Validation of the laryngopharyngeal reflux color and texture recognition compared to pH-probe monitoring

The Use of Deep Learning Software in the Detection of Voice Disorders: A Systematic Review

OBJECTIVE

To summarize the use of deep learning in the detection of voice disorders using acoustic and laryngoscopic input, compare specific neural networks in terms of accuracy, and assess their effectiveness compared to expert clinical visual examination.

DATA SOURCES

Embase, MEDLINE, and Cochrane Central.

REVIEW METHODS

Databases were screened through November 11, 2023 for relevant studies. The inclusion criteria required studies to utilize a specified deep learning method, use laryngoscopy or acoustic input, and measure accuracy of binary classification between healthy patients and those with voice disorders.

RESULTS

Thirty-four studies met the inclusion criteria, with 18 focusing on voice analysis, 15 on imaging analysis, and 1 both. Across the 18 acoustic studies, 21 programs were used for identification of organic and functional voice disorders. These technologies included 10 convolutional neural networks (CNNs), 6 multilayer perceptrons (MLPs), and 5 other neural networks. The binary classification systems yielded a mean accuracy of 89.0% overall, including 93.7% for MLP programs and 84.5% for CNNs. Among the 15 imaging analysis studies, a total of 23 programs were utilized, resulting in a mean accuracy of 91.3%. Specifically, the twenty CNNs achieved a mean accuracy of 92.6% compared to 83.0% for the 3 MLPs.

CONCLUSION

Deep learning models were shown to be highly accurate in the detection of voice pathology, with CNNs most effective for assessing laryngoscopy images and MLPs most effective for assessing acoustic input. While deep learning methods outperformed expert clinical exam in limited comparisons, further studies integrating external validation are necessary.

Pharyngeal reflux episodes in patients with suspected laryngopharyngeal reflux versus healthy subjects: a prospective cohort study

PURPOSE

This study aimed to analyze pharyngeal reflux episodes in patients with suspected LPR versus healthy subjects using 24-h MII-pH monitoring.

METHODS

One hundred twenty-one patients who visited our clinic with a chief complaint of LPR-related symptoms and underwent 24-h MII-pH monitoring were enrolled prospectively. Also, 27 healthy subjects were enrolled and underwent 24-h MII-pH monitoring during the same period. We analyzed sensitivity, specificity, and accuracy comprehensively to determine appropriate cut-off values of pharyngeal reflux episodes in 24-h MII-pH monitoring to diagnose patients with LPR.

RESULTS

Twenty-nine of 121 patients with suspected LPR showed no pharyngeal reflux episodes, while 92 showed more than one pharyngeal reflux event. In contrast, the 22 healthy subjects showed no pharyngeal reflux episodes, three showed one reflux event, and two showed two reflux events. A cut-off value of ≥ 1 showed best accuracy reflected by combined sensitivity and specificity values, while ≥ 2 demonstrated better specificity with slight loss of sensitivity and slightly lower overall accuracy, suggesting cut-off value of ≥ 1 pharyngeal reflux episodes is a good clinical indicator.

CONCLUSION

A cut-off value of ≥ 1 in pharyngeal reflux episodes on 24-h MII-pH monitoring in patients with suspected LPR might be an acceptable diagnostic tool for LPR.

Quantitative laryngoscopy with computer-aided diagnostic system for laryngeal lesions

Laryngoscopes are widely used in the clinical diagnosis of laryngeal lesions, but such diagnosis relies heavily on the physician's subjective experience. The purpose of this study was to develop a computer-aided diagnostic system for the detection of laryngeal lesions based on objective criteria. This study used the distinct features of the image contour to find the clearest image in the laryngoscopic video. First to reduce the illumination problem caused by the laryngoscope lens, which could not fix the position of the light source, this study proposed image compensation to provide the image with a consistent brightness range for better performance. Second, we also proposed a method to automatically screen clear images from laryngoscopic film. Third, we used ACM to segment automatically them based on structural features of the pharynx and larynx, using hue and geometric analysis in the vocal cords and other zones. Finally, the support vector machine was used to classify laryngeal lesions based on a decision tree. This study evaluated the performance of the proposed system by assessing the laryngeal images of 284 patients. The accuracy of the detection for vocal cord polyps, cysts, leukoplakia, tumors, and healthy vocal cords were 93.15%, 95.16%, 100%, 96.42%, and 100%, respectively. The cross-validation accuracy for the five classes were 93.1%, 94.95%, 99.4%, 96.01% and 100%, respectively, and the average test accuracy for the laryngeal lesions was 93.33%. Our results showed that it was feasible to take the hue and geometric features of the larynx as signs to identify laryngeal lesions and that they could effectively assist physicians in diagnosing laryngeal lesions.

