft and bone tissues could be diagnosed simultaneously by using Mono images with 70 to 90 keV in DECT.

CONCLUSION: Mono images with 70 to 90 keV has the ability to diagnose both soft and bone tissues, effectively reducing metal artifacts while excellent CNR can be maintained.

並列關鍵字

dual energy computed tomography (DECT) ； Mono images ； MARS ； iterative algorithm ； image quality

參考文獻

1. Hegazy, M.A.A., et al., Dual-energy-based metal segmentation for metal artifact reduction in dental computed tomography. Med Phys, 2018. 45(2): p. 714-724.

2. Barrett, J.F. and N. Keat, Artifacts in CT: recognition and avoidance. Radiographics, 2004. 24(6): p. 1679-91.

3. Cha, J., et al., Dual-energy CT with virtual monochromatic images and metal artifact reduction software for reducing metallic dental artifacts. Acta Radiol, 2017. 58(11): p. 1312-1319.

4. De Crop, A., et al., Analysis of metal artifact reduction tools for dental hardware in CT scans of the oral cavity: kVp, iterative reconstruction, dual-energy CT, metal artifact reduction software: does it make a difference? Neuroradiology, 2015. 57(8): p. 841-9.

5. Long, Z., et al., Evaluation of projection- and dual-energy-based methods for metal artifact reduction in CT using a phantom study. J Appl Clin Med Phys, 2018. 19(4): p. 252-260.