[The effect of narrow band imaging in CO2 laser therapy in early-stage glottic cancer].

Objective:To explore efficacy of narrow band imaging（NBI） technique in CO2laser therapy in Early-Stage Glottic cancer. Methods:The clinical data of patients with Early-Stage Glottic cancer who underwent CO2laser vocal cord resection from June 2011 to August 2022 were retrospectively analyzed. Among these, 27 patients who underwent surgery assisted by NBI were assigned to the observation group, while 25 patients who underwent conventional CO2 laser microsurgery with a suspension laryngoscope were assigned to the control group. The differences between the two groups were analyzed in terms of intraoperative frozen pathology results, postoperative recurrence rates, 5-year cumulative disease-free survival rates, complications, and voice recovery. Results:All 52 patients were operated successfully. Temporary tracheostomy and serious complications did not occur during the operation. The postoperative patient's pronunciation was satisfactory. One patient experienced vocal cord adhesion, but there were no severe complications such as breathing difficulties or bleeding, with an overall complication rate of 1.92%. Postoperative follow-up was 1-5 years. The 5 years recurrence free survival in the general group was 77.90％, and the 5 years recurrence free survival in the NBI group was 100％, the difference was statistically significant（P<0.05）. NBI endoscopy is safer and more accurate than the general group in determining the safe margin of tumor mucosal resection（P<0.05）. Among the patients who accepted the voice analysis, the difference was no statistically significant（P>0.05）. Conclusion:Compared with conventional CO2laser surgery under microscope, NBI guided laser resection of Early-Stage Glottic cancer is more accurate. NBI guided laser resection could improve 5 years recurrence free survival rate. In a word, narrow-band imaging endoscopy can has very high value in clinical application.

Evaluation of neck ultrasound measurements as a difficult airway predictor: A prospective observational study.

This study aimed to investigate the relationship between cervical measurements and difficult airways using ultrasonographic measurements. American Society of Anesthesiologists grade I to III, male or female, 120 adult patients, undergoing elective surgery were enrolled in the study. The study involved measuring the distance of the trachea, cricoid cartilage, thyroid cartilage, vocal cord anterior commissure, and hyoid bone to the skin using a 10 to 13 MHz linear ultrasound probe in the transverse plane. Additionally, the length of the cricothyroid and thyrohyoid membranes, along with their distance from the skin, were measured using the probe in the sagittal plane. Subsequently, another experienced anesthesiologist conducted mask ventilation and intubation after the patient's induction of general anesthesia. Throughout this process, the patient was assessed for difficulties in mask ventilation, laryngoscopy, and intubation. 28 (23.3%) patients had a difficult airway. Analyzing the measurements associated with difficult airways, the most reliable predictor was the epiglottis midline-skin distance [AUC (area under the curve): 0.847, P < .001, cutoff: >19.9, sensitivity: 78.6%, specificity: 79.4%]. Additionally, other factors such as hyoid bone to skin distance, thyroid cartilage to skin distance, thyrohyoid membrane to skin distance, and vocal cord anterior commissure-skin distance were also identified as predictors for a difficult airway. The increase in the distance of the epiglottis midline, vocal cord anterior commissure, hyoid bone, thyrohyoid membrane, and thyroid cartilage to the skin at the level of the isthmus measured by ultrasonography is predictive of difficult airways. Based on our study outcomes, we assert that ultrasonographic evaluation can be used in the prediction of difficult airways.

A machine learning approach for vocal fold segmentation and disorder classification based on ensemble method.

