Dataset Information

Semi-automatic tracking, smoothing and segmentation of hyoid bone motion from videofluoroscopic swallowing study.

ABSTRACT: Motion analysis of the hyoid bone via videofluoroscopic study has been used in clinical research, but the classical manual tracking method is generally labor intensive and time consuming. Although some automatic tracking methods have been developed, masked points could not be tracked and smoothing and segmentation, which are necessary for functional motion analysis prior to registration, were not provided by the previous software. We developed software to track the hyoid bone motion semi-automatically. It works even in the situation where the hyoid bone is masked by the mandible and has been validated in dysphagia patients with stroke. In addition, we added the function of semi-automatic smoothing and segmentation. A total of 30 patients' data were used to develop the software, and data collected from 17 patients were used for validation, of which the trajectories of 8 patients were partly masked. Pearson correlation coefficients between the manual and automatic tracking are high and statistically significant (0.942 to 0.991, P-value<0.0001). Relative errors between automatic tracking and manual tracking in terms of the x-axis, y-axis and 2D range of hyoid bone excursion range from 3.3% to 9.2%. We also developed an automatic method to segment each hyoid bone trajectory into four phases (elevation phase, anterior movement phase, descending phase and returning phase). The semi-automatic hyoid bone tracking from VFSS data by our software is valid compared to the conventional manual tracking method. In addition, the ability of automatic indication to switch the automatic mode to manual mode in extreme cases and calibration without attaching the radiopaque object is convenient and useful for users. Semi-automatic smoothing and segmentation provide further information for functional motion analysis which is beneficial to further statistical analysis such as functional classification and prognostication for dysphagia. Therefore, this software could provide the researchers in the field of dysphagia with a convenient, useful, and all-in-one platform for analyzing the hyoid bone motion. Further development of our method to track the other swallowing related structures or objects such as epiglottis and bolus and to carry out the 2D curve registration may be needed for a more comprehensive functional data analysis for dysphagia with big data.

SUBMITTER: Kim WS

PROVIDER: S-EPMC5705154 | biostudies-literature | 2017

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Semi-automatic tracking, smoothing and segmentation of hyoid bone motion from videofluoroscopic swallowing study.

Kim Won-Seok WS Zeng Pengcheng P Shi Jian Qing JQ Lee Youngjo Y Paik Nam-Jong NJ

PloS one 20171128 11

Motion analysis of the hyoid bone via videofluoroscopic study has been used in clinical research, but the classical manual tracking method is generally labor intensive and time consuming. Although some automatic tracking methods have been developed, masked points could not be tracked and smoothing and segmentation, which are necessary for functional motion analysis prior to registration, were not provided by the previous software. We developed software to track the hyoid bone motion semi-automat ...[more]

PMID: 29182687

Similar Datasets

Project description:PURPOSE:To automate the estimation of swallowing motion from 2D MR cine images using deformable registration for future applications of personalized margin reduction in head and neck radiotherapy and outcome assessment of radiation-associated dysphagia. METHODS:Twenty-one patients with serial 2D FSPGR-MR cine scans of the head and neck conducted through the course of definitive radiotherapy for oropharyngeal cancer. Included patients had at least one cine scan before, during, or after radiotherapy, with a total of 52 cine scans. Contours of 7 swallowing related regions-of-interest (ROIs), including pharyngeal constrictor, epiglottis, base of tongue, geniohyoid, hyoid, soft palate, and larynx, were manually delineated from consecutive frames of the cine scan covering at least one swallowing cycle. We applied a modified thin-plate-spline robust-point-matching algorithm to register the point sets of each ROI automatically over frames. The deformation vector fields from the registration were then used to estimate the motion during swallowing for each ROI. Registration errors were estimated by comparing the deformed contours with the manual contours. RESULTS:On average 22 frames of each cine scan were contoured. The registration for one cine scan (7 ROIs over 22 frames) on average took roughly 22 minutes. A number of 8018 registrations were successfully batch processed without human interaction after the contours were drawn. The average registration error for all ROIs and all patients was 0.36 mm (range: 0.06 mm- 2.06 mm). Larynx had the average largest motion in superior direction of all structures under consideration (range: 0.0 mm- 58.7 mm). Geniohyoid had the smallest overall motion of all ROIs under consideration and the superior-inferior motion was larger than the anterior-posterior motion for all ROIs. CONCLUSION:We developed and validated a deformable registration framework to automate the estimation of swallowing motion from 2D MR cine scans.

