Deep-learning-based image acquisition support tool for endoscopic narrow Band Imaging of the Larynx

Introduction Narrow band imaging (NBI) enables a contrast-enhanced imaging of mucosal blow-vessels. Nowadays NBI is a standard feature in many endoscopes. NBI is increasingly being applies in clinical investigations of the head-neck region. Using flexible laryngoscopes different laryngeal lesions can be investigated in awake patients. NBI enables a better recognition and differentiation of different pathologies than white light endoscopy.

It is essential for Material and methods Our dataset consists of 74.915 NBI-images of 41 patients that have been acquired using flexible laryngoscopes (ENF-VH, Olympus). The NBI-images have been labeld by experts and served as ground truth for training and validation of our deep-learning-pipeline.

Results We developed a deep-learning-method that shows the physician in real time on the screen if the current NBI image is usable for the classification of laryngeal lesions or not. The physician can then adjust e.g. the position of the endoscope to acquire better images.

Discussion Our study is a first step towards making NBI available for ENT-physicians that are not experts in this technique. In the future the method will be extended to also detect and highlight suspect tissue regions.

Publication History

Article published online:
12 May 2023