Introduction. Flexible endoscopes are essential for diagnostics . Digital endoscopes are connected to a video processor that can perform various operations to improve the image. One of those operations is edge enhancement that sharpens the image [2, 3]. Sharpening is applied to compensate for the blur that results from enlarging the image captured by the endoscope chip. The image on screen is typically enlarged by a factor of four. Sharpness has a major impact on the perception of image quality by ENT-professionals . Unfortunately, the algorithm and parameters that the processors apply are unknown, the name varies per video processor and the units that express the strength of the operation are arbitrary. These unknowns make it difficult to compare the image quality or standardize the level of edge enhancement. We solved this by uniformly measuring the level of edge enhancement with test images [5, 6]. Now we would like to optimize the level of edge enhancement with respect to the perception by ENT-professionals. Images without edge enhancement will be perceived as vague, whereas excessive levels of edge enhancement yield sharper images but contain objectionable artifacts and too much noise. Method. Edge enhancement is studied in three types of flexible digital ENT endoscopes by taking in vitro pictures of the Rez checker target, while systematically varying the levels of edge enhancement that are available on the video processors. The level of edge enhancement, sharpness and noise can be measured using these test images. We then collected series of in vivo images of the larynx of a healthy test subject while systematically varying the levels of edge enhancement that are available on the video processors. Each series of in vivo images will be presented to ENT-professionals for a forced pairwise comparison test, in which the participants have to select the image with the best image quality for diagnostic purposes and ignore variables like orientation and position of the endoscopic tip. The numbers of votes are converted to a psychometric scale of just noticeable differences according to the Thurstone V model. Results. We expect to present preliminary results at BME 2023. References  B. C. Paul, S. Chen, S. Sridharan, Y. Fang, M. R. Amin and R. C. Branski, “Diagnostic accuracy of history, laryngoscopy, and stroboscopy,” The Laryngoscope, vol. 123, no. 1, pp. 215-219, 2012.  M. Kawaida, F. Hiroyuki and N. Kohno, “Observations of laryngeal lesions with a rhinolarynx electronic videoendoscope system and digital image processing,” Ann Otol Rhinol Laryngol, vol. 107, no. 10 Pt 1, pp. 855-9, 1998.  M. Kawaida, H. Fukuda and N. Kohno, “Digital image processing of laryngeal lesions by electronic videoendoscopy,” Laryngoscope, vol. 112, no. 3, pp. 559-64, 2002.  G. Geleijnse, L. L. Veder, M. M. Hakkesteegt and R. M. Metselaar, “The objective measurement and subjective perception of image quality of flexible ENT endoscopes,” Journal of Image Science & Technology, vol. 66, no. 3, pp. 1-6, 2022.  G. Geleijnse and B. Rieger, “Influence of edge enhancement applied in endoscopic systems on sharpness and noise,” Journal of Biomedical Optics, vol. 27, no. 10, 2022.  G. Geleijnse and B. Rieger, “Edge enhancement applied in ENT-endoscopic systems,” in 9th Dutch Bio-Medical Engineering Conference, Egmond aan Zee, 2022.
|Period||27 Jan 2023|
|Event title||9th Dutch Bio-Medical Engineering Conference|
|Location||Egmond aan Zee, NetherlandsShow on map|
|Degree of Recognition||International|