Elevating Patient Care With Deep Learning: High-Resolution Cervical Auscultation Signals for Swallowing Kinematic Analysis in Nasogastric Tube Patients
Bibliographic record
Abstract
Patients with nasogastric (NG) tubes require careful monitoring due to the potential impact of the tube on their ability to swallow safely. This study aimed to investigate the utility of high-resolution cervical auscultation (HRCA) signals in assessing swallowing functionality of patients using feeding tubes. HRCA, capturing swallowing vibratory and acoustic signals, has been explored as a surrogate for videofluoroscopy image analysis in previous research. In this study, we analyzed HRCA signals recorded from patients with NG tubes to identify swallowing kinematic events within this group of subjects. Machine learning architectures from prior research endeavors, originally designed for participants without NG tubes, were fine-tuned to accomplish three tasks in the target population: estimating the duration of upper esophageal sphincter opening, estimating the duration of laryngeal vestibule closure, and tracking the hyoid bone. The convolutional recurrent neural network proposed for the first task predicted the onset of upper esophageal sphincter opening and closure for 67.61% and 82.95% of patients, respectively, with an error margin of fewer than three frames. The hybrid model employed for the second task successfully predicted the onset of laryngeal vestibule closure and reopening for 79.62% and 75.80% of patients, respectively, with the same error margin. The stacked recurrent neural network identified hyoid bone position in test frames, achieving a 41.27% overlap with ground-truth outputs. By applying established algorithms to an unseen population, we demonstrated the utility of HRCA signals for swallowing assessment in individuals with NG tubes and showcased the generalizability of algorithms developed in our previous studies. Clinical impact: This study highlights the promise of HRCA signals for assessing swallowing in patients with NG tubes, potentially improving diagnosis, management, and care integration in both clinical and home healthcare settings.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".