In the healthcare domain, the essential task is to understand and classify diseases affecting the vocal folds (VFs). The accurate identification of VF disease is the key issue in this domain. Integrating VF segmentation and disease classification into a single system is challenging but important for precise diagnostics. Our study addresses this challenge by combining VF illness categorization and VF segmentation into a single integrated system. We utilized two effective ensemble machine learning methods: ensemble EfficientNetV2L-LGBM and ensemble UNet-BiGRU. We utilized the EfficientNetV2L-LGBM model for classification, achieving a training accuracy of 98.88%, validation accuracy of 97.73%, and test accuracy of 97.88%. These exceptional outcomes highlight the system's ability to classify different VF illnesses precisely. In addition, we utilized the UNet-BiGRU model for segmentation, which attained a training accuracy of 92.55%, a validation accuracy of 89.87%, and a significant test accuracy of 91.47%. In the segmentation task, we examined some methods to improve our ability to divide data into segments, resulting in a testing accuracy score of 91.99% and an Intersection over Union (IOU) of 87.46%. These measures demonstrate skill of the model in accurately defining and separating VF. Our system's classification and segmentation results confirm its capacity to effectively identify and segment VF disorders, representing a significant advancement in enhancing diagnostic accuracy and healthcare in this specialized field. This study emphasizes the potential of machine learning to transform the medical field's capacity to categorize VF and segment VF, providing clinicians with a vital instrument to mitigate the profound impact of the condition. Implementing this innovative approach is expected to enhance medical procedures and provide a sense of optimism to those globally affected by VF disease.

Surgeon-performed transcutaneous laryngeal ultrasound for vocal cord assessment after total thyroidectomy: a prospective study : Original article.

PURPOSE: Assessing vocal cord mobility is crucial for patients undergoing thyroid surgery. We aimed to evaluate the feasibility and efficacy of surgeon-performed transcutaneous laryngeal ultrasound (TLUS) compared to flexible nasolaryngoscopy. METHOD: From February 2022 to December 2022, we conducted a prospective observational study on patients scheduled for total thyroidectomy at our Institution. All patients underwent TLUS followed by flexible nasolaryngoscopy by a blinded otolaryngologist. Findings were classified as normal or vocal cord movement impairment and then compared. Patients evaluable on TLUS were included in Group A, while those not evaluable were included in Group B, and their features were compared. RESULTS: Group A included 180 patients, while Group B included 21 patients. Male sex (p < 0.001), age (p = 0.034), BMI (p < 0.001), thyroid volume (p = 0.038), and neck circumference (p < 0.001) were associated with Group B. TLUS showed a sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of 100%, 99.4%, 94.4%, 100%, and 99.4%, respectively. Cohen's K value was 0.984. CONCLUSION: TLUS is a valid, easy-to-perform, non-invasive, and painless alternative for evaluating vocal cords in selected patients. It can be used either as a first level exam and as screening tool for selecting cases for flexible nasolaryngoscopy. TLUS should be integrated into routine thyroid ultrasound examination.

Objective Comparison of White Light and Narrow-Band Imaging for Detecting Scars, Sulci and Nodules.

INTRODUCTION: Narrow-band imaging (NBI) can improve detection of lesions in the aerodigestive tract. However, its role in benign lesions of the larynx is unclear. This study aims to determine whether NBI improves the detection of scars, sulci, and nodules compared to panchromatic lighting using objective image analysis. METHODS: In total, 120 vocal folds (VFs) were analyzed with and without NBI (21 normal, 15 scars, 16 sulci, and 45 nodules). Each VF image had anterior, middle, and posterior thirds analyzed for brightness using an area morphometry software (Optimas 5.1a). The middle-third with the lesion was analyzed against surrounding VF segments for average and standard deviation (SD) in absolute grayscale. RESULTS: The use of panchromatic light resulted in greater illumination and grayscale values than NBI. All lesions tended to be in the mid-membranous fold. Under panchromatic light, change in brightness when comparing anterior versus middle (A-M) was +6.1% for normal, versus 6.5%, 8.1%, and 7.1% for sulci, nodules, and scars, respectively. Under NBI, they were 9.0% (normal), 12.3% (sulci), 13.7% (nodules), and 13.1% (scars). A greater SD of luminescence was observed at pathology sites (p < 0.05) when using NBI. The change in absolute grayscale at all lesion sites was greater when using NBI than when using panchromatic light (p < 0.05). CONCLUSION: NBI significantly enhanced the area of pathology in patients with nodules, sulci, and scars. Greater SD values in grayscale at pathologic sites were observed compared at normal sites. Thus, NBI may improve the detection of phonotraumatic lesions compared to panchromatic light. LEVEL OF EVIDENCE: N/A Laryngoscope, 134:4066-4070, 2024.

