Machine Learning-Assisted Vocal Cord Ultrasound Examination: Project VIPR
This work addresses diagnostic accuracy issues for patients undergoing vocal cord examinations, but it is incremental as it applies existing machine learning methods to a new medical imaging domain.
The researchers tackled the problem of operator-dependent accuracy in vocal cord ultrasound (VCUS) by developing a machine learning algorithm to automatically identify vocal cords and classify normal vs. paralysis, achieving validation accuracies of 96% for segmentation and 99% for classification.
Intro: Vocal cord ultrasound (VCUS) has emerged as a less invasive and better tolerated examination technique, but its accuracy is operator dependent. This research aims to apply a machine learning-assisted algorithm to automatically identify the vocal cords and distinguish normal vocal cord images from vocal cord paralysis (VCP). Methods: VCUS videos were acquired from 30 volunteers, which were split into still frames and cropped to a uniform size. Healthy and simulated VCP images were used as training data for vocal cord segmentation and VCP classification models. Results: The vocal cord segmentation model achieved a validation accuracy of 96%, while the best classification model (VIPRnet) achieved a validation accuracy of 99%. Conclusion: Machine learning-assisted analysis of VCUS shows great promise in improving diagnostic accuracy over operator-dependent human interpretation.