Shitong Mao

Purpose: Posterior pharyngeal wall (PPW) thickening is a common structural manifestation of treatment-related internal lymphedema in patients with head and neck cancer and can be evaluated using lateral videofluoroscopic images acquired during modified barium swallow (MBS) studies. However, no standardized method currently exists for quantitative measurement. Methods: This two-phase methodological study first conducted a scoping review to characterize existing approaches to PPW thickness measurement and identify the principal factors contributing to variation in PPW thickness measurement. Eight studies were identified with variation observed across these measurement factors. An empirical reliability study was then performed using 107 MBS studies from 93 Head and Neck Cancer patients. Two independent raters annotated 10 anatomical landmarks on pre-swallow and post-swallow resting frames, and a computer-assisted algorithm quantitatively generated seven candidate PPW thickness measurements based on different vertebral reference points and measurement directions. Reliability was assessed intraclass correlation coefficients (ICCs). Results: Inter-rater reliability was highest for the measurement originating from the midpoint of C3 and extending perpendicular to the anterior C3 surface (ICC, 0.78; 95% CI, 0.72-0.83). Among the normalization references, C2-C4 vertebral length demonstrated greater reliability than C3 vertebral length, with an ICC of 0.88 (95% CI, 0.85-0.91). No significant differences were observed between pre-swallow and post-swallow PPW thickness measurements across the evaluated approaches. Conclusions: These findings provide a methodological foundation for more standardized and reliable quantification of PPW thickness, with potential to expand the role of MBS from the functional evaluation of swallowing to include quantitative characterization of pharyngeal structural abnormalities.

Purpose: Modified Barium Swallow (MBS) studies utilize videofluoroscopy, a dynamic X-ray technique for evaluating swallowing anatomy and physiology. Each MBS exam typically includes multiple bolus trials, often involving different bolus consistencies. Accurate classification of bolus types is essential, as swallowing dynamics, aspiration risks, and residue levels vary with bolus consistency. In this preliminary study, we propose a deep learning–based approach for automated bolus type classification in MBS, aiming to provide a standardized and efficient framework for automated processing of swallowing assessments. Methods: A total of 206 patients (Mean ± SD age: 60.24 ± 9.02 years; 89.32% men) underwent MBS examinations, comprising 277 individual MBS studies. The dataset included 2,752 bolus-level video segments, categorized by bolus type as follows: 1,711 liquid (IDDSI 0-3, 62.17%), 521 pudding (IDDSI 4, 18.93%), and 520 solid boluses (IDDSI 7, cookie or cracker, 18.89%). To standardize variable video lengths for the data pipeline, each MBS video was temporally segmented into a fixed-length frame sequence, with shorter videos padded using static frames and longer videos randomly cropped to the target length. We employed an Inflated 3D convolutional neural network to develop the deep learning model. Results: Each video segment contained an average of 273.03 ± 195.81 frames. On the independent test set, the deep learning model achieved an overall accuracy of 96.13%, and the macro F1-score was 95.05% in classifying food bolus types within MBS videos. Conclusions: The developed AI-based system demonstrated effective automated classification of food bolus types in MBS videos, representing an important step toward fully automated MBS analysis for swallowing efficiency assessment. The AI model reduces the reliance on manual labels, thereby promising to streamline clinical and research workflows.

Background: This proof‐of‐concept case–control study examined the feasibility of tongue high‐density surface electromyography (HDS–EMG) to detect hypoglossal neuropathy. Methods: We analyzed tongue HDS–EMG for N = 2 participants. Subjects were graded through clinical tongue functional measures and gold‐standard needle EMG. We collected HDS–EMG in three tongue tasks: relaxation, protrusion, and isometric maximum effort. The HDS–EMG was decomposed into motor unit spike train estimates. Results: Muscle relaxation HDS–EMG contained spikes consistent with fasciculation potential morphology. During tongue protrusion, a case of confirmed CN XII neuropathy exhibited an elevated motor unit (MU) discharge rate and fewer MUs (45.71 peaks/s, 3 estimated MUs) compared to the control (11.63 peaks/s, 11 estimated MUs). During isometric contraction, the case exhibited an average discharge rate of 3.96 peaks/s on 3 MU estimates, whereas the control had a discharge rate of 2.58 peaks/s on 5 MU estimates. Conclusions: Non‐invasive tongue HDS–EMG appears to show potential sensitivity for detecting reduced MU recruitment in hypoglossal neuropathy.

Understanding the effects of reduced frame rates on the reliability of Modified Barium Swallow (MBS) ratings for swallowing safety and efficiency is essential for clinical practice. While previous research has examined frame rate (simulated pulse rates) implications concerning penetration, aspiration, and residue ratings, the impact on summary grading systems like the Dynamic Imaging Grade of Swallowing Toxicity (DIGEST) remains unclear. This study analyzed a total of 315 full MBS videos collected from 278 patients, and 76.25% of patients were diagnosed with head and neck cancer (mean age 62.51 years, SD 8.53; 12.23% female). We performed pairwise comparisons of DIGEST grades and DIGEST grade components (Safety and Efficiency) derived independently across studies with 30, 15, and 7.5 frames per second (FPS). Weighted Cohen’s kappa values consistently exceeded 0.84 across all assessments, indicating “almost perfect” agreement among the different simulated pulse rates. Exact agreement for all comparisons surpassed 85%. These findings suggest that the DIGEST grading system is robust to variations in frame rate, allowing for reliable assessments even under reduced pulse rate conditions.