Investigation of the Position and Angle of the Vocal Fold Projected onto the Thyroid Cartilage.

OBJECTIVE: To perform laryngeal framework surgery for unilateral vocal fold paralysis and obtain favorable voice improvement, it is necessary to accurately determine the vocal fold and arytenoid cartilage positions. Thus, the position and angle of the paralyzed vocal folds and arytenoid cartilage projected onto the affected thyroid plate were measured using computed tomography (CT) before and after surgery. METHODS: Forty-six male patients with thyroid cartilage ossification observed on preoperative CT and vocal fold paralysis were included. Using Adobe Illustrator®, the thyroid plate on the affected side was reconstructed from the continuous images of the sagittal section of the CT examination during participant's quiet breathing (reconstructed affected thyroid plate [RATP]). RESULTS: The anterior commissure mean position was slightly cranial to the midpoint of the thyroid cartilage midline. The paralyzed vocal fold angle was not parallel to the baseline. The average unaffected vocal fold angle during vocalization projected onto the affected thyroid plate was 13.83°, which differed significantly from the average paralyzed vocal fold angle before surgery (19.05°). However, no significant difference was observed in comparison with the average angle of the paralyzed vocal fold after arytenoid adduction. The average distance from the inferior notch of the affected side thyroid cartilage to the affected arytenoid cartilage was 16.7 mm. CONCLUSION: By understanding the positional relationship between the thyroid cartilage plate and internal structure from preoperative CT images, more effective surgery can be performed according to individual differences. LEVEL OF EVIDENCE: IV Laryngoscope, 134:4088-4094, 2024.

Multi-instance learning based artificial intelligence model to assist vocal fold leukoplakia diagnosis: A multicentre diagnostic study.

OBJECTIVE: To develop a multi-instance learning (MIL) based artificial intelligence (AI)-assisted diagnosis models by using laryngoscopic images to differentiate benign and malignant vocal fold leukoplakia (VFL). METHODS: The AI system was developed, trained and validated on 5362 images of 551 patients from three hospitals. Automated regions of interest (ROI) segmentation algorithm was utilized to construct image-level features. MIL was used to fusion image level results to patient level features, then the extracted features were modeled by seven machine learning algorithms. Finally, we evaluated the image level and patient level results. Additionally, 50 videos of VFL were prospectively gathered to assess the system's real-time diagnostic capabilities. A human-machine comparison database was also constructed to compare the diagnostic performance of otolaryngologists with and without AI assistance. RESULTS: In internal and external validation sets, the maximum area under the curve (AUC) for image level segmentation models was 0.775 (95 % CI 0.740-0.811) and 0.720 (95 % CI 0.684-0.756), respectively. Utilizing a MIL-based fusion strategy, the AUC at the patient level increased to 0.869 (95 % CI 0.798-0.940) and 0.851 (95 % CI 0.756-0.945). For real-time video diagnosis, the maximum AUC at the patient level reached 0.850 (95 % CI, 0.743-0.957). With AI assistance, the AUC improved from 0.720 (95 % CI 0.682-0.755) to 0.808 (95 % CI 0.775-0.839) for senior otolaryngologists and from 0.647 (95 % CI 0.608-0.686) to 0.807 (95 % CI 0.773-0.837) for junior otolaryngologists. CONCLUSIONS: The MIL based AI-assisted diagnosis system can significantly improve the diagnostic performance of otolaryngologists for VFL and help to make proper clinical decisions.

Bilateral Vocal Fold Abduction Dysfunction: A Report of Two Neonatal Cases.

