Background and Objectives: In institutional settings, manually segmenting connected speech is a time-consuming and labor-intensive process. This study aims to develop a deep-learning model for automating this process, evaluating its accuracy, and determining the minimum dataset size for effective performance.Materials and Method Voice data from 524 individuals with pathological conditions and 502 individuals with normal conditions, totaling 1026 samples, were used. Each voice sample had 17 chunks, including a “summer” sentence (15 chunks) and vowels /α/ and /i/. The deep-learning model employed in this study is based on the multi-layer perceptron-mixer architecture. This study evaluated performance using the Intersection over Union (IoU) metric, commonly employed in artificial intelligence-based image detection for chunk segmentation. Results: The accuracy of chunk identification at the frame level was 96.47%. Using IoU metrics, chunk segmentation accuracy was 98.15% at IoU ≥0.6, 96.03% at IoU ≥0.7, and 89.78% at IoU ≥0.8. Optimal dataset size exploration indicated that more than 700 connected speech datasets were needed for successful training, maintaining F1-scores up to 95% at IoU ≥0.7. Conclusion The artificial intelligence model is suitable for the development of an automated system that efficiently divides segments in the institutional collection of voice data. This suggests its potential utility in advancing voice research using connected speech.

Purpose: This study aimed to compare the outcomes of primary radiotherapy (RT) versus surgery in early-stage human papilloma virus–positive oropharyngeal squamous cell carcinoma (hpv+OPC), and investigate the preoperative clinical factors that can predict the requirement for postoperative adjuvant treatment. Materials and Methods: This multicenter study included 166 patients with American Joint Committee on Cancer 8th edition-Stages I-II hpv+OPC. Sixty (36.1%) and 106 (63.9%) patients underwent primary (concurrent chemo)radiotherapy [(CC)RT] and surgery, respectively. Seventy-eight patients (73.6%) in the surgery group received postoperative (CC)RT. Results: With a median follow-up of 45.6 months for survivors, the 2-year overall survival (OS), progression-free survival (PFS), and locoregional control (LC) for RT/surgery were 97.8%/96.4%, 91.1%/92.0%, and 92.9%/93.3%, respectively. In multivariate analyses, patients with synchronous radiologic extranodal extension and conglomeration (ENEcong) of metastatic lymph nodes (LNs) showed significantly poorer OS (p=0.047), PFS (p=0.001), and LC (p=0.003). In patients undergoing primary surgery, two or more clinically positive LN metastases (odds ratio [OR], 5.15; p=0.004) and LN metastases with ENEcong (OR, 3.75; p=0.009) were predictors of postoperative chemoradiotherapy. No patient in the primary RT group demonstrated late severe toxicity whereas three (2.8%), one (0.9%), and one (0.9%) patient in the surgery group showed grade 3 dysphagia, grade 3 xerostomia, and fatal oral cavity bleeding. Conclusion We found no differences in OS, PFS, and LC between upfront RT and surgery in stage I-II hpv+OPC which warrants comparison through a prospective trial in the treatment de-escalation era. However, most early-stage hpv+OPC patients undergoing surgery received adjuvant (CC)RT. Pretreatment LN findings were prognostic and predictive for adjuvant treatment.

Objectives: . Owing to the functional and structural complexity of the head and neck area, the reconstruction of defects in these areas is challenging. Free flap surgery has become standard for the reconstruction of the head and neck with improvements in microvascular surgery. The aim of this study was to use the cumulative sum (CUSUM) method to evaluate the learning curve for free-flap head and neck reconstruction performed by a single surgeon. Methods: . We retrospectively reviewed the medical records of 47 patients who underwent free-flap reconstruction from 2017 to 2021. The clinical demographics and surgical outcomes were analyzed. The total operation time was analyzed using the CUSUM method, which is an analytical approach for visualizing patterns in data by converting raw data into an accumulation of deviations from the average value. Results: . CUSUM analysis showed two phases of the learning curve: phase 1 (cases 1–22) and phase 2 (cases 23–47). The operative time in phase 1 (579.9±128.2 minutes) was significantly longer than that in phase 2 (418.6±80.9 minutes) (P<0.001). The re-exploration rate was higher in phase 1 (31.8%) than in phase 1 (4%) (P=0.018). The flap failure rate was higher in phase 1 (9.1%) than in phase 1 (4%), but this difference was not statistically significant (P=0.593). Conclusion . The learning curve of free-flap head and neck reconstruction seems to stabilize after approximately 20 cases.

