Issa S, Scharfetter H
Institute of Medical Engineering, Graz University of Technology, Graz, Austria.
J Electr Bioimpedance. 2018 Dec 31;9(1):163-175. doi: 10.2478/joeb-2018-0021. eCollection 2018 Jan.
In biomedical MITS, slight unintentional movements of the patient during measurement can contaminate the aimed images to a great extent. This study deals with measurement optimization in biomedical MITS through the detection of these unpredictable movements during measurement and the elimination of the resulting movement artefacts in the images to be reconstructed after measurement. The proposed detection and elimination (D&E) methodology requires marking the surface of the object under investigation with specific electromagnetically perturbing markers during multi-frame measurements. In addition to the active marker concept already published, a new much simpler passive marker concept is presented. Besides the biological signal caused by the object, the markers will perturb the primary magnetic field inducing their own signals. The markers' signals will be used for the detection of any unwanted object movements and the signal frames corrupted thereby. The corrupted signal frames will be then excluded from image reconstruction in order to prevent any movement artefacts from being imaged with the object. In order to assess the feasibility of the developed D&E technique, different experiments followed by image reconstruction and quantitative analysis were performed. Hereof, target movements were provoked during multifrequency, multiframe measurements in the β-dispersion frequency range on a saline phantom of physiological conductivity. The phantom was marked during measurement with either a small single-turn coil, an active marker, or a small soft-ferrite plate, a passive marker. After measurement, the erroneous phantom signals were corrected according to the suggested D&E strategy, and images of the phantom before and after correction were reconstructed. The corrected signals and images were then compared to the erroneous ones on the one hand, and to other true ones gained from reference measurements wherein no target movements were provoked on the other hand. The obtained qualitative and quantitative measurement and image reconstruction results showed that the erroneous phantom signals could be accurately corrected, and the movement artefacts could be totally eliminated, verifying the applicability of the novel D&E technique in measurement optimization in biomedical MITS and supporting the proposed aspects.
在生物医学磁共振成像技术（MITS）中，测量过程中患者轻微的无意运动可能会在很大程度上污染目标图像。本研究通过检测测量过程中这些不可预测的运动，并消除测量后待重建图像中由此产生的运动伪影，来处理生物医学MITS中的测量优化问题。所提出的检测与消除（D&E）方法要求在多帧测量期间用特定的电磁干扰标记物标记被研究对象的表面。除了已发表的有源标记物概念外，还提出了一种新的、更为简单的无源标记物概念。除了由对象产生的生物信号外，标记物还会干扰主磁场，从而感应出它们自身信号。标记物的信号将用于检测任何不需要的对象运动以及由此损坏的信号帧。然后将损坏的信号帧排除在图像重建之外，以防止任何运动伪影与对象一起成像。为了评估所开发的D&E技术的可行性，进行了不同的实验，随后进行图像重建和定量分析。在此过程中，在生理电导率的盐水体模上，在β-色散频率范围内进行多频、多帧测量时引发目标运动。在测量过程中，用一个小的单匝线圈（有源标记物）或一个小的软铁氧体板（无源标记物）对体模进行标记。测量后，根据所建议的D&E策略校正错误的体模信号，并重建校正前后的体模图像。然后将校正后的信号和图像一方面与错误的信号和图像进行比较，另一方面与从参考测量中获得的其他真实信号和图像进行比较，在参考测量中未引发目标运动。所获得的定性和定量测量及图像重建结果表明，错误的体模信号能够被准确校正，运动伪影能够被完全消除，这验证了新型D&E技术在生物医学MITS测量优化中的适用性，并支持了所提出的观点。