Stridor is caused by oscillation of the narrowed upper airway. The most common cause of neonatal stridor is laryngomalacia, followed by vocal fold abduction dysfunction. Herein, we present two neonatal cases of idiopathic dysfunction of vocal fold abduction. A neonate was admitted to the neonatal intensive care unit (NICU) on day 4 of life for inspiratory stridor, intermittent subcostal retraction, and cyanosis. A second neonate was admitted to the NICU on day 7 of life for inspiratory stridor and cyanosis when crying. Neither patient had dysmorphic features or unusual cardiac ultrasonography findings. The diagnosis was confirmed by laryngo-bronchoscopy. Conservative treatment with biphasic positive airway pressure was effective in both cases and symptoms resolved within a few months. Resolution of vocal fold abduction dysfunction was confirmed by repeat endoscopy. Clinical manifestations of vocal fold abduction dysfunction vary widely. Although most cases resolve spontaneously, prolonged tube feeding, or even tracheostomy, is needed in some severe cases. Diagnosis of vocal fold abduction dysfunction requires a laryngo-bronchoscopy study; thus, there may be a large number of undiagnosed patients. Vocal fold abduction dysfunction should be considered in the differential diagnosis for neonatal inspiratory stridor.

How reliable is assessment of true vocal cord-arytenoid unit mobility in patients affected by laryngeal cancer? a multi-institutional study on 366 patients from the ARYFIX collaborative group.

PURPOSE: In clinical practice the assessment of the "vocal cord-arytenoid unit" (VCAU) mobility is crucial in the staging, prognosis, and choice of treatment of laryngeal squamous cell carcinoma (LSCC). The aim of the present study was to measure repeatability and reliability of clinical assessment of VCAU mobility and radiologic analysis of posterior laryngeal extension. METHODS: In this multi-institutional retrospective study, patients with LSCC-induced impairment of VCAU mobility who received curative treatment were included; pre-treatment endoscopy and contrast-enhanced imaging were collected and evaluated by raters. According to their evaluations, concordance, number of assigned categories, and inter- and intra-rater agreement were calculated. RESULTS: Twenty-two otorhinolaryngologists evaluated 366 videolaryngoscopies (total evaluations: 2170) and 6 radiologists evaluated 237 imaging studies (total evaluations: 477). The concordance of clinical rating was excellent in only 22.7% of cases. Overall, inter- and intra-rater agreement was weak. Supraglottic cancers and transoral endoscopy were associated with the lowest inter-observer reliability values. Radiologic inter-rater agreement was low and did not vary with imaging technique. Intra-rater reliability of radiologic evaluation was optimal. CONCLUSIONS: The current methods to assess VCAU mobility and posterior extension of LSCC are flawed by weak inter-observer agreement and reliability. Radiologic evaluation was characterized by very high intra-rater agreement, but weak inter-observer reliability. The relevance of VCAU mobility assessment in laryngeal oncology should be re-weighted. Patients affected by LSCC requiring imaging should be referred to dedicated radiologists with experience in head and neck oncology.

Automatic Segmentation of Membranous Glottal Gap Area with U-Net-Based Architecture.

BACKGROUND: While videostroboscopy is recognized as the most popular approach for investigating vocal fold function, evaluating the numerical values, such as the membranous glottal gap area, remains too time consuming for clinical applications. METHODS: We used a total of 2507 videostroboscopy images from 137 patients and developed five U-Net-based deep-learning image segmentation models for automatic masking of the membranous glottal gap area. To further validate the models, we used another 410 images from 41 different patients. RESULTS: During development, all five models exhibited acceptable and similar metrics. While the VGG19 U-Net had a long inference time of 1654 ms, the other four models had more practical inference times, ranging from 16 to 138 ms. During further validation, Efficient U-Net demonstrated the highest intersection over union of 0.8455, the highest Dice coefficient of 0.9163, and the lowest Hausdorff distance of 1.5626. The normalized membranous glottal gap area index was also calculated and validated. Efficient U-Net and VGG19 U-Net exhibited the lowest mean squared errors (3.5476 and 3.3842) and the lowest mean absolute errors (1.8835 and 1.8396). CONCLUSIONS: Automatic segmentation of the membranous glottal gap area can be achieved through U-net-based architecture. Considering the segmentation quality and speed, Efficient U-Net is a reasonable choice for this task, while the other four models remain valuable competitors. The models' masked area enables possible calculation of the normalized membranous glottal gap area and analysis of the glottal area waveform, revealing promising clinical applications for this model. LEVEL OF EVIDENCE: NA Laryngoscope, 134:2835-2843, 2024.

