Automated assessment of hyoid movement during normal swallow using ultrasound.
Ontology highlight
ABSTRACT:
SUBMITTER: Ma JK
PROVIDER: S-EPMC9314830 | biostudies-literature | 2022 May
REPOSITORIES: biostudies-literature
Publications
Automated assessment of hyoid movement during normal swallow using ultrasound.
Ma Joan K-Y JK Wrench Alan A AA
International journal of language & communication disorders 20220314 3
<h4>Background</h4>The potential for using ultrasound by speech and language therapists (SLTs) as an adjunct clinical tool to assess swallowing function has received increased attention during the COVID-19 pandemic, with a recent review highlighting the need for further research on normative data, objective measurement, elicitation protocol and training. The dynamic movement of the hyoid, visible in ultrasound, is crucial in facilitating bolus transition and protection of the airway during a swa ...[more]
Similar Datasets
Project description:Hyoid bone movement is an important physiological event during swallowing that contributes to normal swallowing function. In order to determine the adequate hyoid bone movement, clinicians conduct an X-ray videofluoroscopic swallowing study, which even though it is the gold-standard technique, has limitations such as radiation exposure and cost. Here, we demonstrated the ability to track the hyoid bone movement using a non-invasive accelerometry sensor attached to the surface of the human neck. Specifically, deep neural networks were used to mathematically describe the relationship between hyoid bone movement and sensor signals. Training and validation of the system were conducted on a dataset of 400 swallows from 114 patients. Our experiments indicated the computer-aided hyoid bone movement prediction has a promising performance when compared with human experts' judgements, revealing that the universal pattern of the hyoid bone movement is acquirable by the highly nonlinear algorithm. Such a sensor-supported strategy offers an alternative and widely available method for online hyoid bone movement tracking without any radiation side-effects and provides a pronounced and flexible approach for identifying dysphagia and other swallowing disorders.
| S-EPMC6689594 | biostudies-literature
Project description:We are developing a system for long term Semi-Automated Rehabilitation At the Home (SARAH) that relies on low-cost and unobtrusive video-based sensing. We present a cyber-human methodology used by the SARAH system for automated assessment of upper extremity stroke rehabilitation at the home. We propose a hierarchical model for automatically segmenting stroke survivor's movements and generating training task performance assessment scores during rehabilitation. The hierarchical model fuses expert therapist knowledge-based approaches with data-driven techniques. The expert knowledge is more observable in the higher layers of the hierarchy (task and segment) and therefore more accessible to algorithms incorporating high level constraints relating to activity structure (i.e., type and order of segments per task). We utilize an HMM and a Decision Tree model to connect these high level priors to data driven analysis. The lower layers (RGB images and raw kinematics) need to be addressed primarily through data driven techniques. We use a transformer based architecture operating on low-level action features (tracking of individual body joints and objects) and a Multi-Stage Temporal Convolutional Network(MS-TCN) operating on raw RGB images. We develop a sequence combining these complimentary algorithms effectively, thus encoding the information from different layers of the movement hierarchy. Through this combination, we produce a robust segmentation and task assessment results on noisy, variable and limited data, which is characteristic of low cost video capture of rehabilitation at the home. Our proposed approach achieves 85% accuracy in per-frame labeling, 99% accuracy in segment classification and 93% accuracy in task completion assessment. Although the methodology proposed in this paper applies to upper extremity rehabilitation using the SARAH system, it can potentially be used, with minor alterations, to assist automation in many other movement rehabilitation contexts (i.e., lower extremity training for neurological accidents).
