STAMPS: development and verification of swallowing kinematic analysis software

Abstract

BackgroundSwallowing impairment is a common complication in various geriatric and neurodegenerative diseases. Swallowing kinematic analysis is essential to quantitatively evaluate the swallowing motion of the oropharyngeal structures. This study aims to develop a novel swallowing kinematic analysis software, called spatio-temporal analyzer for motion and physiologic study (STAMPS), and verify its validity and reliability.MethodsSTAMPS was developed in MATLAB, which is one of the most popular platforms for biomedical analysis. This software was constructed to acquire, process, and analyze the data of swallowing motion. The target of swallowing structures includes bony structures (hyoid bone, mandible, maxilla, and cervical vertebral bodies), cartilages (epiglottis and arytenoid), soft tissues (larynx and upper esophageal sphincter), and food bolus. Numerous functions are available for the spatiotemporal parameters of the swallowing structures. Testing for validity and reliability was performed in 10 dysphagia patients with diverse etiologies and using the instrumental swallowing model which was designed to mimic the motion of the hyoid bone and the epiglottis.ResultsThe intra- and inter-rater reliability tests showed excellent agreement for displacement and moderate to excellent agreement for velocity. The Pearson correlation coefficients between the measured and instrumental reference values were nearly 1.00 (P < 0.001) for displacement and velocity. The Bland–Altman plots showed good agreement between the measurements and the reference values.ConclusionsSTAMPS provides precise and reliable kinematic measurements and multiple practical functionalities for spatiotemporal analysis. The software is expected to be useful for researchers who are interested in the swallowing motion analysis.Electronic supplementary materialThe online version of this article (doi:10.1186/s12938-017-0412-1) contains supplementary material, which is available to authorized users.

Automatic Detection of Airway Invasion from Videofluoroscopy via Deep Learning Technology

“…To the authors' knowledge, the fully automatic detection of any kind of abnormality from VFSS images has never previously been attempted. Most previous studies have used obsolete semiautomated identification and/or tracking algorithms [12][13][14][15], in which salient landmarks must first be demarcated by humans. Two previous studies have attempted fully automated analysis of VFSS images, but did not address pathological findings.…”

“…Hossain et al attempted to track the motion of the hyoid bone semi-automatically by determining the region of interest using the Haar classifier, but manual identification of templates was required to increase accuracy [13]. Lee et al developed MATLAB-based software for kinematic analysis of swallowing with high correlation and reliability, but specific events and target structures should be marked manually [14]. Most existing models use automatic tracking of the salient anatomical structures following human manual demarcation in the first few frames.…”

Online Learning for the Hyoid Bone Tracking During Swallowing With Neck Movement Adjustment Using Semantic Segmentation

“…The length of the VFSS videos ranged from approximately 2 to 5 seconds. Swallowing motion analysis software, called the spatio-temporal analyzer for motion and physiologic study (STAMPS; https://github.com/cmookj/stamps), was used to obtain positional data of the hyoid bone [17]. The position of the hyoid bone was determined by two experienced examiners who were blinded to the results of the developed algorithm.…”

Swallowing kinematic analysis might be helpful in predicting aspiration and pyriform sinus stasis

“…A more objective analysis of swallowing biomechanical events is VFSS overlaid with swallowing kinematic evaluation. This is also known to have excellent interrater and intrarater reliability 6 . VFSS has several disadvantages, however, including radiation exposure, altered food consistency and sensory feedback due to the use of contrast medium, and limited patient accessibility 7 .…”

“…Dysphagia is a critical clinical issue that can occur in various disease groups, including stroke, head and neck cancer, Parkinson’s disease, and motor neuron disease 1 – 4 . Aspiration resulting from compromised swallowing can lead to aspiration pneumonia, which increases mortality and negatively affects the quality of life of patients 5 , 6 .…”