A deep learning pipeline for automated classification of vocal fold polyps in flexible laryngoscopy.

PURPOSE: To develop and validate a deep learning model for distinguishing healthy vocal folds (HVF) and vocal fold polyps (VFP) on laryngoscopy videos, while demonstrating the ability of a previously developed informative frame classifier in facilitating deep learning development. METHODS: Following retrospective extraction of image frames from 52 HVF and 77 unilateral VFP videos, two researchers manually labeled each frame as informative or uninformative. A previously developed informative frame classifier was used to extract informative frames from the same video set. Both sets of videos were independently divided into training (60%), validation (20%), and test (20%) by patient. Machine-labeled frames were independently verified by two researchers to assess the precision of the informative frame classifier. Two models, pre-trained on ResNet18, were trained to classify frames as containing HVF or VFP. The accuracy of the polyp classifier trained on machine-labeled frames was compared to that of the classifier trained on human-labeled frames. The performance was measured by accuracy and area under the receiver operating characteristic curve (AUROC). RESULTS: When evaluated on a hold-out test set, the polyp classifier trained on machine-labeled frames achieved an accuracy of 85% and AUROC of 0.84, whereas the classifier trained on human-labeled frames achieved an accuracy of 69% and AUROC of 0.66. CONCLUSION: An accurate deep learning classifier for vocal fold polyp identification was developed and validated with the assistance of a peer-reviewed informative frame classifier for dataset assembly. The classifier trained on machine-labeled frames demonstrates improved performance compared to the classifier trained on human-labeled frames.

Diagnostic Accuracy of Laryngeal Ultrasound for Evaluating Vocal Fold Movement Impairment in Children.

PURPOSE: Vocal fold movement impairment (VFMI) secondary to recurrent laryngeal nerve (RLN) injury is a common source of morbidity after pediatric cervical, thoracic, and cardiac procedures. Flexible laryngoscopy (FL) is the gold standard to diagnose VFMI yet can be challenging to perform and/or risks possible clinical decompensation in some children and is an aerosolizing procedure. Laryngeal ultrasound (LUS) is a potential non-invasive alternative, but limited data exists in the pediatric surgical population regarding its efficacy. We aimed to investigate the diagnostic accuracy of LUS compared to FL in evaluating VFMI. METHODS: A prospective, single-center, single-blinded (rater) cohort study was undertaken on perioperative pediatric patients at risk for RLN injury. Patients underwent FL and LUS. Cohen's kappa was used to determine chance-corrected agreement. RESULTS: Between 2021 and 2023, 85 paired evaluations were performed with patients having a median (IQR) age of 10 (4, 42) months and weight of 7.5 (5.4, 13.4) kilograms. The prevalence of VFMI was 27.1%. Absolute agreement between evaluations was 98.8% (kappa 0.97, 95% CI: 0.91-1.00, P < 0.001). The sensitivity and specificity of LUS in detecting VFMI was 95.7% and 100%, yielding a positive predictive value (PPV) of 100% and negative predictive value (NPV) of 98.4% (95% CI: 90-100%). Diagnostic accuracy was 98.8% (95% CI: 93-100%). CONCLUSION: LUS is a highly accurate modality in evaluating VFMI in children. While FL remains the gold standard for diagnosis, LUS offers a low-risk screening modality for children at risk for VFMI such that only those with an abnormal LUS or presence of clinical symptoms discordant with LUS findings should undergo FL. TYPE OF STUDY: Prospective, single-center, single blinded (rater), cohort study. LEVEL OF EVIDENCE: Level II.

Vocal Cord Leukoplakia Classification Using Siamese Network Under Small Samples of White Light Endoscopy Images.

