Laryngoscopy, stroboscopy and other tools for the evaluation of voice disorders

Evaluating the Diagnostic Potential of Connected Speech for Benign Laryngeal Disease Using Deep Learning Analysis

OBJECTIVES

This study aimed to evaluate the performance of artificial intelligence (AI) models using connected speech and vowel sounds in detecting benign laryngeal diseases.

STUDY DESIGN

Retrospective.

METHODS

Voice samples from 772 patients, including 502 with normal voices and 270 with vocal cord polyps, cysts, or nodules, were analyzed. We employed deep learning architectures, including convolutional neural networks (CNNs) and time series models, to process the speech data. The primary endpoint was the area under the receiver's operating characteristic curve for binary classification.

RESULTS

CNN models analyzing speech segments significantly outperformed those using vowel sounds in distinguishing patients with and without benign laryngeal diseases. The best-performing CNN model achieved areas under the receiver operating characteristic curve of 0.895 and 0.845 for speech and vowel sounds, respectively. Correlations between AI-generated disease probabilities and perceptual assessments were more pronounced in the connected-speech analyses. However, the time series models performed worse than the CNNs.

CONCLUSION

Connected speech analysis is more effective than traditional vowel sound analysis for the diagnosis of laryngeal voice disorders. This study highlights the potential of AI technologies in enhancing the diagnostic capabilities of speech, advocating further exploration, and validation in this field.

Do-It-Yourself Low-Cost Stroboscopy

This case series tests the construction of an extremely low-cost stroboscope and assesses its strengths and weaknesses.

The Occurrence of Laryngeal Pathologies in a Treatment-Seeking Pediatric Population

OBJECTIVES

The purpose of this study was to 1) describe the age- and sex-specific occurrence of laryngeal pathologies in a treatment-seeking pediatric population in the voice unit of Ghent University Hospital, Belgium, and 2) describe this population in terms of vocal parameters, vocal complaints, influencing factors, and treatment history and recommendation.

STUDY DESIGN

Retrospective, observational study.

METHODS

All patient records were analyzed for children (0-18 years) who consulted the ear, nose, and throat department of Ghent University Hospital for the first time between July 2015 and June 2021 with complaints of dysphonia. In total, 103 children (66 males, 37 females) with a mean age of 10.01 years (SD: 3.4, range 3.93-17.96) were included in this study. Laryngeal pathology was diagnosed using a flexible videolaryngo(strobo)scopy. The influence of age and sex on laryngeal etiology (organic/functional voice disorder) was examined using a Welch-modified t test and a Fisher's exact test, respectively.

RESULTS

Organic lesions were observed in 77.7% of the participants, with vocal fold nodules (VFNs) being the most common diagnosis (66.0%). A functional voice disorder was diagnosed in 22.3% of the children. Children with a functional voice disorder are significantly older than children with an organic voice disorder. There was no statistically significant difference between males and females in laryngeal etiology. Mean dysphonia severity index was -2.7 (SD: 3.2, range -9.3 to +3.7), the mean acoustic voice quality index 4.70 (SD: 1.5, range 2.35-8.27), and the mean pediatric voice handicap index 29.8 (SD: 13.6, range 5-60). The occurrence of vocal misuse was mentioned in 80.6% of the patient records.

CONCLUSION

Organic voice disorders, especially VFNs, are predominant in treatment-seeking children with dysphonia. Functional voice disorders become more common with increasing age during childhood. A disordered vocal quality, reduced vocal capabilities and reduced voice-related quality of life were found.

An Experimental Analysis on Multicepstral Projection Representation Strategies for Dysphonia Detection

Biometrics-based authentication has become the most well-established form of user recognition in systems that demand a certain level of security. For example, the most commonplace social activities stand out, such as access to the work environment or to one's own bank account. Among all biometrics, voice receives special attention due to factors such as ease of collection, the low cost of reading devices, and the high quantity of literature and software packages available for use. However, these biometrics may have the ability to represent the individual impaired by the phenomenon known as dysphonia, which consists of a change in the sound signal due to some disease that acts on the vocal apparatus. As a consequence, for example, a user with the flu may not be properly authenticated by the recognition system. Therefore, it is important that automatic voice dysphonia detection techniques be developed. In this work, we propose a new framework based on the representation of the voice signal by the multiple projection of cepstral coefficients to promote the detection of dysphonic alterations in the voice through machine learning techniques. Most of the best-known cepstral coefficient extraction techniques in the literature are mapped and analyzed separately and together with measures related to the fundamental frequency of the voice signal, and its representation capacity is evaluated on three classifiers. Finally, the experiments on a subset of the Saarbruecken Voice Database prove the effectiveness of the proposed material in detecting the presence of dysphonia in the voice.

The Comparison of Narrow Band Imaging, White Light Laryngoscopy and Videolaryngostroboscopy in the Evaluation of Benign Vocal Fold Lesions

BACKGROUND

The aim of the study was comparing the diagnostic role of the white light (WL) endoscopy, video laryngostroboscopy (VLS), and narrow-band imaging (NBI) in the evaluation of the benign vocal fold lesions.

MATERIAL AND METHODS

From March 2018 to June 2019, a total of 118 cases were enrolled in this study. Ninety-eight patients were suspected with vocal fold nodules, cysts, polyps, and sulcus vocalis and 20 patients without a history of dysphonia. Each patient was examined by WL, VLS, and NBI endoscopy. Recorded images and videos were analyzed and scored by three otorhinolaryngologists who had at least 3 years of experience in phoniatrics field. The evaluation results were compared between the methods and the physicians.

RESULTS

A total of 118 cases were grouped by their confirmed diagnosis, vocal fold nodules (n = 28), vocal fold cyst (n = 24), vocal fold polyp (n = 9), and sulcus vocalis (n = 37). When the correct diagnosis rates of the physicians were compared, the statistical significance was found between the physicians in the WL and VLS method (P= 0.014, P= 0.027). No statistically significant difference was found among physicians in NBI method (P = 0.368). The difference between the diagnostic methods was found to be statistically significant in reaching the accurate diagnosis for benign vocal fold lesions (P< 0.001). While the difference between NBI-WL and VLS-WL was statistically significant (P< 0.001 and P< 0.001). The difference between NBI-VLS was not statistically significant while evaluating the vocal fold nodules, cysts, and sulcus vocalis separately (P= 0.102, P = 0.026, P = 0.157). Otherwise, it was statistically significant (P= 0.002) while evaluating total benign lesions in the study. The difference between NBI-VLS combination and VLS, NBI-VLS combination and VLS-WL combination were statistically significant (P< 0.001 and P= 0.001).

CONCLUSION

This study showed that NBI assessments have similar accuracy and sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) to VLS and have a higher value than WL endoscopy in reaching the diagnosis for benign vocal fold lesions. Therefore, NBI can be accepted as a promising approach to identify benign laryngeal lesions due to its optical properties.

Videokymographic index of glottic function: an analysis of diagnostic accuracy

ABSTRACT Purpose To develop the Videokymographic Index of Glottic Function (VIGF), a composite indicator from digital videokymography parameters, captured by high-speed videolaryngoscopy exams of women with and without laryngeal alterations of behavioral etiology. Methods The sample consisted of 92 women aged between 18 and 45 years. Fifty-five (55) women with behavioral dysphonia, presenting with laryngeal and voice alterations, and thirty-seven (37) women without any laryngeal and voice alterations. Voice evaluation was performed by consensus via an auditory-perceptual analysis of the sustained vowel /a/ at a habitual pitch and loudness. Voice classification was obtained by means of a general degree of dysphonia, where G0 indicated neutral voice quality and G1 to G3 indicated altered voice quality. Laryngeal images were captured via digital videokymography analysis of a sustained vowel /i/ at a habitual pitch and loudness. The VIGF was based on the midpoint of the glottal region for analysis. Logistic regression was performed using the MINITAB 19 program. Results Logistic regression was composed of two stages: Stage 1 consisted of the analysis of all variables, where the maximum opening and closed quotient variables showed statistical significance (p-value <0.05) and the model was well adjusted according to the Hosmer-Lemeshow test (p-value=0.794). Stage 2 consisted of the re-analysis of the selected variables, also showing a well-adjusted model (p-value=0.198). The VIGF was defined as follows: VIGF=e^(8.1318-0.2941AbMax-0.0703FechGlo)/1+e^(8.1318-0.2941AbMax-0.0703FechGlo). Conclusion The VIGF demonstrated a cut-off value equal to 0.71. The probability of success was 81.5%, sensitivity 76.4%, and specificity 89.2%.

Videokymographic index of glottic function: an analysis of diagnostic accuracy

PURPOSE

To develop the Videokymographic Index of Glottic Function (VIGF), a composite indicator from digital videokymography parameters, captured by high-speed videolaryngoscopy exams of women with and without laryngeal alterations of behavioral etiology.

METHODS

The sample consisted of 92 women aged between 18 and 45 years. Fifty-five (55) women with behavioral dysphonia, presenting with laryngeal and voice alterations, and thirty-seven (37) women without any laryngeal and voice alterations. Voice evaluation was performed by consensus via an auditory-perceptual analysis of the sustained vowel /a/ at a habitual pitch and loudness. Voice classification was obtained by means of a general degree of dysphonia, where G0 indicated neutral voice quality and G1 to G3 indicated altered voice quality. Laryngeal images were captured via digital videokymography analysis of a sustained vowel /i/ at a habitual pitch and loudness. The VIGF was based on the midpoint of the glottal region for analysis. Logistic regression was performed using the MINITAB 19 program.

RESULTS

Logistic regression was composed of two stages: Stage 1 consisted of the analysis of all variables, where the maximum opening and closed quotient variables showed statistical significance (p-value <0.05) and the model was well adjusted according to the Hosmer-Lemeshow test (p-value=0.794). Stage 2 consisted of the re-analysis of the selected variables, also showing a well-adjusted model (p-value=0.198). The VIGF was defined as follows: VIGF=e^(8.1318-0.2941AbMax-0.0703FechGlo)/1+e^(8.1318-0.2941AbMax-0.0703FechGlo).

CONCLUSION

The VIGF demonstrated a cut-off value equal to 0.71. The probability of success was 81.5%, sensitivity 76.4%, and specificity 89.2%.

Minor Structural Alterations of the Vocal Fold Cover: Vocal Quality before and after Treatment

Introduction  Minor structural alterations of the vocal fold cover are important causes of dysphonia. The variability in the type of alteration and the grade of vocal deviation affects the definition of the conduit and the results of treatment. Objective  To characterize the occurrence, the selected treatments adopted and vocal quality before and after treatment in patients with minor structural alterations of the vocal folds cover. Methods  This was a cross-sectional study based on the records of patients treated by an interdisciplinary team at the laryngology outpatient clinic of a public university hospital from 2010 to 2018. Data collection consisted of access to a database of information on otorhinolaryngological diagnostic hypotheses, intervention and perceptual-auditory vocal assessments before and after the treatment. Data from 102 subjects were analyzed. Association tests were applied between the perceptual-auditory vocal results and the different alterations found and between these and the adopted treatments. The results of the degrees of vocal deviation before and after treatment were also compared. Results  The degree of roughness was associated with the sulcus vocalis, and in this alteration the highest occurrence was mild degree of roughness. There was an improvement in the breathiness and general grade of vocal deviation after treatment. Conclusion  Cysts were the most frequent structural alteration in the population studied. There was an association between the degree of general deviation and that of roughness in sulcus vocalis cases. The breathiness and the general grade of vocal deviation improved after treatment regardless of the type of treatment and alteration.

Hoarseness

Hoarseness is a common problem, typically of transient nature. When hoarseness does not resolve, or when it is associated with concerning symptoms, it is important to consider a wide differential and refer to an otolaryngologist. This article discusses the physiology of the voice and possible causes of dysphonia, and explores when it warrants further work-up by ENT. A discussion of diagnostic techniques and the myriad of tools to treat hoarseness follows. Additionally, the role of reflux in dysphonia is examined with a critical eye to aid in accurate assessment of the patient's complaint.

High-resolution magnetic resonance and mass spectrometry imaging of the human larynx

High-resolution, noninvasive and nondestructive imaging of the subepithelial structures of the larynx would enhance microanatomic tissue assessment and clinical decision making; similarly, in situ molecular profiling of laryngeal tissue would enhance biomarker discovery and pathology readout. Towards these goals, we assessed the capabilities of high-resolution magnetic resonance imaging (MRI) and matrix-assisted laser desorption/ionisation-mass spectrometry (MALDI-MS) imaging of rarely reported paediatric and adult cadaveric larynges that contained pathologies. The donors were a 13-month-old male, a 10-year-old female with an infraglottic mucus retention cyst and a 74-year-old female with advanced polypoid degeneration and a mucus retention cyst. MR and molecular imaging data were corroborated using whole-organ histology. Our MR protocols imaged the larynges at 45-117 μm² in-plane resolution and capably resolved microanatomic structures that have not been previously reported radiographically-such as the vocal fold superficial lamina propria, vocal ligament and macula flavae; age-related tissue features-such as intramuscular fat deposition and cartilage ossification; and the lesions. Diffusion tensor imaging characterised differences in water diffusivity, primary tissue fibre orientation, and fractional anisotropy between the intrinsic laryngeal muscles, mucosae and lesions. MALDI-MS imaging revealed peptide signatures and putative protein assignments for the polypoid degeneration lesion and the N-glycan constituents of one mucus retention cyst. These imaging approaches have immediate application in experimental research and, with ongoing technology development, potential for future clinical application.