Laryngopharyngeal reflux image quantization and analysis of its severity

Laryngopharyngeal reflux (LPR) is a prevalent disease affecting a high proportion of patients seeking laryngology consultation. Diagnosis is made subjectively based on history, symptoms, and endoscopic assessment. The results depend on the examiner's interpretation of endoscopic images. There are still no consistent objective diagnostic methods. The aim of this study is to use image processing techniques to quantize the laryngeal variation caused by LPR, to judge and analyze its severity. This study proposed methods of screening sharp images automatically from laryngeal endoscopic images and using throat eigen structure for automatic region segmentation. The proposed image compensation improved the illumination problems from the use of laryngoscope lens. Fisher linear discriminant was used to find out features and classification performance while support vector machine was used as the classifier for judging LPR. Evaluation results were 97.16% accuracy, 98.11% sensitivity, and 3.77% false positive rate. To evaluate the severity, quantized data of the laryngeal variation was used. LPR images were combined with reflux symptom index score chart, and severity was graded using a neural network. The results indicated 96.08% accuracy. The experiment indicated that laryngeal variation induced by LPR could be quantized by using image processing techniques to assist in diagnosing and treating LPR.

Changing Trends of Color of Different Laryngeal Regions in Laryngopharyngeal Reflux Disease

INTRODUCTION

In our previous study, we demonstrated that there might be correlations between laryngopharyngeal reflux disorder and the hue value of laryngoscopic images. And we found that different regions of larynx have different hue values. It was hypothesized that the degree of inflammation varies between different laryngeal regions, due to an acid reflux pattern.

OBJECTIVE

The objective of this study was to compare the changing trends of hue values of different laryngeal regions in patients with laryngopharyngeal reflux (LPR) disease.

METHODS

Ninety-seven patients, including 20 pH-positive, 19 pH-negative were tested for LPR through multichannel intraluminal impedance 24-hour pH monitoring, and 58 controls with reflux symptom index less than 13. Laryngoscopic images of all patients were obtained. The hue values of 7 areas of interest, including both sides of the true vocal folds, the false vocal folds, the arytenoids, and the interarytenoid space, were quantified using a hue calculation. The analysis of variance analysis was applied to find if there was significant difference between different groups within each region.

RESULTS

(1) In the regions of both sides of the true vocal folds and interarytenoid, there was no significant difference between positive group and negative group; (2) in the regions of both sides of false vocal folds, there was no significant difference between negative group and control group; (3) in the regions of both sides of arytenoids, there was neither significant difference between positive group and negative group nor between negative group and control group; (4) in other comparisons, there were significant differences.

CONCLUSION

Hue values of separate laryngeal regions are different. In negative group, the hue values of interarytenoid region are similar with positive group, and the sensitivity to the acid are different from different regions. The sensitivity in the true vocal folds may be present.

Evaluation and Management of Laryngopharyngeal Reflux Disease: State of the Art Review

OBJECTIVE

To review the current literature about the epidemiology, clinical presentation, diagnosis, and treatment of laryngopharyngeal reflux (LPR).

DATA SOURCES

PubMed, Cochrane Library, and Scopus.

METHODS

A comprehensive review of the literature on LPR epidemiology, clinical presentation, diagnosis, and treatment was conducted. Using the PRISMA statement, 3 authors selected relevant publications to provide a critical analysis of the literature.

CONCLUSIONS

The important heterogeneity across studies in LPR diagnosis continues to make it difficult to summarize a single body of thought. Controversies persist concerning epidemiology, clinical presentation, diagnosis, and treatment. No recent epidemiologic study exists regarding prevalence and incidence with the use of objective diagnostic tools. There is no survey that evaluates the prevalence of symptoms and signs on a large number of patients with confirmed LPR. Regarding diagnosis, an increasing number of authors used multichannel intraluminal impedance-pH monitoring, although there is no consensus regarding standardization of the diagnostic criteria. The efficiency of proton pump inhibitor (PPI) therapy remains poorly demonstrated and misevaluated by incomplete clinical tools that do not take into consideration many symptoms and extralaryngeal findings. Despite the recent advances in knowledge about nonacid LPR, treatment protocols based on PPIs do not seem to have evolved.

IMPLICATIONS FOR PRACTICE

The development of multichannel intraluminal impedance-pH monitoring and pepsin and bile salt detection should be considered for the establishment of a multiparameter diagnostic approach. LPR treatment should evolve to a more personalized regimen, including diet, PPIs, alginate, and magaldrate according to individual patient characteristics. Multicenter international studies with a standardized protocol could improve scientific knowledge about LPR.