OBJECTIVE: Accurate vocal cord leukoplakia classification is instructive for clinical diagnosis and surgical treatment. This article introduces a reliable very deep Siamese network for accurate vocal cord leukoplakia classification. STUDY DESIGN: A study of a classification network based on a retrospective database. SETTING: Academic university and hospital. METHODS: The white light image datasets of vocal cord leukoplakia used in this article were classified into 6 classes: normal tissues, inflammatory keratosis, mild dysplasia, moderate dysplasia, severe dysplasia, and squamous cell carcinoma. The classification performance was assessed by comparing it with 6 classical deep learning models, including AlexNet, VGG Net, Google Inception, ResNet, DenseNet, and Vision Transformer. RESULTS: Experiments show the superior classification performance of our proposed network compared to state-of-the-art methods. The overall accuracy is 0.9756. The values of sensitivity and specificity are very high as well. The confusion matrix provides information for the 6-class classification task and demonstrates the superiority of our proposed network. CONCLUSION: Our very deep Siamese network can provide accurate classification results of vocal cord leukoplakia, which facilitates early detection, clinical diagnosis, and surgical treatment. The excellent performance obtained in white light images can reduce the cost for patients, especially those living in developing countries.

Ultrasound Screening After Cardiac Surgery Shows Vocal Fold Impairment and Predicts Aspiration.

INTRODUCTION: Vocal fold motion impairment (VFMI) is a known consequence after high-risk cardiac surgery. We implemented a universal laryngeal ultrasound (LUS) screening protocol for VFMI after the Norwood and aortic arch surgery. We hypothesized that LUS would accurately identify VFMI and predict postoperative aspiration. METHODS: We implemented a screening algorithm with LUS for patients undergoing high-risk cardiac surgery at a tertiary care pediatric hospital. Positively screened patients underwent flexible nasolaryngoscopy (FNL). Patients with an abnormal FNL underwent a video-fluoroscopic swallow study (VFSS). Patient demographics, length of stay, and swallowing outcomes were assessed. Two-tailed chi square and Wilcoxon rank sum tests were used to assess for differences. RESULTS: Sixty-seven patients underwent either Norwood or arch reconstruction over a 16-month period and underwent universal LUS. The average birth weight was 3.24 kg (SD 0.57). Of the 67 patients, VFMI was identified by LUS and 100% confirmed on FNL in 58.21% (n = 39/67) of patients. Aspiration and penetration on VFSS were higher in the group with VFMI as compared with those without VFMI (53.8% vs. 21.4%, p = 0.008). There was no difference in length of stay between patients who did not have a diagnosis of VFMI and those found to have VFMI (41.0 days vs 45.3 days p = 0.73). CONCLUSIONS: Universal LUS screening for patients following high-risk cardiac surgery may lead to earlier identification of postoperative VFMI and aspiration. Recognition of VFMI through this universal screening program could lead to earlier interventions and possibly improved swallowing outcomes. LEVEL OF EVIDENCE: 3 Laryngoscope, 134:1939-1944, 2024.

Effects of vocal fold lesions on particle deposition in a mouth-throat model: A numerical study.

OBJECTIVE: Dysphonia is very common worldwide and aerosol drug inhalation is an important treatment for patients with dysphonia. This study aimed to explore the effects of vocal fold (VF) lesions on the particle deposition pattern using computational modeling. METHODS: A realistic mouth-throat (MT) model of a healthy adult was constructed based on computed tomography images. Small and large vocal fold lesions were incorporated in the original model. A steady inhalation flowrate of 15 and 30 liter per minute (LPM) was used as the velocity inlet and monodisperse particles with diameters of 5 to 10 µm were simulated. RESULTS: Particles of larger size are more likely to be deposited in MT models, most of them distributed in oral cavity, oropharynx and supraglottis. The ideal sizes at 30 LPM ranged over 7-10 µm for healthy VFs and 6-8 µm for VF lesions. The best sizes at 15 LPM ranged over 6-8 µm for healthy VFs and 8-9 µm for VF lesions. CONCLUSION: Based on this study, VF lesions influence the deposition pattern in the glottis obviously. The ideal sizes differ at the flow rates of 15 and 30 LPM.

Dynamic changes of vocal tract dimensions with sound pressure level during messa di vocea).