Vocal fold leukoplakia poses a challenge to otolaryngologists due to its various spectrum of pathologic diagnosis. The degree of dysplasia is associated with malignancy risk and the new 2017 WHO classification system changed from the 3-tier system to a 2-tier system consisting of low and high grades. Infections including candidiasis, cryptococcosis, and tuberculosis should also be included in the differential diagnosis. Efforts have been made to evaluate risks using endoscopic technologies such as narrow band imaging, and surgery is essential for histopathological diagnosis. Regarding management, it is important to make an accurate diagnosis and find a balance between oncologic safety and functional outcome.

Background and Objectives: The quality of life after thyroidectomy, such as voice change, is considered to be as important as control of the disease. In this study, we aimed to evaluate changes in both subjective and objective voice parameters after thyroidectomy resulting in normal morbidity of the vocal cords.Materials and Method In this prospective cohort study, 204 patients who underwent thyroidectomy with or without central neck dissection at a single referral center from Feb 2015 to Aug 2016 were enrolled. All patients underwent prospective voice evaluations including both subjective and objective assessments preoperatively and then at 2 weeks, 3, 6, and 12 months postoperatively. Temporal changes of the voice parameters were analyzed. Results: Values of the subjective assessment tool worsened during the early postoperative follow-up period and did not recover to the preoperative values at 12 months postoperatively. The maximal phonation time gradually decreased, whereas most objective parameters, including maximal vocal pitch (MVP), reached preoperative values at 3–6 months postoperatively. The initial decrease in MVP was significantly greater in patients undergoing total thyroidectomy, and their MVP recovery time was faster than that of patients undergoing lobectomy (p=0.001). Patients whose external branch of the superior laryngeal nerve was confirmed intact by electroidentification showed no difference in recovery speed compared with patients without electroindentification (p=0.102), although the initial decrease in MVP was lower with electroidentification. Conclusion Subjective assessment in voice quality and maximal phonation time after thyroidectomy did not show recovery to preoperative values. Aggravation of MVP was associated with surgical extent and electroidentification.

Diagnosis using voice is non-invasive and can be implemented through various voice recording devices; therefore, it can be used as a screening or diagnostic assistant tool for laryngeal voice disease to help clinicians. The development of artificial intelligence algorithms, such as machine learning, led by the latest deep learning technology, began with a binary classification that distinguishes normal and pathological voices; consequently, it has contributed in improving the accuracy of multi-classification to classify various types of pathological voices. However, no conclusions that can be applied in the clinical field have yet been achieved. Most studies on pathological speech classification using speech have used the continuous short vowel /ah/, which is relatively easier than using continuous or running speech. However, continuous speech has the potential to derive more accurate results as additional information can be obtained from the change in the voice signal over time. In this review, explanations of terms related to artificial intelligence research, and the latest trends in machine learning and deep learning algorithms are reviewed; furthermore, the latest research results and limitations are introduced to provide future directions for researchers.

Objectives: . We aimed to assess the genetic differences between cases of early-stage tongue cancer that were positive or negative for lymph node metastasis. Methods: . In total, 35 cases of tongue cancer with RNA sequencing data were enrolled in this study. The gene expression profile of the following two groups was compared: N0 group (T stage 1 or 2 with N0 stage) and N+ group (T stage 1 or 2 with N+ stage). Using the R and limma packages in the Bioconductor program, we extracted the differentially expressed genes (DEGs). Gene ontology and pathway enrichment analysis were performed using the Database for Annotation, Visualization and Integration Discovery (DAVID) online tool. Immune cell infiltration was analyzed using the CIBERSORT online program. Immunochemical staining of the cancer tissue was evaluated and The Cancer Genome Atlas (TCGA) data were analyzed to validate the identified DEGs. Results: . No significant differences were found in the infiltration of 22 types of immune cells. Among a total of 51 identified DEGs, 14 genes were significantly upregulated, while 37 genes were significantly downregulated (P<0.01; fold change >2). Pathway analysis revealed significant associations with the arachidonic acid metabolism-related pathway, calcium signaling, and the muscle contraction pathway. The following DEGs were the most significantly different between the two groups: DEFB4A, SPRR2B, DEFB103B, SPRR2G, DEFB4B, and FAM25A. TCGA data showed that DEFB4A and DEFB103B were more highly expressed in the N0 group than in the N+ group, although the difference did not achieve statistical significance. Immunochemical staining of cancer tissue revealed significantly higher expression of defensin in the N0 group. Conclusions . Defensin (DEFB4A, DEFB103B, DEFB4B) may be a novel biomarker for early regional metastasis in T1/2 tongue cancer.

