Veterinary Research Communications, 27 Suppl. 1 (2003) 749–753 © 2003 Kluwer Academic Publishers. Printed in the Netherlands Endoscopic Examination of Equine Upper Respiratory Tract by Computed Image Analysis D. Fava, G. Bitelli*, M. Cappi, A. Diana, M. Pietra and S. Cinotti Department of Veterinary Clinical Science (Section of Internal Medicine), Faculty of Veterinary Medicine, *D.I.S.T.A.R.T. Faculty of Engineering, University of Bologna, 40064 Ozzano dell’Emilia(BO), Italy *Correspondence: Dipartimento Clinico Veterinario, Universita ` degli Studi di Bologna, Via Tolara di Sopra 50, 40064, Ozzano dell’Emilia (BO), Italy E-mail: mpietra@vet.unibo.it Keywords: computed assisted image analysis, endoscopy, horse, larynx, pharynx INTRODUCTION Upper respiratory airway endoscopy of horses at rest is the most reliable diagnostic tool to evaluate pharynx and larynx morphological and functional alterations. Unfortunately this examination doesn’t elicit the correct evaluation of dysfunctions appearing during physical activity (Ducharme, 1998; Kastner et al., 1998). The sensi- tivity of this technique can be increased by dynamic examination during treadmill exercise, by using specific arytenoids stimulation and, in addition, by increasing the pharyngeal inspiratory negative pressure (Slap Test, Lobelina administration, rebreathing technique) (Stick, 1998). A major obstacle to diagnosis is the difficulty in comparing images of different breathing phases from the same horse. A study per- formed by Lafortuna and colleagues (1999) showed the possibility of post processing the endoscopic images through the use of algorithms that are able to correct the optical distortion of the laryngoscope image. The aim of the present study is to compare pharyngeal and laryngeal endoscopic images obtained in horses at rest and after laryngeal stress using different techniques and a computed image correction method. MATERIALS AND METHODS Ten healthy horses of different breeds, five males and five females, weighing between 400 and 500 kg, were examined. The endoscopic images were obtained using a Pentax EG290P videoendoscope after the intravenous administration of detomidine (0.01 mg/ kg); in one horse the exam was repeated without tranquillization to evaluate the influence of the drug on pharyngeal and laryngeal motility. The endoscopic tube was placed in the rhinopharynx to focus the larynx perpendicularly. The distance 749