The messa di voce (MdV), which consists of a continuous crescendo and subsequent decrescendo on one pitch is one of the more difficult exercises of the technical repertoire of Western classical singing. With rising lung pressure, regulatory adjustments both on the level of the glottis and the vocal tract are required to keep the pitch stable. The dynamic changes of vocal tract dimensions with the bidirectional variation of sound pressure level (SPL) during MdV were analyzed by two-dimensional real-time magnetic resonance imaging (25 frames/s) and synchronous audio recordings in 12 professional singer subjects. Close associations in the respective articulatory kinetics were found between SPL and lip opening, jaw opening, pharynx width, uvula elevation, and vertical larynx position. However, changes in vocal tract dimensions during plateaus of SPL suggest that perceived loudness could have been varied beyond the dimension of SPL. Further multimodal investigation, including the analysis of sound spectra, is needed for a better understanding of the role of vocal tract resonances in the control of vocal loudness in human phonation.

Optical Flow Glottovibrogram for the examination of vocal fold pathology.

Laryngeal high-speed video endoscopy is performed to examine the cycles of vocal fold vibrations in detail and to diagnose voice abnormalities. One of the recent image processing techniques for visualizing vocal fold vibration is optical flow-based playbacks, which include optical flow kymograms (OFKG) for local dynamics, optical flow glottovibrogram (OFGVG) and glottal optical flow waveforms (GOFW) for global dynamics. In recent times, various optical flow computing algorithms have been developed. In this paper, we used four well-known optical flow algorithms Horn Schunk, Lucas Kanade, Gunnar Farneback, and TVL1 to construct the optical flow playbacks. The proposed playback reliability is examined by comparing them to traditional representations such as Phonovibrogram (PVG). Since PVG and OFGVG are interconnected, a comparison study was carried out to better comprehend their interaction.Clinical Relevance- Both OFGVG and PVG add to the precision of interpreting pathological conditions by offering complementary information to the conventional spatiotemporal representations.

Classifying Vocal Folds Fixation from Endoscopic Videos with Machine Learning.

Vocal folds motility evaluation is paramount in both the assessment of functional deficits and in the accurate staging of neoplastic disease of the glottis. Diagnostic endoscopy, and in particular videoendoscopy, is nowadays the method through which the motility is estimated. The clinical diagnosis, however, relies on the examination of the videoendoscopic frames, which is a subjective and professional-dependent task. Hence, a more rigorous, objective, reliable, and repeatable method is needed. To support clinicians, this paper proposes a machine learning (ML) approach for vocal cords motility classification. From the endoscopic videos of 186 patients with both vocal cords preserved motility and fixation, a dataset of 558 images relative to the two classes was extracted. Successively, a number of features was retrieved from the images and used to train and test four well-grounded ML classifiers. From test results, the best performance was achieved using XGBoost, with precision = 0.82, recall = 0.82, F1 score = 0.82, and accuracy = 0.82. After comparing the most relevant ML models, we believe that this approach could provide precise and reliable support to clinical evaluation.Clinical Relevance- This research represents an important advancement in the state-of-the-art of computer-assisted otolaryngology, to develop an effective tool for motility assessment in the clinical practice.

Development and validation of an artificial intelligence algorithm for detecting vocal cords in video laryngoscopy.

Airway procedures in life-threatening situations are vital for saving lives. Video laryngoscopy (VL) is commonly performed during endotracheal intubation (ETI) in the emergency department. Artificial intelligence (AI) is widely used in the medical field, particularly to detect anatomical structures. This study aimed to develop an AI algorithm that detects vocal cords from VL images acquired during emergent situations. This retrospective study used VL images acquired in the emergency department to facilitate the ETI. The vocal cord image was labeled with a ground-truth bounding box. The dataset was divided into training and validation datasets. The algorithm was developed from a training dataset using the YOLOv4 model. The performance of the algorithm was evaluated using a test set. The test set was further divided into specific environments during the ETI for clinical subgroup analysis. In total, 20,161 images from 84 patients were used in this study. A total of 10,287, 5766, and 4108 images were used for the model training, validation, and test sets, respectively. The developed algorithm achieved F1 score 0.906, sensitivity 0.963, and specificity 0.842 in the validation set. The performance in the test set was F1 score 0.808, sensitivity 0.823, and specificity 0.804. We developed and validated an AI algorithm to detect vocal cords in VL. This algorithm demonstrated a high performance. The algorithm can be used to determine the vocal cord to ensure safe ETI.