Objective: Fiberoptic endoscopic evaluation of swallowing (FEES) is a standard diagnostic tool for swallowing disorders. However, it has not been used frequently in Korea because of the long test time, low cost, and the absence of a standard evaluation system. The purpose of this study was to suggest a standard fill-out form for the FEES result. Methods: From February 2019 to June 2020, a total of 98 FEES tests were performed by an otolaryngologist (JYJ) at the Wonkwang University Hospital. After the exclusion of 68 cases, 30 cases were analyzed twice by 4 raters with over 5 years of experience as otolaryngologists working in various hospitals. The results were measured for the rater’s test-retest reliability and inter-rater consistency. Results: Cohen’s kappa values for measuring the intra-rater consistency of the four raters were 0.984, 0.887, 0.848, and 0.930, respectively, meaning very good alignment of 0.8 or more, respectively. The Fleiss Kappa value for measuring inter-rater consistency was 0.276, meaning ‘fair’ for values of 0.2 or more. To examine consistency, an intraclass correlation coefficient (ICC) analysis conducted by assuming the grading score to be a constant continuous variable gave an ICC value of 0.729 (P＜0.001), showing a very reliable tendency. Conclusion In this study, all the items of the fill-out form were rated using a three-step grading scale, so the degree of agreement was high when performed twice by the same rater, but the degree of agreement among raters was relatively low. Therefore, our fill-out form for FEES will be useful in evaluating the improvement of a patient over the course of clinical treatment.

The Korean Society of Laryngology, Phoniatrics and Logopedics appointed a task force to establish clinical practice guidelines for the management of unilateral vocal fold paralysis (UVFP). These guidelines cover a comprehensive range of management-related factors, including the diagnosis and treatment of UVFP, and provide in-depth information based on current, up-to-date knowledge. Detailed evidence profiles are provided for each recommendation. The CORE databases, including OVID Medline, Embase, the Cochrane Library, and KoreaMed, were searched to identify all relevant papers, using a predefined search strategy. When insufficient evidence existed, expert opinions and Delphi questionnaires were used to fill the evidence gap. The committee developed 16 evidence-based recommendations in six categories: initial evaluation (R1–4), spontaneous recovery (R5), medical treatment (R6), surgical treatment (R7–14), voice therapy (R15), and aspiration prevention (R16). The goal of these guidelines is to assist general otolaryngologists and speech-language pathologists who are primarily responsible for treating patients with UVFP. These guidelines are also intended to facilitate understanding of the condition among other health-care providers, including primary care physicians, nurses, and policy-makers.

BACKGROUND/AIMS: Since patients with human papillomavirus (HPV)-associated head and neck squamous cell carcinoma (HNSCC) have favorable outcomes after treatment, treatment de-escalation for these patients is being actively investigated. However, not all HPV-positive HNSCCs are curable, and some patients have a poor prognosis. The purpose of this study was to identify poor prognostic factors in patients with HPV-positive HNSCC. METHODS: Patients who received a diagnosis of HNSCC and tested positive for HPV from 2000 to 2015 at a single hospital site (n = 152) were included in this retrospective analysis. HPV typing was conducted using the HPV DNA chip assay or liquid bead microarray system. Expression of p16 in the tumors was assessed by immunohistochemistry. To determine candidate factors associated with overall survival (OS), univariate and multivariable Cox regression analyses were performed. RESULTS: A total of 152 patients with HPV-positive HNSCC were included in this study; 82.2% were male, 43.4% were current or former smokers, and 84.2% had oropharyngeal cancer. By univariate analysis, old age, performance status ≥ 1, non-oropharyngeal location, advanced T classification (T3–4), and HPV genotype 18 were significantly associated with poor OS. By multivariable analysis, performance status ≥ 1 and non-oropharyngeal location were independently associated with shorter OS (hazard ratio [HR], 4.36, p = 0.015; HR, 11.83, p = 0.002, respectively). Furthermore, HPV genotype 18 positivity was also an independent poor prognostic factor of OS (HR, 10.87, p < 0.001). CONCLUSIONS Non-oropharyngeal cancer, poor performance status, and HPV genotype 18 were independent poor prognostic factors in patients with HPV-positive HNSCC. Patients with these risk factors might not be candidates for de-escalation treatment.