Project description:Purpose: To evaluate the accuracy of deep-learning-based auto-segmentation of the superior constrictor, middle constrictor, inferior constrictor, and larynx in comparison with a traditional multi-atlas-based method. Methods and Materials: One hundred and five computed tomography image datasets from 83 head and neck cancer patients were retrospectively collected and the superior constrictor, middle constrictor, inferior constrictor, and larynx were analyzed for deep-learning versus multi-atlas-based segmentation. Eighty-three computed tomography images (40 diagnostic computed tomography and 43 planning computed tomography) were used for training the convolutional neural network, and for atlas-based model training. The remaining 22 computed tomography datasets were used for validation of the atlas-based auto-segmentation versus deep-learning-based auto-segmentation contours, both of which were compared with the corresponding manual contours. Quantitative measures included Dice similarity coefficient, recall, precision, Hausdorff distance, 95th percentile of Hausdorff distance, and mean surface distance. Dosimetric differences between the auto-generated contours and manual contours were evaluated. Subjective evaluation was obtained from 3 clinical observers to blindly score the autosegmented structures based on the percentage of slices that require manual modification. Results: The deep-learning-based auto-segmentation versus atlas-based auto-segmentation results were compared for the superior constrictor, middle constrictor, inferior constrictor, and larynx. The mean Dice similarity coefficient values for the 4 structures were 0.67, 0.60, 0.65, and 0.84 for deep-learning-based auto-segmentation, whereas atlas-based auto-segmentation has Dice similarity coefficient results at 0.45, 0.36, 0.50, and 0.70, respectively. The mean 95th percentile of Hausdorff distance (cm) for the 4 structures were 0.41, 0.57, 0.59, and 0.54 for deep-learning-based auto-segmentation, but 0.78, 0.95, 0.96, and 1.23 for atlas-based auto-segmentation results, respectively. Similar mean dose differences were obtained from the 2 sets of autosegmented contours compared to manual contours. The dose-volume discrepancies and the average modification rates were higher with the atlas-based auto-segmentation contours. Conclusion: Swallowing-related structures are more accurately generated with DL-based versus atlas-based segmentation when compared with manual contours.

Project description:PurposeIt is essential that clinicians have evidence-based benchmarks to support accurate diagnosis and clinical decision making. Recent studies report poor reliability for diagnostic judgments and identifying mechanisms of impairment from videofluoroscopy (VFSS). Establishing VFSS reference values for healthy swallowing would help resolve such discrepancies. Steele et al. (2019) released preliminary reference data for quantitative VFSS measures in healthy adults aged < 60 years. Here, we extend that work to provide reference percentiles for VFSS measures across a larger age span.MethodData for 16 VFSS parameters were collected from 78 healthy adults aged 21-82 years (39 male). Participants swallowed three comfortable sips each of thin, slightly, mildly, moderately, and extremely thick barium (20% w/v). VFSS recordings were analyzed in duplicate by trained raters, blind to participant and task, using the Analysis of Swallowing Physiology: Events, Kinematics and Timing (ASPEKT) Method. Reference percentiles (p2.5, 5, 25, 50, 75, 95, and 97.5) were determined as per Clinical and Laboratory Standards Institute EP28-A3c guidelines.ResultsWe present VFSS reference percentile tables, by consistency, for (a) timing parameters (swallow reaction time; the hyoid burst-to-upper esophageal sphincter (UES)-opening interval; UES opening duration; time-to-laryngeal vestibule closure (LVC); and LVC duration) and (b) anatomically scaled pixel-based measures of maximum UES diameter, pharyngeal area at maximum pharyngeal constriction and rest, residue (vallecular, pyriform, other pharyngeal locations, total), and hyoid kinematics (X, Y, XY coordinates of peak position; speed). Clinical decision limits are proposed to demarcate atypical values of potential clinical concern.ConclusionThese updated reference percentiles and proposed clinical decision limits are intended to support interpretation and reliability for VFSS assessment data.Supplemental materialhttps://doi.org/10.23641/asha.24043041.

Project description:Reduced range of hyoid and laryngeal movement is thought to contribute to aspiration risk and pharyngeal residues in dysphagia. Our aim was to determine the extent to which movements of the hyoid and larynx are correlated in the superior and anterior directions in swallowing, and whether movement range is predictive of penetration-aspiration or pharyngeal residue.Prospective, single-blind study of penetration-aspiration and pharyngeal residue with objective frame-by-frame measures of hyoid and laryngeal excursion from videofluoroscopy.Tertiary hospital and rehabilitation teaching hospital.Twenty-eight participants referred for videofluoroscopy: 13 women, aged 57-77; 15 men, aged 54-70. Individuals with known neurodegenerative diseases or prior surgery to the neck were excluded. Each swallowed three boluses of 40% w/v thin liquid barium suspension.Two speech-language pathologists independently rated penetration-aspiration, vallecular and pyriform sinus residue. Cervical spine length, hyoid and laryngeal displacement were traced frame-by-frame. Predictive power was calculated.Cervical spine length was significantly greater in men. Hyoid displacement ranged from 34-63% of the C2-4 distance. Arytenoid displacement ranged from 18-66%, with significantly smaller anterior displacement in men. Positive hyoid-laryngeal movement correlations in both axes were the most common pattern observed. Participants with reduced displacement ranges (? first quartile) and with abnormal correlation patterns were more likely to display penetration-aspiration. Those with reduced anterior hyoid displacement and abnormal correlation patterns had a greater risk of post-swallow pharyngeal residues.It is difficult for clinicians to make on-line appraisals of the extent to which hyoid and laryngeal movement may be contributing to functional swallowing consequences during videofluoroscopy. This study suggests that it is most important for clinicians to discern whether reduced anterior displacement of these structures is contributing to a patient's swallowing impairment. Measures of structural displacement in thin liquid swallowing should be corrected for variations in participant height. Reductions in anterior hyoid and laryngeal movement below the first-quartile boundaries are statistically associated with increased risk for penetration-aspiration and post-swallow residues.

Dataset Information

Semi-automatic tracking, smoothing and segmentation of hyoid bone motion from videofluoroscopic swallowing study.

Publications

Semi-automatic tracking, smoothing and segmentation of hyoid bone motion from videofluoroscopic swallowing study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets