Hyperspectral imaging (HSI) is increasingly gaining acceptance in the medical field. Up until now, HSI has been used in conjunction with rigid endoscopy to detect cancer in vivo. The logical next step is to pair HSI with flexible endoscopy, since it improves access to hard-to-reach areas. While the flexible endoscope’s fiber optic cables provide the advantage of flexibility, they also introduce an interfering honeycomb-like pattern onto images. Due to the substantial impact this pattern has on locating cancerous tissue, it must be removed before the HS data can be further processed. Thereby, the loss of information is to minimize avoiding the suppression of small-area variations of pixel values. We have developed a system that uses flexible endoscopy to record HS cubes of the larynx and designed a special filtering technique to remove the honeycomb-like pattern with minimal loss of information. We have confirmed its feasibility by comparing it to conventional filtering techniques using an objective metric and by applying unsupervised and supervised classifications to raw and pre-processed HS cubes. Compared to conventional techniques, our method successfully removes the honeycomb-like pattern and considerably improves classification performance, while preserving image details.
Hyperspectral imaging (HSI) is a technology with high potential in the field of non-invasive detection of cancer. However, in complex imaging situations like HSI of the larynx with a rigid endoscope, various image interferences can disable a proper classification of cancerous tissue. We identified three main problems: i) misregistration of single images in a HS cube due to patient heartbeat ii) image noise and iii) specular reflections (SR). Consequently, an image pre-processor is developed in the current paper to overcome these image interferences. It encompasses i) image registration ii) noise removal by minimum noise fraction (MNF) transformation and iii) a novel SR detection method. The results reveal that the pre-processor improves classification performance, while the newly developed SR detection method outperforms global thresholding technique hitherto used by 46%. The novel pre-processor will be used for future studies towards the development of an operational scheme for HS-based larynx cancer detection. RGB image of the larynx derived from the hyperspectral cube and corresponding specular reflections (a) manually segmented and (b) detected by a novel specular reflection detection method.
Currently, there are no fast and accurate screening methods available for head and neck cancer, the eighth most common tumor entity. For this study, we used hyperspectral imaging, an imaging technique for quantitative and objective surface analysis, combined with deep learning methods for automated tissue classification. As part of a prospective clinical observational study, hyperspectral datasets of laryngeal, hypopharyngeal and oropharyngeal mucosa were recorded in 98 patients before surgery in vivo. We established an automated data interpretation pathway that can classify the tissue into healthy and tumorous using convolutional neural networks with 2D spatial or 3D spatio‐spectral convolutions combined with a state‐of‐the‐art Densenet architecture. Using 24 patients for testing, our 3D spatio‐spectral Densenet classification method achieves an average accuracy of 81%, a sensitivity of 83% and a specificity of 79%.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.