Background: Quantifying head and neck lymphedema and fibrosis (HN‐LEF) is crucial in the investigation and management of treatment sequelae in head and neck cancer (HNC).Methods: The T1‐ and T2‐weighted MRI signal intensity (SI) was examined in relation to HN‐LEF categories per physical/tactile examination (No‐LEF, A‐B = edema, C = edema + fibrosis, D = fibrosis), and MRI structural volumes were examined in relation to a novel 10‐point HN‐LEF score in the intraoral and submental regions.Results: We identified differences in ranks among HN‐LEF categories in relation to the MRI SI (A‐B and C are higher than D and No‐LEF for T2 SI, and A‐B is the highest for T1). Furthermore, six pairs of FOM volumes on MRI demonstrated a strong negative correlation (p < 0.05) with the HN‐LEF score at adjacent palpable sites.Conclusion: Both MRI SI and structural volumes can potentially be imaging biomarkers of edematous soft tissue states in HNC patients.

Background Radiation-associated lymphedema and fibrosis (LEF) is a significant toxicity following radiation therapy (RT) for head and neck cancer (HNC) patients. Recently, the CT Lymphedema and Fibrosis Assessment Tool (CT-LEFAT) was developed to standardize LEF diagnosis through fat stranding visualized on CT. This study aims to evaluate the inter-observer reliability and diagnostic accuracy of the CT-LEFAT criteria. Materials and Methods This study retrospectively evaluated 26 HNC patients treated with RT that received a minimum of two contrast-enhanced CT scans. Qualitative review was conducted by five physician raters to assess the fat stranding observed on CT according to the CT-LEFAT criteria. Fleiss’ kappa analysis was used to assess the inter- and intra-rater reliability, and Receiver Operating Characteristic (ROC) Area Under the Curve (AUC) analysis was used to evaluate diagnostic accuracy. Results The inter-rater reliability across the six CT-LEFAT regions generally indicated a slight to fair agreement across all raters (0.04 ≤ kappa ≤ 0.36). Intra-observer agreement was generally fair to moderate (overall kappa=0.44). The ROC AUC analysis varied based on aggregation method used (0.60 ≤ average AUC ≤ 0.70). Conclusion This specific use-case evaluating CT-LEFAT criteria displays limited performance. This suggests that additional materials, such as further training, refinement of imaging methods, or other processes may be required before achieving clinically-ready diagnostic performance of LEF diagnosis.

Background: This pilot study analyzed correlations between tongue electrical impedance myography (EIM), standard tongue electromyography (EMG), and tongue functional measures in N = 4 long‐term oropharyngeal cancer (OPC) survivors.Methods: Patients were screened for a supportive care trial (NCT04151082). Hypoglossal nerve function was evaluated with genioglossus needle EMG, functional measures with the Iowa oral performance instrument (IOPI), and multi‐frequency tissue composition with tongue EIM.Results: Tongue EIM conductivity was higher for patients with EMG‐confirmed cranial nerve XII neuropathy than those without (p = 0.005) and in patients with mild versus normal EMG reinnervation ratings (16 kHz EIM: p = 0.051). Tongue EIM correlated with IOPI strength measurements (e.g., anterior maximum isometric lingual strength: r² = 0.62, p = 0.020).Conclusions: Tongue EIM measures related to tongue strength and the presence of XII neuropathy. Noninvasive tongue EIM may be a convenient adjunctive biomarker to assess tongue health in OPC survivors.

Background: Upper esophageal sphincter opening (UESO), and laryngeal vestibule closure (LVC) are two essential kinematic events whose timings are crucial for adequate bolus clearance and airway protection during swallowing. Their temporal characteristics can be quantified through time‐consuming analysis of videofluoroscopic swallow studies (VFSS).Objectives: We sought to establish a model to predict the odds of penetration or aspiration during swallowing based on 15 temporal factors of UES and laryngeal vestibule kinematics.Methods: Manual temporal measurements and ratings of penetration and aspiration were conducted on a videofluoroscopic dataset of 408 swallows from 99 patients. A generalized estimating equation model was deployed to analyze association between individual factors and the risk of penetration or aspiration.Results: The results indicated that the latencies of laryngeal vestibular events and the time lapse between UESO onset and LVC were highly related to penetration or aspiration. The predictive model incorporating patient demographics and bolus presentation showed that delayed LVC by 0.1 s or delayed LVO by 1% of the swallow duration (average 0.018 s) was associated with a 17.19% and 2.68% increase in odds of airway invasion, respectively.Conclusion: This predictive model provides insight into kinematic factors that underscore the interaction between the intricate timing of laryngeal kinematics and airway protection. Recent investigation in automatic noninvasive or videofluoroscopic detection of laryngeal kinematics would provide clinicians access to objective measurements not commonly quantified in VFSS. Consequently, the temporal and sequential understanding of these kinematics may interpret such measurements to an estimation of the risk of aspiration or penetration which would give rise to rapid computer‐assisted dysphagia diagnosis.Level of Evidence: 2 Laryngoscope, 133:521–527, 2023

Artificial intelligence and machine learning techniques have progressed dramatically and become powerful tools required to solve complicated tasks, such as computer vision, speech recognition, and natural language processing. Since these techniques have provided promising and evident results in these fields, they emerged as valuable methods for applications in human physiology and healthcare. General physiological recordings are time-related expressions of bodily processes associated with health or morbidity. Sequence classification, anomaly detection, decision making, and future status prediction drive the learning algorithms to focus on the temporal pattern and model the nonstationary dynamics of the human body. These practical requirements give birth to the use of recurrent neural networks (RNNs), which offer a tractable solution in dealing with physiological time series and provide a way to understand complex time variations and dependencies. The primary objective of this article is to provide an overview of current applications of RNNs in the area of human physiology for automated prediction and diagnosis within different fields. Finally, we highlight some pathways of future RNN developments for human physiology.