Guilfoile Connor, Rampant Peter, House Michael
Genesis Cancer Care WA, 24 Salvado Road, Wembley, WA, 6014, Australia.
School of Physics, University of Western Australia, 35 Stirling Hwy, Crawley, WA, 6009, Australia.
Australas Phys Eng Sci Med. 2017 Jun;40(2):385-394. doi: 10.1007/s13246-017-0543-5. Epub 2017 Apr 4.
The presence of metal artefacts in computed tomography (CT) create issues in radiation oncology. The loss of anatomical information and incorrect Hounsfield unit (HU) values produce inaccuracies in dose calculations, providing suboptimal patient treatment. Metal artefact reduction (MAR) algorithms were developed to combat these problems. This study provides a qualitative and quantitative analysis of the "Smart MAR" software (General Electric Healthcare, Chicago, IL, USA), determining its usefulness in a clinical setting. A detailed analysis was conducted using both patient and phantom data, noting any improvements in HU values and dosimetry with the GE-MAR enabled. This study indicates qualitative improvements in severity of the streak artefacts produced by metals, allowing for easier patient contouring. Furthermore, the GE-MAR managed to recover previously lost anatomical information. Additionally, phantom data showed an improvement in HU value with GE-MAR correction, producing more accurate point dose calculations in the treatment planning system. Overall, the GE-MAR is a useful tool and is suitable for clinical environments.
计算机断层扫描（CT）中金属伪影的存在给放射肿瘤学带来了问题。解剖信息的丢失和不正确的亨氏单位（HU）值会导致剂量计算不准确，从而使患者治疗效果欠佳。为解决这些问题开发了金属伪影减少（MAR）算法。本研究对“智能MAR”软件（美国伊利诺伊州芝加哥市通用电气医疗集团）进行了定性和定量分析，以确定其在临床环境中的实用性。使用患者和体模数据进行了详细分析，记录启用GE-MAR后HU值和剂量测定的任何改善情况。本研究表明，金属产生的条纹伪影的严重程度在质量上有所改善，便于更轻松地进行患者轮廓勾画。此外，GE-MAR成功恢复了先前丢失的解剖信息。此外，体模数据显示经GE-MAR校正后HU值有所改善，在治疗计划系统中产生了更准确的点剂量计算结果。总体而言，GE-MAR是一种有用的工具，适用于临床环境。
