Endoscopic Colour Analysis of Nasal Mucosa: Methodology and Clinical Data

Abstract

Abstract Diseases of the nasal mucosa are common in clinical practice; on average, every third ENT (ear, nose, throat) patient complains of nasal mucosa problems. Mucosa pathologies change the structure and composition of tissue; the presence of different cells, variation in biochemical composition and dilation or constriction of blood vessels lead to mucosa colour changes that help setting the diagnosis. However, experience and colour vision vary between doctors, so a more objective mucosa colour examination could be helpful. The aim of this study was to investigate whether colour analysis of camera-captured endoscopic images can distinguish healthy from pathological nasal mucosa and how specific such analysis could be for various diseases. A total of 66 participants (four groups, including healthy volunteers) underwent anterior nasal endoscopy, and the taken photos and videos were further analysed using self-developed Matlab software. The procedure involved selection of the region of interest (inferior nasal turbinate), segmentation and averaging the colour parameters over selected areas, exploiting the RGB and L*a*b colour scales. Allergic, chronic hyperplastic and inflamed (chronic and acute) nasal mucosa conditions in comparison with healthy mucosa were studied in this work. A relatively wide spread of colour parameter values was observed in the healthy volunteer group as well as in all patient groups. The most sensitive colour parameters for each of the pathology groups were identified.