Department of Biomedical Engineering, The University of Iowa, Iowa City, IA 52242, USA.
Dysphagia. 2010 Dec;25(4):298-306. doi: 10.1007/s00455-009-9261-9. Epub 2009 Oct 24.
Videofluoroscopic examination of swallowing remains the standard for evaluation of patients with swallowing complaints. Although attempts have been made to objectify aspects of the study, findings from the study are largely subjective and reliant on clinician training and judgment leading to considerable inter-rater variability. We describe a computerized image analysis program designed to objectify one component of the swallow study, the movement of the hyoid bone. Hyoid motion has been shown to be different in dysphagic versus non-dysphagic patients. Reduced hyoid elevation is also considered a risk factor for aspiration; however, there has not been much work done on actually quantifying hyoid motion and associating it with other aspects of the swallow study. The clinician is prompted to define the hyoid bone in a calibration frame, and the system then tracks that region of interest throughout the rest of the study. This system shows strong correlations with manual analysis and can account for head position changes during the study. While the hyoid bone was reported on in this study, other regions of interest within the image field could also be tracked using this technique. A more quantitative analysis such as this has the opportunity to improve the inter-rater reliability of the test and therefore lead to more consistent findings from swallow studies.
吞咽的视频荧光检查仍然是评估吞咽问题患者的标准。尽管已经尝试使研究的某些方面客观化，但研究结果在很大程度上是主观的，并且依赖于临床医生的培训和判断，这导致了相当大的观察者间变异性。我们描述了一种计算机图像分析程序，旨在使吞咽研究的一个组成部分——舌骨的运动客观化。已经证明，吞咽困难患者与非吞咽困难患者的舌骨运动不同。舌骨运动减少也被认为是吸入的危险因素；然而，实际上很少有关于定量舌骨运动并将其与吞咽研究的其他方面相关联的工作。临床医生被提示在校准框架中定义舌骨，然后系统在整个研究过程中跟踪该感兴趣区域。该系统与手动分析具有很强的相关性，并可以解释研究过程中头部位置的变化。虽然在这项研究中报告了舌骨，但也可以使用这种技术来跟踪图像场中的其他感兴趣区域。这样更定量的分析有机会提高测试的观察者间可靠性，从而使吞咽研究的结果更加一致。