Comparison between the acoustic fundamental frequency of the voice and the vibration frequency of the vocal folds analyzed by digital kymography. / Comparação entre a frequência fundamental acústica da voz e a frequência de vibração das pregas vocais analisada pela videoquimografia digital.

PURPOSE: To compare the frequency of vocal fold opening variation, analyzed by digital kymography, with the fundamental voice frequency obtained by acoustic analysis, in individuals without laryngeal alteration. METHODS: Observational analytical cross-sectional study. The participants were forty-eight women and 38 men from 18 to 55 years of age. The evaluation was made by voice acoustic analysis, by the habitual emission of the vowel /a/ for 3 seconds, and days of the week, and digital kymography (DKG), by the habitual emission of the vowels /i/ and /ɛ/. The measurements analyzed were acoustic fundamental frequency (f0), extracted by the Computerized Speech Lab (CSL) program, and dominant frequency of the variation of right (R-freq) and left (L-freq) vocal fold opening, obtained through the KIPS image processing program. The mounting of the kymograms consisted in the manual demarcation of the region by vertical lines delimiting width and horizontal lines separating the posterior, middle and anterior thirds of the Rima glottidis. In the statistical analysis, the Anderson-Darling test was used to verify the normality of the sample. The ANOVA and Tukey tests were performed for the comparison of measurements between the groups. For the comparison of age between the groups, the Mann-Whitney test was used. RESULTS: There are no differences between the values of the frequency measurement analyzed by digital kymography, with the acoustic fundamental frequency, in individuals without laryngeal alteration. CONCLUSION: The values of the dominant frequency of the vocal folds opening variation, as assessed by digital kymography, and the acoustic fundamental frequency of the voice are similar, allowing comparison between these measurements in the multidimensional evaluation of the voice, in individuals without laryngeal alteration.

OBJETIVO: Comparar a frequência da variação da abertura das pregas vocais, analisada pela videoquimografia digital, com a frequência fundamental da voz, obtida através da análise acústica, em indivíduos sem alteração laríngea. MÉTODO: Trata-se de um estudo observacional analítico transversal. Participaram 48 mulheres e 38 homens, de 18 a 55 anos. A avaliação foi composta por análise acústica da voz, obtida pela emissão habitual da vogal /a/ durante 3 segundos, e os dias da semana, e pela videoquimografia digital (DKG), obtida pela emissão habitual das vogais /i/ e /ɛ/. As medidas analisadas foram a frequência fundamental acústica (f0), extraída pelo programa Computerized Speech Lab (CSL), e a frequência dominante da variação de abertura da prega vocal direita (D-freq) e esquerda (E-freq), obtidas através do programa de processamento de imagens KIPS. A montagem dos quimogramas constou na demarcação manual da região, compostas por linhas verticais que delimitaram largura da prega vocal e linhas horizontais que marcaram os terços posterior, médio e anterior da rima glótica. Na análise estatística, o teste Anderson-Darling foi utilizado para verificar a normalidade da amostra. Os testes ANOVA e Tukey foram realizados para a comparação das medidas entre os grupos. Para a comparação da idade entre os grupos, foi utilizado o teste Mann-Whitney. RESULTADOS: Não existem diferenças entre os valores da medida de frequência analisada pela videoquimografia digital, com a frequência fundamental acústica, em indivíduos sem alteração laríngea. CONCLUSÃO: Os valores da frequência dominante da variação de abertura das pregas vocais, avaliada pela videoquimografia digital, e a frequência fundamental acústica da voz são similares, permitindo uma comparação entre estas medidas na avaliação multidimensional da voz, em indivíduos sem alteração laríngea.