| S-EPMC8417323 | biostudies-literature
Project description:IntroductionThe significance of hyoid dynamics in OSA pathophysiology remains unclear. Drug-induced sleep endoscopy (DISE) is often used for evaluating patients intolerant to positive airway pressure (PAP) therapy. We performed DISE with concurrent hyoid-focused ultrasonography to quantify hyoid dynamics during obstructive and non-obstructive breathing.MethodsA cross-sectional analysis from a prospective cohort of patients undergoing DISE with PAP titration (DISE-PAP) and hyoid-focused ultrasound was conducted. Hyoid ultrasound was performed during obstructive breathing, and non-obstructive breathing after PAP administration. Motion was quantified by generating displacement curves based on echo-tracking hyoid movement. The image analysis protocol for quantifying hyoid displacement was performed independently by two researchers, and reliability of measures was assessed. Univariate and multivariate regressions were performed for various clinical data and hyoid displacement during obstructive breathing.ResultsTwenty patients met inclusion criteria. On average, the cohort was male (75%), elderly (65.9 ± 10 years), overweight (29.3 ± 3.99 kg/m2 ), and with moderate-to-severe OSA (29.3 ± 12.5 events/h). Mean hyoid displacement during obstructive breathing was 5.81 mm (±3.48). In all patients, hyoid displacement decreased after PAP administration (-3.94 mm [95% CI: -5.10, -2.78]; p < 0.0001). Inter-rater reliability for measures of hyoid displacement was excellent. After multivariate regression, hyoid displacement at baseline was associated with higher AHI (β [95% CI] = 0.18 [0.03, 0.33], p = 0.020).ConclusionDuring DISE, hyoid displacement is greater during obstructive breathing with significant variability amongst patients. Further, these ultrasonographic measurements had excellent intra- and inter-rater reliability. Additional, larger studies are needed to understand contributors to hyoid mobility.Level of evidence4 Laryngoscope, 133:3221-3227, 2023.
| S-EPMC10592545 | biostudies-literature
Project description:Breast imaging techniques are used to assess the tumor response to neoadjuvant treatment (NAT), which is increasingly one of the preferred therapeutic options and increases the rate of breast conservation for breast cancer. Herein, we report a case in which a woman was diagnosed with invasive ductal carcinoma in the left breast and received NAT before surgery. Automated breast ultrasound (AB US) was regularly performed before and during the NAT to evaluate the tumor response to NAT by measuring diameter changes and volume reductions of the tumor. Images showed that the tumor size was significantly reduced and disappeared after 7 cycles of NAT, except for macrocalcification. Postoperative histopathological examination confirmed that there were no residual tumor cells. We found that AB US overcame the limitations of handheld US, such as operator dependence, poor reproducibility and limited field of view, and can be an alternative modality to assess the tumor response of NAT in the absence of magnetic resonance imaging (MRI) instruments.
| S-EPMC9411026 | biostudies-literature
Project description:BackgroundEarly recognition of volume overload is essential for heart failure patients. Volume overload can often be easily treated if caught early but causes significant morbidity if unrecognized and allowed to progress. Intravascular volume status can be assessed by ultrasound-based estimation of right atrial pressure (RAP), but the availability of this diagnostic modality is limited by the need for experienced physicians to accurately interpret these scans.ObjectivesWe sought to evaluate whether machine learning can accurately estimate echocardiogram-measured RAP.MethodsWe developed fully automated deep learning models for identifying inferior vena cava scans with rapid inspiration in echocardiogram studies and estimating RAP from those scans. The RAP estimation model was trained and evaluated using 15,828 ultrasound videos of the inferior vena cava and coupled cardiologist-assessed RAP estimates as well as 319 RAP measurements from right heart catheterization.ResultsOur model agreed with cardiologist estimates 80.3% of the time (area under the receiver-operating characteristic of 0.844) in a test data set, at the upper end of interoperator agreement rates found in the literature of 70 to 75%. Our model's RAP estimates were statistically indistinguishable from cardiologists' ultrasound-based RAP estimates (P = 0.98) when compared against the gold standard of right heart catheterization RAP measurements in a subset of patients. Our model also generalized well to an external data set of echocardiograms from a different institution (area under the receiver-operating characteristic of 0.854 compared to cardiologist RAP estimates).ConclusionsMachine learning is capable of accurately and robustly interpreting RAP from echocardiogram videos. This algorithm could be used to perform automated assessments of intravascular volume status.