Components of Voice Evaluation

This article provides a concise review of contemporary options for evaluating voice disorders. Focus is given to patient history and patient-derived voice handicap and quality of life assessments, clinician-derived perceptual analysis of voice, and finally flexible and rigid, high-definition laryngoscopy with videostroboscopy to fully evaluate laryngeal function and biomechanics.

Role of Narrow Band Imaging in Relation to Stroboscopy and White Light Laryngoscopy in Diagnosis of Benign Nonvascular Glottic Lesions

Objective

To evaluate the role of narrow band imaging (NBI) in detecting benign nonvascular glottic lesions.

Materials and methods

Our study is a retrospective and prospective analysis of 247 patients with suspected benign glottic lesions who presented to our voice clinic over a 6-month duration. Patients were diagnosed on stroboscopy (by the first author) and divided into three groups consisting of leukoplakia, sulci, and cysts. A white light (WL) laryngoscopy and NBI examination was performed by an independent laryngologist (not an author) followed by a comparative analysis. The final gold standard for diagnosis was microlaryngoscopy with or without histopathology. Sensitivity of each modality for diagnosis of the three groups was calculated.

Results

The NBI is more sensitive for diagnosing leukoplakia. Stroboscopy is more sensitive for diagnosing sulcus.

Conclusion

Small vocal fold lesions may be missed on WL laryngoscopy. Stroboscopy helps in the diagnosis of structural glottic lesions affecting mucosal wave pattern. Many studies have shown that NBI light highlights the vasculature in superficial mucosal and subepithelial layers. This study is an analysis of the value of NBI in detecting relatively avascular glottic lesions, such as leukoplakia, sulci, and cysts.

How to cite this article

Nerurkar NK, Shedge AE, Gupta HK, Arya AA. Role of Narrow Band Imaging in Relation to Stroboscopy and White Light Laryngoscopy in Diagnosis of Benign Nonvascular Glottic Lesions. Int J Phonosurg Laryngol 2017;7(1):1-5.

Feasibility of Clinical Endoscopy and Stroboscopy in Children With Bilateral Vocal Fold Lesions

PURPOSE

The purpose of this study was to examine the utility of flexible and rigid endoscopy and stroboscopy for the identification of anatomical and physiological features in children with bilateral vocal fold lesions. The secondary purpose was to describe the age distribution of patients who could tolerate use of the different types of endoscopes.

METHOD

This cross-sectional clinic-based study included 38 children (ages 5 to 12 years) diagnosed with bilateral vocal fold lesions via videoendoscopy. Vocal fold vibratory characteristics (e.g., mucosal wave) were rated by 4 clinicians by consensus.

RESULTS

Bilateral vocal fold lesions could be well described anatomically after visualization with both flexible and rigid endoscopes and were most commonly described as symmetrical and broad based. However, the clinicians' confidence in the accuracy of stroboscopy for rating vocal fold vibratory characteristics was limited for both flexible and rigid stroboscopes.

CONCLUSIONS

Videoendoscopy was adequate for viewing and characterizing anatomical structures of bilateral vocal fold lesions in pediatric patients; however, vibratory characteristics were often not fully visualized with videostroboscopy. In view of the importance of visualizing vocal fold vibration in the differential diagnosis and treatment of vocal fold lesions, other imaging modalities, such as high-speed videoendoscopy, may provide more accurate descriptions of vocal fold vibratory characteristics in this population.

High- and ultrahigh-field magnetic resonance imaging of naïve, injured and scarred vocal fold mucosae in rats

Subepithelial changes to the vocal fold mucosa, such as fibrosis, are difficult to identify using visual assessment of the tissue surface. Moreover, without suspicion of neoplasm, mucosal biopsy is not a viable clinical option, as it carries its own risk of iatrogenic injury and scar formation. Given these challenges, we assessed the ability of high- (4.7 T) and ultrahigh-field (9.4 T) magnetic resonance imaging to resolve key vocal fold subepithelial tissue structures in the rat, an important and widely used preclinical model in vocal fold biology. We conducted serial in vivo and ex vivo imaging, evaluated an array of acquisition sequences and contrast agents, and successfully resolved key anatomic features of naïve, acutely injured, and chronically scarred vocal fold mucosae on the ex vivo scans. Naïve lamina propria was hyperintense on T1-weighted imaging with gadobenate dimeglumine contrast enhancement, whereas chronic scar was characterized by reduced lamina propria T1 signal intensity and mucosal volume. Acutely injured mucosa was hypointense on T2-weighted imaging; lesion volume steadily increased, peaked at 5 days post-injury, and then decreased – consistent with the physiology of acute, followed by subacute, hemorrhage and associated changes in the magnetic state of hemoglobin and its degradation products. Intravenous administration of superparamagnetic iron oxide conferred no T2 contrast enhancement during the acute injury period. These findings confirm that magnetic resonance imaging can resolve anatomic substructures within naïve vocal fold mucosa, qualitative and quantitative features of acute injury, and the presence of chronic scar.

