Pediatric Postoperative Outcomes for Severe and Very Severe Obstructive Sleep Apnea Syndrome.

OBJECTIVE: To determine if increasing obstructive sleep apnea syndrome (OSAS) severity, as determined by preoperative polysomnography data, is an independent risk for respiratory complications and level of follow-up care after adenotonsillectomy or tonsillectomy. METHODS: A retrospective analysis of patients ≤21 years of age with severe OSAS (obstructive apnea-hypopnea index [OAHI] >10) undergoing adenotonsillectomy or tonsillectomy. Patients were categorized based on preoperative polysomnography data (PSG). Outcome measures including respiratory complications were collected via chart review. Logistic regression was used in the analysis of all parameters, and Wilcoxon Rank Sum tests were used for analysis of both OAHI and oxygen saturation nadir as continuous variables. All surgeries were performed at Johns Hopkins Hospital, a tertiary care center. RESULTS: We identified 358 patients with severe OSAS who had adenotonsillectomy or tonsillectomy. OAHI >40 and oxygen saturation nadir <80% were significantly associated with postoperative respiratory complications. Increasing OAHI and O2 saturation <80% was each associated with unplanned continuous positive airway pressure (CPAP) initiations postoperatively. There was no association between hypercarbia and presence of any complications. CONCLUSION: Patients with very severe OSAS (preoperative OAHI ≥40) as determined by preoperative PSG may be at higher risk of developing respiratory complications postoperatively. However, there does not appear to be a linear association with increasing severity of OAHI on regression analysis. Further research is needed to understand factors associated with complications in severe and very severe OAHI. LEVEL OF EVIDENCE: 4 Laryngoscope, 134:4148-4155, 2024.

Evaluating machine learning architectures for sound event detection for signals with variable signal-to-noise-ratios in the Beaufort Sea.

This paper explores the challenging polyphonic sound event detection problem using machine learning architectures applied to data recorded in the Beaufort Sea during the Canada Basin Acoustic Propagation Experiment. Four candidate architectures were investigated and evaluated on nine classes of signals broadcast from moored sources that were recorded on a vertical line array of hydrophones over the course of the yearlong experiment. These signals represent a high degree of variability with respect to time-frequency characteristics, changes in signal-to-noise ratio (SNR) associated with varying signal levels as well as fluctuating ambient sound levels, and variable distributions, which resulted in class imbalances. Within this context, binary relevance, which decomposes the multi-label learning task into a number of independent binary learning tasks, was examined as an alternative to the conventional multi-label classification (MLC) approach. Binary relevance has several advantages, including flexible, lightweight model configurations that support faster model inference. In the experiments presented, binary relevance outperformed conventional MLC approach on classes with the most imbalance and lowest SNR. A deeper investigation of model performance as a function of SNR showed that binary relevance significantly improved recall within the low SNR range for all classes studied.

Evaluation of Parental Perspectives and Concerns about Pediatric Cochlear Implantation: A Social Media Analysis.

OBJECTIVE: Parents often experience uncertainty during decision-making for their child's cochlear implantation (CI) surgery, and online forums provide insight on parental opinions that might not be expressed in clinic. This study aims to evaluate parental perspectives and concerns about pediatric CI using social media analysis. STUDY DESIGN: Qualitative study. SETTING: Three online forums involving parental posts about pediatric CI. INTERVENTION/METHODS: Forums were queried using keywords (e.g., "cochlear implant") to gather all U.S. parent-initiated posts about pediatric CI from 2006 to 2021. Thematic content analysis was performed to classify posts by overarching domain, themes, and subthemes. Posts were reviewed for thematic synthesis and double coded. Descriptive statistics were calculated for each theme by unique users. RESULTS: A total of 79 posts by 41 unique users were analyzed. Themes relating to decision for CI included facilitators, inhibitors, resources, and feelings. Parents posted about lack of benefit from hearing aids promoting decision for CI and high cost as an inhibitor. Some expressed concern about making a major decision for a minor. Parents often mentioned their child's healthcare providers and social media as resources. CONCLUSION: Through social media posts, parental priorities and concerns for decision-making of CI surgery were identified. Findings may guide discussions between physicians and parents and facilitate shared decision-making about CI.