| S-EPMC11450965 | biostudies-literature
Project description:ObjectivesTo use superb microvascular imaging (SMI) to evaluate longitudinally spiral artery (SA) and uterine artery (UtA) vascular adaptation in normal human pregnancy, and to develop reference ranges for use at various gestational ages throughout pregnancy.MethodsThe data for this study were obtained from the National Institutes of Health (NIH)-funded Human Placenta Project. Women aged 18-35 years, with a body mass index < 30 kg/m2 , without comorbidities, with a singleton gestation conceived spontaneously, and gestational age at or less than 13 + 6 weeks were eligible for inclusion. The current analysis was restricted to uncomplicated pregnancies carried to term. Exclusion criteria included maternal or neonatal complications, fetal or umbilical cord anomalies, abnormal placental implantation or delivery < 37 weeks. Women who fulfilled the inclusion criteria formed the reference population of the Human Placenta Project study. Each participant underwent eight ultrasound examinations during pregnancy. The pulsatility index (PI) of both the left and right UtA were obtained twice for each artery and the presence or absence of a notch was noted. Using SMI technology, the total number of SA imaged was recorded in a sagittal placental section at the level of cord insertion. The PI and peak systolic velocity (PSV) were also measured in a total of six SA, including two in the central portion of the placenta, two peripherally towards the uterine fundal portion, and two peripherally towards the lower uterine segment.ResultsA total of 90 women fulfilled the study criteria. Maternal UtA-PI decreased throughout the first half of pregnancy from a mean ± SD of 1.39 ± 0.50 at 12-13 weeks' gestation to 0.88 ± 0.24 at 20-21 weeks' gestation. The mean number of SA visualized in a sagittal plane of the placenta increased from 8.83 ± 2.37 in the first trimester to 16.99 ± 3.31 in the late-third trimester. The mean SA-PI was 0.57 ± 0.12 in the first trimester and decreased progressively during the second trimester, reaching a nadir of 0.40 ± 0.10 at 24-25 weeks, and remaining constant until the end of pregnancy. SA-PSV was highest in early pregnancy with a mean of 57.16 ± 14.84 cm/s at 12-13 weeks' gestation, declined to a mean of 49.38 ± 17.88 cm/s at 20-21 weeks' gestation and continued to trend downward for the remainder of pregnancy, reaching a nadir of 34.50 ± 15.08 cm/s at 36-37 weeks' gestation. A statistically significant correlation was noted between SA-PI and UtA-PI (r = 0.5633; P < 0.001). Multilevel regression models with natural cubic splines were used to create reference ranges of SA-PSV and SA-PI for given gestational ages.ConclusionFrom early gestation, we have demonstrated the ability to image and quantify SA blood flow in normal pregnancy, and have developed reference ranges for use at various gestational ages throughout pregnancy. © 2023 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.
| S-EPMC10801897 | biostudies-literature
Project description:To understand the functional significance of skeletal muscle anatomy, a method of quantifying local shape changes in different tissue structures during dynamic tasks is required. Taking advantage of the good spatial and temporal resolution of B-mode ultrasound imaging, we describe a method of automatically segmenting images into fascicle and aponeurosis regions and tracking movement of features, independently, in localized portions of each tissue. Ultrasound images (25 Hz) of the medial gastrocnemius muscle were collected from eight participants during ankle joint rotation (2° and 20°), isometric contractions (1, 5, and 50 Nm), and deep knee bends. A Kanade-Lucas-Tomasi feature tracker was used to identify and track any distinctive and persistent features within the image sequences. A velocity field representation of local movement was then found and subdivided between fascicle and aponeurosis regions using segmentations from a multiresolution active shape model (ASM). Movement in each region was quantified by interpolating the effect of the fields on a set of probes. ASM segmentation results were compared with hand-labeled data, while aponeurosis and fascicle movement were compared with results from a previously documented cross-correlation approach. ASM provided good image segmentations (<1 mm average error), with fully automatic initialization possible in sequences from seven participants. Feature tracking provided similar length change results to the cross-correlation approach for small movements, while outperforming it in larger movements. The proposed method provides the potential to distinguish between active and passive changes in muscle shape and model strain distributions during different movements/conditions and quantify nonhomogeneous strain along aponeuroses.