Summary: Magnetic resonance imaging is shown to allow the nondestructive assessment of acute injury and scar formation in vocal fold mucosa, as demonstrated ex vivo using a preclinical rat model.

Management of Dysphonia After Radiation Therapy

Radiation-induced dysphonia can develop after radiation for primary laryngeal cancer or when the larynx is in the radiation field for nonlaryngeal malignancy. The effects are dose dependent and lead to variable degrees of dysphonia in both short- and long-term follow-up. Rehabilitation of the irradiated larynx can prove frustrating but can be facilitated through behavioral, pharmacologic, or surgical interventions.

Change in diagnosis and treatment following specialty voice evaluation: A national database analysis

OBJECTIVE

We evaluated the association between specialty voice evaluation and changes in laryngeal diagnosis and treatment in patients with laryngeal/voice disorders.

STUDY DESIGN

Retrospective analysis of a large, national administrative U.S. claims database.

METHODS

Patients included were identified with a laryngeal disorder based on International Classification of Diseases, Ninth Revision, Clinical Modification codes, from January 1, 2004, to December 31, 2008, and had been seen by an otolaryngologist as an outpatient and had a specialty voice evaluation designated by videolaryngostroboscopy (VLS) within 90 days of the last laryngoscopy. Patient diagnosis at the last laryngoscopy visit and the subsequent initial VLS visit were collected. Specific treatment modalities were tabulated for the 30-day period after the last laryngoscopy and for 30 days after the VLS.

RESULTS

A total of 168,444 unique patients saw an otolaryngologist for 273,616 outpatient visits. Of those, 6.1% had a VLS performed, of which 4,000 (23.8%) occurred within 90 days of the last laryngoscopy, with a median interval of 30 days (interquartile range 15-50 days). Half of the patient visits had a change in laryngeal diagnosis. Changes in use of antibiotics, proton pump inhibitors, voice therapy, and surgical intervention were seen after specialty voice evaluation.

CONCLUSIONS

Specialty voice evaluation was associated with changes in laryngeal diagnosis and treatment. Further study is needed to assess the impact on health care costs and patient outcomes.

Frequency and factors associated with use of videolaryngostroboscopy in voice disorder assessment

OBJECTIVES/HYPOTHESIS

Videolaryngostroboscopy (VLS) is considered an important diagnostic tool in the evaluation of patients with laryngeal/voice disorders. We evaluated the frequency of, diagnoses associated with, and factors related to VLS use in the evaluation of outpatients with laryngeal/voice disorders.

STUDY DESIGN

Retrospective analysis of a large, national administrative U.S. claims database.

METHODS

Patients with a laryngeal disorder based on International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes from January 1, 2004, to December 31, 2008, seen by an otolaryngologist were included. Patient age, gender, geographic region, laryngeal diagnosis, comorbid conditions, and whether laryngoscopy or VLS was performed during the outpatient otolaryngology visit were collected.

RESULTS

A total of 168,444 unique patients saw an otolaryngologist for 272,112 outpatient visits. Of those, 6.2% of outpatient otolaryngology visits had a VLS performed. Patient age was related to VLS use, with lower odds in the elderly (≥ 65 years of age) and those 0 to 17 years of age. Geographic variation was noted, with higher odds of VLS use in urban versus rural areas and greater odds in the Northeast versus the South. Laryngeal diagnosis was associated with VLS use with greatest odds for patients with multiple diagnoses, vocal fold paralysis, and paresis-followed by nonspecific dysphonia and benign vocal fold/laryngeal pathology and followed by acute and chronic laryngitis and laryngeal cancer. Patients with gastro-esophageal reflux (GER) had greater odds of VLS use than patients without GER.

CONCLUSIONS

VLS was used in 6.2% of outpatient otolaryngology outpatient visits; and its use was influenced by multiple factors.