| S-EPMC3349610 | biostudies-literature
Project description:There has been little attention given to the relationship between variations in normal craniofacial morphology and swallowing physiology. This preliminary investigation evaluated the relationship between the Frankfort-mandibular plane angle (FMA) and hyoid displacement during swallowing. Hyoid movement was evaluated during 12-ml and 24-ml swallows of liquid barium in 12 healthy subjects (age = 20-29 years, median = 23 years). Lateral projection videofluorography was utilized. Positions of the hyoid at maximum forward displacement, maximum upward displacement, starting position, and ending position were determined using image analysis software. The mean FMA was 28.92 degrees +/- 4.08 degrees (mean +/- SD, range = 20 degrees -34 degrees ). A Pearson correlation (<or=0.05) demonstrated that hyoid forward displacement was significantly inversely correlated with the FMA [R = -0.68, p = 0.015 (12 ml) and R = -0.72, p = 0.009 (24 ml)]; thus, the greater the FMA, the smaller the hyoid forward displacement. Upward displacement of the hyoid was not significantly correlated with FMA for 12-ml (R = -0.41, p = 0.55) or 24-ml swallows (R = 0.21, p = 0.512). In addition, there was no significant correlation between hyoid starting or ending positions. In conclusion, the results of this preliminary study suggest that normal variations in morphology, as measured by the FMA, may influence hyoid movement and therefore affect swallowing physiology.
| S-EPMC2689550 | biostudies-literature
Project description:Purpose:Because it shows the movement of different parts of the tongue in real time, ultrasound biofeedback therapy is a promising technology for speech research and remediation. One limitation is the difficulty of interpreting real-time ultrasound images of tongue motion. Our image processing system, TonguePART, tracks the tongue surface and allows for the acquisition of quantitative tongue part trajectories. Method:TonguePART automatically identifies the tongue contour based on ultrasound image brightness and tracks motion of the tongue root, dorsum, and blade in real time. We present tongue part trajectory data from 2 children with residual sound errors on /r/ and 2 children with typical speech, focusing on /r/ (International Phonetic Alphabet ?) in the phonetic context /?r/. We compared the tongue trajectories to magnetic resonance images of sustained vowel /?/ and /r/. Results:Measured trajectories show larger overall displacement and greater differentiation of tongue part movements for children with typical speech during the production of /?r/, compared to children with residual speech sound disorders. Conclusion:TonguePART is a fast, reliable method of tracking articulatory movement of tongue parts for syllables such as /?r/. It is extensible to other sounds and phonetic contexts. By tracking tongue parts, clinical researchers can investigate lingual coordination. TonguePART is suitable for real-time data collection and biofeedback. Ultrasound biofeedback therapy users may make more progress using simplified biofeedback of tongue movement.
| S-EPMC7286630 | biostudies-literature
Project description:Digital breast tomosynthesis (DBT) offers poor image quality along the depth direction. This paper presents a new method that improves the image quality of DBT considerably through the a priori information from automated ultrasound (AUS) images.DBT and AUS images of a complex breast-mimicking phantom are acquired by a DBT/AUS dual-modality system. The AUS images are taken in the same geometry as the DBT images and the gradient information of the in-slice AUS images is adopted into the new loss functional during the DBT reconstruction process. The additional data allow for new iterative equations through solving the optimization problem utilizing the gradient descent method. Both visual comparison and quantitative analysis are employed to evaluate the improvement on DBT images. Normalized line profiles of lesions are obtained to compare the edges of the DBT and AUS-corrected DBT images. Additionally, image quality metrics such as signal difference to noise ratio (SDNR) and artifact spread function (ASF) are calculated to quantify the effectiveness of the proposed method.In traditional DBT image reconstructions, serious artifacts can be found along the depth direction (Z direction), resulting in the blurring of lesion edges in the off-focus planes parallel to the detector. However, by applying the proposed method, the quality of the reconstructed DBT images is greatly improved. Visually, the AUS-corrected DBT images have much clearer borders in both in-focus and off-focus planes, fewer Z direction artifacts and reduced overlapping effect compared to the conventional DBT images. Quantitatively, the corrected DBT images have better ASF, indicating a great reduction in Z direction artifacts as well as better Z resolution. The sharper line profiles along the Y direction show enhancement on the edges. Besides, noise is also reduced, evidenced by the obviously improved SDNR values.The proposed method provides great improvement on the quality of DBT images. This improvement makes it easier to locate and to distinguish a lesion, which may help improve the accuracy of the diagnosis using DBT imaging.
| S-EPMC4032424 | biostudies-other